Khronosopen access

OpenGL ES 3.2 Specification — OpenGL ES 3.2 Specification

OpenGL® ES Version 3.2 (May 5, 2022)

Editor: Jon Leech

Copyright © 2006-2022 The Khronos Group Inc. All Rights Reserved.

This specification is protected by copyright laws and contains material proprietary to Khronos. Except as described by these terms, it or any components may not be reproduced, republished, distributed, transmitted, displayed, broadcast or otherwise exploited in any manner without the express prior written permission of Khronos. This specification has been created under the Khronos Intellectual Property Rights Policy, which is Attachment A of the Khronos Group Membership Agreement available at www.khronos.org/files/member agreement.pdf. Khronos grants a conditional copyright license to use and reproduce the unmodified specification for any purpose, without fee or royalty, EXCEPT no licenses to any patent, trademark or other intellectual property rights are granted under these terms. Parties desiring to implement the specification and make use of Khronos trademarks in relation to that implementation, and receive reciprocal patent license protection under the Khronos IP Policy must become Adopters and confirm the implementation as conformant under the process defined by Khronos for this specification; see https://www.khronos.org/adopters. Khronos makes no, and expressly disclaims any, representations or warranties, express or implied, regarding this specification, including, without limitation: merchantability, fitness for a particular purpose, non-infringement of any intellectual property, correctness, accuracy, completeness, timeliness, and reliability. Under no circumstances will Khronos, or any of its Promoters, Contributors or Members, or their respective partners, officers, directors, employees, agents or representatives be liable for any damages, whether direct, indirect, special or consequential damages for lost revenues, lost profits, or otherwise, arising from or in connection with these materials. Vulkan and Khronos are registered trademarks, and WebGL and EGL are trademarks of The Khronos Group Inc. ASTC is a trademark of ARM Holdings PLC; OpenCL is a trademark of Apple Inc.; and OpenGL is a registered trademark, and the OpenGL ES and OpenGL SC logos are trademarks of Hewlett Packard Enterprise, all used under license by Khronos. All other product names, trademarks, and/or company names are used solely for identification and belong to their respective owners.

Contents 1

Introduction 1.1 Formatting of the OpenGL ES Specification . . . . . . . . . . . . 1.2 What is the OpenGL ES Graphics System? . . . . . . . . . . . . 1.3 Programmer’s View of OpenGL ES . . . . . . . . . . . . . . . . 1.4 Implementor’s View of OpenGL ES . . . . . . . . . . . . . . . . 1.5 Our View . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Related APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.1 OpenGL ES Shading Language . . . . . . . . . . . . . . 1.6.2 WebGL . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6.3 Window System Bindings . . . . . . . . . . . . . . . . . 1.6.4 OpenCL . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 Filing Bug Reports . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 1 2 2 3 3 3 4 4 4 5

2

OpenGL ES Fundamentals 2.1 OpenGL ES Fundamentals . . . . . . . . . . . . . . . . . . . . . 2.2 Command Syntax . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Data Conversion For State-Setting Commands . . . . . . 2.2.2 Data Conversions For State Query Commands . . . . . . 2.3 Command Execution . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Graphics Reset Recovery . . . . . . . . . . . . . . . . . . 2.3.3 Flush and Finish . . . . . . . . . . . . . . . . . . . . . . 2.3.4 Numeric Representation and Computation . . . . . . . . . 2.3.5 Fixed-Point Data Conversions . . . . . . . . . . . . . . . 2.4 Rendering Commands . . . . . . . . . . . . . . . . . . . . . . . 2.5 Context State . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5.1 Generic Context State Queries . . . . . . . . . . . . . . . 2.6 Objects and the Object Model . . . . . . . . . . . . . . . . . . . 2.6.1 Object Management . . . . . . . . . . . . . . . . . . . .

6 6 8 10 12 13 13 16 17 18 22 24 24 25 25 25

i

CONTENTS 2.6.2 2.6.3 2.6.4 2.6.5 2.6.6 2.6.7 2.6.8 2.6.9 2.6.10 2.6.11 2.6.12 2.6.13

ii Buffer Objects . . . . . . . . . . . . . . . . . . . . . . . Shader Objects . . . . . . . . . . . . . . . . . . . . . . . Program Objects . . . . . . . . . . . . . . . . . . . . . . Program Pipeline Objects . . . . . . . . . . . . . . . . . Texture Objects . . . . . . . . . . . . . . . . . . . . . . . Sampler Objects . . . . . . . . . . . . . . . . . . . . . . Renderbuffer Objects . . . . . . . . . . . . . . . . . . . . Framebuffer Objects . . . . . . . . . . . . . . . . . . . . Vertex Array Objects . . . . . . . . . . . . . . . . . . . . Transform Feedback Objects . . . . . . . . . . . . . . . . Query Objects . . . . . . . . . . . . . . . . . . . . . . . Sync Objects . . . . . . . . . . . . . . . . . . . . . . . .

26 27 27 27 27 28 28 28 28 29 29 29

3

Dataflow Model

30

4

Event Model 4.1 Sync Objects and Fences . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Waiting for Sync Objects . . . . . . . . . . . . . . . . . . 4.1.2 Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.3 Sync Object Queries . . . . . . . . . . . . . . . . . . . . 4.2 Query Objects and Asynchronous Queries . . . . . . . . . . . . . 4.2.1 Query Object Queries . . . . . . . . . . . . . . . . . . .

33 33 35 37 38 39 42

5

Shared Objects and Multiple Contexts 5.1 Object Deletion Behavior . . . . . . . . . . . . . . . . . . . . . . 5.1.1 Side Effects of Shared Context Destruction . . . . . . . . 5.1.2 Automatic Unbinding of Deleted Objects . . . . . . . . . 5.1.3 Deleted Object and Object Name Lifetimes . . . . . . . . 5.2 Sync Objects and Multiple Contexts . . . . . . . . . . . . . . . . 5.3 Propagating Changes to Objects . . . . . . . . . . . . . . . . . . 5.3.1 Determining Completion of Changes to an object . . . . . 5.3.2 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.3 Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . .

44 44 44 45 45 46 46 47 47 48

6

Buffer Objects 6.1 Creating and Binding Buffer Objects . . . . . . . . . . . . . . . . 6.1.1 Binding Buffer Objects to Indexed Targets . . . . . . . . . 6.2 Creating and Modifying Buffer Object Data Stores . . . . . . . . 6.3 Mapping and Unmapping Buffer Data . . . . . . . . . . . . . . . 6.3.1 Unmapping Buffers . . . . . . . . . . . . . . . . . . . . .

50 51 53 54 56 60

OpenGL ES 3.2 (May 5, 2022)

CONTENTS

iii

6.3.2 Effects of Mapping Buffers on Other GL Commands . . . Effects of Accessing Outside Buffer Bounds . . . . . . . . . . . . Copying Between Buffers . . . . . . . . . . . . . . . . . . . . . . Buffer Object Queries . . . . . . . . . . . . . . . . . . . . . . . . 6.6.1 Indexed Buffer Object Limits and Binding Queries . . . . Buffer Object State . . . . . . . . . . . . . . . . . . . . . . . . .

60 61 61 62 63 65

7

Programs and Shaders 7.1 Shader Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Shader Binaries . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Program Objects . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Program Interfaces . . . . . . . . . . . . . . . . . . . . . 7.4 Program Pipeline Objects . . . . . . . . . . . . . . . . . . . . . . 7.4.1 Shader Interface Matching . . . . . . . . . . . . . . . . . 7.4.2 Program Pipeline Object State . . . . . . . . . . . . . . . 7.5 Program Binaries . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Uniform Variables . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6.1 Loading Uniform Variables In The Default Uniform Block 7.6.2 Uniform Blocks . . . . . . . . . . . . . . . . . . . . . . . 7.6.3 Uniform Buffer Object Bindings . . . . . . . . . . . . . . 7.7 Atomic Counter Buffers . . . . . . . . . . . . . . . . . . . . . . . 7.7.1 Atomic Counter Buffer Object Storage . . . . . . . . . . 7.7.2 Atomic Counter Buffer Bindings . . . . . . . . . . . . . . 7.8 Shader Buffer Variables and Shader Storage Blocks . . . . . . . . 7.9 Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.10 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.11 Shader Memory Access . . . . . . . . . . . . . . . . . . . . . . . 7.11.1 Shader Memory Access Ordering . . . . . . . . . . . . . 7.11.2 Shader Memory Access Synchronization . . . . . . . . . 7.12 Shader, Program, and Program Pipeline Queries . . . . . . . . . . 7.13 Required State . . . . . . . . . . . . . . . . . . . . . . . . . . . .

66 67 70 71 79 94 97 99 100 102 108 111 115 116 116 117 117 119 120 120 121 123 127 135

8

Textures and Samplers 8.1 Texture Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Sampler Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3 Sampler Object Queries . . . . . . . . . . . . . . . . . . . . . . . 8.4 Pixel Rectangles . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.1 Pixel Storage Modes and Pixel Buffer Objects . . . . . . . 8.4.2 Transfer of Pixel Rectangles . . . . . . . . . . . . . . . . 8.5 Texture Image Specification . . . . . . . . . . . . . . . . . . . .

137 138 140 143 144 144 146 157

6.4 6.5 6.6 6.7

OpenGL ES 3.2 (May 5, 2022)

CONTENTS

9

iv

8.5.1 Required Texture Formats . . . . . . . . . . . . . . . . . 8.5.2 Encoding of Special Internal Formats . . . . . . . . . . . 8.5.3 Texture Image Structure . . . . . . . . . . . . . . . . . . 8.6 Alternate Texture Image Specification Commands . . . . . . . . . 8.6.1 Texture Copying Feedback Loops . . . . . . . . . . . . . 8.7 Compressed Texture Images . . . . . . . . . . . . . . . . . . . . 8.8 Multisample Textures . . . . . . . . . . . . . . . . . . . . . . . . 8.9 Buffer Textures . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.10 Texture Parameters . . . . . . . . . . . . . . . . . . . . . . . . . 8.11 Texture Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.11.1 Active Texture . . . . . . . . . . . . . . . . . . . . . . . 8.11.2 Texture Parameter Queries . . . . . . . . . . . . . . . . . 8.11.3 Texture Level Parameter Queries . . . . . . . . . . . . . . 8.12 Depth Component Textures . . . . . . . . . . . . . . . . . . . . . 8.13 Cube Map Texture Selection . . . . . . . . . . . . . . . . . . . . 8.13.1 Seamless Cube Map Filtering . . . . . . . . . . . . . . . 8.14 Texture Minification . . . . . . . . . . . . . . . . . . . . . . . . 8.14.1 Scale Factor and Level of Detail . . . . . . . . . . . . . . 8.14.2 Coordinate Wrapping and Texel Selection . . . . . . . . . 8.14.3 Mipmapping . . . . . . . . . . . . . . . . . . . . . . . . 8.14.4 Manual Mipmap Generation . . . . . . . . . . . . . . . . 8.15 Texture Magnification . . . . . . . . . . . . . . . . . . . . . . . . 8.16 Combined Depth/Stencil Textures . . . . . . . . . . . . . . . . . 8.17 Texture Completeness . . . . . . . . . . . . . . . . . . . . . . . . 8.17.1 Effects of Sampler Objects on Texture Completeness . . . 8.17.2 Effects of Completeness on Texture Application . . . . . . 8.17.3 Effects of Completeness on Texture Image Specification . 8.18 Immutable-Format Texture Images . . . . . . . . . . . . . . . . . 8.19 Texture State . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.20 Texture Comparison Modes . . . . . . . . . . . . . . . . . . . . . 8.20.1 Depth Texture Comparison Mode . . . . . . . . . . . . . 8.21 sRGB Texture Color Conversion . . . . . . . . . . . . . . . . . . 8.22 Shared Exponent Texture Color Conversion . . . . . . . . . . . . 8.23 Texture Image Loads and Stores . . . . . . . . . . . . . . . . . . 8.23.1 Image Unit Queries . . . . . . . . . . . . . . . . . . . . .

160 160 164 167 177 177 183 185 188 191 191 191 192 193 194 194 195 195 197 202 204 205 205 205 207 207 207 207 211 212 212 213 215 215 221

Framebuffers and Framebuffer Objects 9.1 Framebuffer Overview . . . . . . . . . . . . . . . . . . . . . . . 9.2 Binding and Managing Framebuffer Objects . . . . . . . . . . . . 9.2.1 Framebuffer Object Parameters . . . . . . . . . . . . . .

222 222 224 227

OpenGL ES 3.2 (May 5, 2022)

CONTENTS

9.3

9.4

9.5 9.6 9.7 9.8

v

9.2.2 Attaching Images to Framebuffer Objects . . . . . . . . . 9.2.3 Framebuffer Object Queries . . . . . . . . . . . . . . . . 9.2.4 Renderbuffer Objects . . . . . . . . . . . . . . . . . . . . 9.2.5 Required Renderbuffer Formats . . . . . . . . . . . . . . 9.2.6 Renderbuffer Object Queries . . . . . . . . . . . . . . . . 9.2.7 Attaching Renderbuffer Images to a Framebuffer . . . . . 9.2.8 Attaching Texture Images to a Framebuffer . . . . . . . . Feedback Loops Between Textures and the Framebuffer . . . . . . 9.3.1 Rendering Feedback Loops . . . . . . . . . . . . . . . . . 9.3.2 Texture Copying Feedback Loops . . . . . . . . . . . . . Framebuffer Completeness . . . . . . . . . . . . . . . . . . . . . 9.4.1 Framebuffer Attachment Completeness . . . . . . . . . . 9.4.2 Whole Framebuffer Completeness . . . . . . . . . . . . . 9.4.3 Required Framebuffer Formats . . . . . . . . . . . . . . . 9.4.4 Effects of Framebuffer Completeness on Framebuffer Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.5 Effects of Framebuffer State on Framebuffer Dependent Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mapping between Pixel and Element in Attached Image . . . . . . Conversion to Framebuffer-Attachable Image Components . . . . Conversion to RGBA Values . . . . . . . . . . . . . . . . . . . . Layered Framebuffers . . . . . . . . . . . . . . . . . . . . . . . .

10 Vertex Specification and Drawing Commands 10.1 Primitive Types . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.1 Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.2 Line Strips . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.3 Line Loops . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.4 Separate Lines . . . . . . . . . . . . . . . . . . . . . . . 10.1.5 Triangle Strips . . . . . . . . . . . . . . . . . . . . . . . 10.1.6 Triangle Fans . . . . . . . . . . . . . . . . . . . . . . . . 10.1.7 Separate Triangles . . . . . . . . . . . . . . . . . . . . . 10.1.8 Lines with Adjacency . . . . . . . . . . . . . . . . . . . 10.1.9 Line Strips with Adjacency . . . . . . . . . . . . . . . . . 10.1.10 Triangles with Adjacency . . . . . . . . . . . . . . . . . 10.1.11 Triangle Strips with Adjacency . . . . . . . . . . . . . . . 10.1.12 Separate Patches . . . . . . . . . . . . . . . . . . . . . . 10.1.13 General Considerations For Polygon Primitives . . . . . . 10.2 Current Vertex Attribute Values . . . . . . . . . . . . . . . . . . . 10.2.1 Current Generic Attributes . . . . . . . . . . . . . . . . . OpenGL ES 3.2 (May 5, 2022)

229 230 234 237 238 238 240 245 245 247 247 248 249 252 252 253 253 254 254 254 256 258 258 258 258 258 259 260 260 260 260 262 263 263 264 265 265

CONTENTS

vi

10.2.2 Vertex Attribute Queries . . . . . . . . . . . . . . . . . . 10.2.3 Required State . . . . . . . . . . . . . . . . . . . . . . . 10.3 Vertex Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.1 Specifying Arrays for Generic Vertex Attributes . . . . . . 10.3.2 Vertex Attribute Divisors . . . . . . . . . . . . . . . . . . 10.3.3 Transferring Array Elements . . . . . . . . . . . . . . . . 10.3.4 Primitive Restart . . . . . . . . . . . . . . . . . . . . . . 10.3.5 Robust Buffer Access . . . . . . . . . . . . . . . . . . . . 10.3.6 Packed Vertex Data Formats . . . . . . . . . . . . . . . . 10.3.7 Vertex Arrays in Buffer Objects . . . . . . . . . . . . . . 10.3.8 Array Indices in Buffer Objects . . . . . . . . . . . . . . 10.3.9 Indirect Commands in Buffer Objects . . . . . . . . . . . 10.4 Vertex Array Objects . . . . . . . . . . . . . . . . . . . . . . . . 10.5 Drawing Commands Using Vertex Arrays . . . . . . . . . . . . . 10.6 Vertex Array and Vertex Array Object Queries . . . . . . . . . . . 10.7 Required State . . . . . . . . . . . . . . . . . . . . . . . . . . . .

266 266 266 267 271 272 273 273 274 274 275 275 276 277 284 286

11 Programmable Vertex Processing 11.1 Vertex Shaders . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1.1 Vertex Attributes . . . . . . . . . . . . . . . . . . . . . . 11.1.2 Vertex Shader Variables . . . . . . . . . . . . . . . . . . 11.1.3 Shader Execution . . . . . . . . . . . . . . . . . . . . . . 11.2 Tessellation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.1 Tessellation Control Shaders . . . . . . . . . . . . . . . . 11.2.2 Tessellation Primitive Generation . . . . . . . . . . . . . 11.2.3 Tessellation Evaluation Shaders . . . . . . . . . . . . . . 11.3 Geometry Shaders . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3.1 Geometry Shader Input Primitives . . . . . . . . . . . . . 11.3.2 Geometry Shader Output Primitives . . . . . . . . . . . . 11.3.3 Geometry Shader Variables . . . . . . . . . . . . . . . . . 11.3.4 Geometry Shader Execution Environment . . . . . . . . .

288 288 288 293 297 308 309 314 322 327 328 329 330 330

12 Fixed-Function Vertex Post-Processing 12.1 The Last Vertex Processing Stage . . . . . . . . . . . . . . . . . . 12.2 Transform Feedback . . . . . . . . . . . . . . . . . . . . . . . . 12.2.1 Transform Feedback Objects . . . . . . . . . . . . . . . . 12.2.2 Transform Feedback Primitive Capture . . . . . . . . . . 12.3 Primitive Queries . . . . . . . . . . . . . . . . . . . . . . . . . . 12.4 Flatshading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.5 Primitive Clipping . . . . . . . . . . . . . . . . . . . . . . . . . .

335 335 336 336 339 344 345 346

OpenGL ES 3.2 (May 5, 2022)

CONTENTS

vii

12.5.1 Clipping Shader Outputs . . . . . . . . . . . . . . . . . . 346 12.6 Coordinate Transformations . . . . . . . . . . . . . . . . . . . . 347 12.6.1 Controlling the Viewport . . . . . . . . . . . . . . . . . . 347 13 Fixed-Function Primitive Assembly and Rasterization 13.1 Discarding Primitives Before Rasterization . . . . . . . . . . . . 13.2 Primitive Bounding Box . . . . . . . . . . . . . . . . . . . . . . 13.3 Invariance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4 Multisampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4.1 Sample Shading . . . . . . . . . . . . . . . . . . . . . . 13.5 Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5.1 Basic Point Rasterization . . . . . . . . . . . . . . . . . . 13.5.2 Point Multisample Rasterization . . . . . . . . . . . . . . 13.6 Line Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.6.1 Basic Line Segment Rasterization . . . . . . . . . . . . . 13.6.2 Other Line Segment Features . . . . . . . . . . . . . . . . 13.6.3 Line Rasterization State . . . . . . . . . . . . . . . . . . 13.6.4 Line Multisample Rasterization . . . . . . . . . . . . . . 13.7 Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7.1 Basic Polygon Rasterization . . . . . . . . . . . . . . . . 13.7.2 Depth Offset . . . . . . . . . . . . . . . . . . . . . . . . 13.7.3 Polygon Multisample Rasterization . . . . . . . . . . . . 13.7.4 Polygon Rasterization State . . . . . . . . . . . . . . . . 13.8 Early Per-Fragment Tests . . . . . . . . . . . . . . . . . . . . . . 13.8.1 Pixel Ownership Test . . . . . . . . . . . . . . . . . . . . 13.8.2 Scissor Test . . . . . . . . . . . . . . . . . . . . . . . . . 13.8.3 Multisample Fragment Operations . . . . . . . . . . . . . 13.8.4 The Early Fragment Test Qualifier . . . . . . . . . . . . .

350 352 352 353 354 356 356 357 357 357 358 360 362 362 363 363 366 367 367 367 368 368 369 370

14 Programmable Fragment Processing 14.1 Fragment Shader Variables . . . . . . . . . . . . . . . . . . . . . 14.2 Shader Execution . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2.1 Texture Access . . . . . . . . . . . . . . . . . . . . . . . 14.2.2 Shader Inputs . . . . . . . . . . . . . . . . . . . . . . . . 14.2.3 Shader Outputs . . . . . . . . . . . . . . . . . . . . . . . 14.2.4 Early Fragment Tests . . . . . . . . . . . . . . . . . . . .

371 371 372 373 373 376 378

15 Writing Fragments and Samples to the Framebuffer 379 15.1 Per-Fragment Operations . . . . . . . . . . . . . . . . . . . . . . 379 15.1.1 Alpha To Coverage . . . . . . . . . . . . . . . . . . . . . 379

OpenGL ES 3.2 (May 5, 2022)

CONTENTS

viii

15.1.2 Stencil Test . . . . . . . . . . . . . . . . . . . . . . . . . 15.1.3 Depth Buffer Test . . . . . . . . . . . . . . . . . . . . . . 15.1.4 Occlusion Queries . . . . . . . . . . . . . . . . . . . . . 15.1.5 Blending . . . . . . . . . . . . . . . . . . . . . . . . . . 15.1.6 sRGB Conversion . . . . . . . . . . . . . . . . . . . . . 15.1.7 Dithering . . . . . . . . . . . . . . . . . . . . . . . . . . 15.1.8 Additional Multisample Fragment Operations . . . . . . . 15.2 Whole Framebuffer Operations . . . . . . . . . . . . . . . . . . . 15.2.1 Selecting Buffers for Writing . . . . . . . . . . . . . . . . 15.2.2 Fine Control of Buffer Updates . . . . . . . . . . . . . . 15.2.3 Clearing the Buffers . . . . . . . . . . . . . . . . . . . . 15.2.4 Invalidating Framebuffer Contents . . . . . . . . . . . . .

381 383 383 384 395 395 396 397 397 399 400 403

16 Reading and Copying Pixels 16.1 Reading Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.1.1 Selecting Buffers for Reading . . . . . . . . . . . . . . . 16.1.2 ReadPixels . . . . . . . . . . . . . . . . . . . . . . . . . 16.1.3 Obtaining Pixels from the Framebuffer . . . . . . . . . . 16.1.4 Conversion of RGBA values . . . . . . . . . . . . . . . . 16.1.5 Final Conversion . . . . . . . . . . . . . . . . . . . . . . 16.1.6 Placement in Pixel Pack Buffer or Client Memory . . . . . 16.2 Copying Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.2.1 Blitting Pixel Rectangles . . . . . . . . . . . . . . . . . . 16.2.2 Copying Between Images . . . . . . . . . . . . . . . . . 16.3 Pixel Draw and Read State . . . . . . . . . . . . . . . . . . . . .

406 406 406 407 409 410 410 411 412 412 415 420

17 Compute Shaders 421 17.1 Compute Shader Variables . . . . . . . . . . . . . . . . . . . . . 423 18 Debug Output 18.1 Debug Messages . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2 Debug Message Callback . . . . . . . . . . . . . . . . . . . . . . 18.3 Debug Message Log . . . . . . . . . . . . . . . . . . . . . . . . 18.4 Controlling Debug Messages . . . . . . . . . . . . . . . . . . . . 18.5 Externally Generated Messages . . . . . . . . . . . . . . . . . . . 18.6 Debug Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.7 Debug Labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.8 Asynchronous and Synchronous Debug Output . . . . . . . . . . 18.9 Debug Output Queries . . . . . . . . . . . . . . . . . . . . . . .

OpenGL ES 3.2 (May 5, 2022)

424 425 427 428 429 430 430 432 433 434

CONTENTS

ix

19 Special Functions 437 19.1 Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 20 Context State Queries 20.1 Simple Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.2 Pointer, String, and Related Context Queries . . . . . . . . . . . . 20.3 Internal Format Queries . . . . . . . . . . . . . . . . . . . . . . . 20.3.1 Internal Format Query Parameters . . . . . . . . . . . . .

439 439 441 443 444

21 State Tables

446

A Invariance A.1 Repeatability . . . . . . . . . . . . . . . . . . . . . . . . . . . . A.2 Multi-pass Algorithms . . . . . . . . . . . . . . . . . . . . . . . A.3 Invariance Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . A.4 Tessellation Invariance . . . . . . . . . . . . . . . . . . . . . . . A.5 Atomic Counter Invariance . . . . . . . . . . . . . . . . . . . . . A.6 What All This Means . . . . . . . . . . . . . . . . . . . . . . . .

504 504 505 505 507 509 510

B Corollaries

511

C Compressed Texture Image Formats 513 C.1 ASTC Compressed Texture Image Formats . . . . . . . . . . . . 513 C.2 ETC Compressed Texture Image Formats . . . . . . . . . . . . . 515 D Version 3.0 and Before D.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . D.2 Change Log for 3.0.3 . . . . . . . . . . . . . . . . . . . . . . . . D.3 Change Log for 3.0.2 . . . . . . . . . . . . . . . . . . . . . . . . D.4 Change Log for 3.0.1 . . . . . . . . . . . . . . . . . . . . . . . . D.5 Credits and Acknowledgements . . . . . . . . . . . . . . . . . .

516 516 518 520 521 523

E Version 3.1 E.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . E.2 Change Log for Released Specifications . . . . . . . . . . . . . . E.3 Credits and Acknowledgements . . . . . . . . . . . . . . . . . .

526 526 527 533

F Version 3.2 F.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . F.2 Change Log Descriptions . . . . . . . . . . . . . . . . . . . . . . F.3 Change Log for Released Specifications . . . . . . . . . . . . . .

536 536 538 538

OpenGL ES 3.2 (May 5, 2022)

CONTENTS F.4

x

Credits and Acknowledgements . . . . . . . . . . . . . . . . . . 549

G Backwards Compatibility 551 G.1 Legacy Features . . . . . . . . . . . . . . . . . . . . . . . . . . . 551 G.2 Differences in Runtime Behavior . . . . . . . . . . . . . . . . . . 552 Index

553

OpenGL ES 3.2 (May 5, 2022)

List of Figures 3.1

Block diagram of the OpenGL ES pipeline. . . . . . . . . . . . .

31

8.1 8.2 8.3 8.4 8.5 8.6 8.7

Transfer of pixel rectangles. . . . . . . . . . . . . . . . . . . . . Selecting a subimage from an image . . . . . . . . . . . . . . . . UNSIGNED_SHORT formats . . . . . . . . . . . . . . . . . . . . . UNSIGNED_INT formats . . . . . . . . . . . . . . . . . . . . . . FLOAT_UNSIGNED_INT formats . . . . . . . . . . . . . . . . . . A texture image and the coordinates used to access it. . . . . . . . Example of the components returned for textureGather. . . . .

146 152 154 155 155 167 199

10.1 Vertex processing and primitive assembly. . . . . . . . . . . . . . 10.2 Triangle strips, fans, and independent triangles. . . . . . . . . . . 10.3 Lines with adjacency. . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Triangles with adjacency. . . . . . . . . . . . . . . . . . . . . . . 10.5 Triangle strips with adjacency. . . . . . . . . . . . . . . . . . . . 10.6 Packed component layout . . . . . . . . . . . . . . . . . . . . . .

256 259 260 260 262 274

11.1 Domain parameterization for tessellation. . . . . . . . . . . . . . 11.2 Inner triangle tessellation. . . . . . . . . . . . . . . . . . . . . . . 11.3 Inner quad tessellation. . . . . . . . . . . . . . . . . . . . . . . . 11.4 Isoline tessellation. . . . . . . . . . . . . . . . . . . . . . . . . .

314 318 320 322

13.1 Rasterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2 Visualization of Bresenham’s algorithm. . . . . . . . . . . . . . . 13.3 Rasterization of wide lines. . . . . . . . . . . . . . . . . . . . . . 13.4 The region used in rasterizing a multisampled line segment. . . . .

350 358 360 362

15.1 Per-fragment operations. . . . . . . . . . . . . . . . . . . . . . . 379 16.1 Operation of ReadPixels. . . . . . . . . . . . . . . . . . . . . . . 406

xi

List of Tables 2.1 2.2 2.3

GL command suffixes . . . . . . . . . . . . . . . . . . . . . . . . GL data types . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary of GL errors . . . . . . . . . . . . . . . . . . . . . . .

9 11 15

4.1 4.2

Initial properties of a sync object created with FenceSync. . . . . Asynchronous query targets . . . . . . . . . . . . . . . . . . . .

34 39

6.1 6.2 6.3 6.4 6.5

Buffer object binding targets. . . . . . . . . . . . . . . . . . . . . Buffer object parameters and their values. . . . . . . . . . . . . . Buffer object initial state. . . . . . . . . . . . . . . . . . . . . . . Buffer object state set by MapBufferRange. . . . . . . . . . . . Indexed buffer object limits and binding queries . . . . . . . . . .

52 52 55 58 64

7.1 7.2 7.3 7.4 7.5 7.6 7.7

CreateShader type values and the corresponding shader stages. . 68 GetProgramResourceiv properties and supported interfaces . . . 87 OpenGL ES Shading Language type tokens . . . . . . . . . . . . 92 Query targets for default uniform block storage, in components. . 103 Query targets for combined uniform block storage, in components. 104 GetProgramResourceiv properties used by GetActiveUniformsiv. 107 GetProgramResourceiv properties used by GetActiveUniformBlockiv. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

8.1 8.2 8.3 8.4 8.5 8.6 8.7

PixelStorei parameters. . . . . . . . . . . . . . . . . . . . . . . . Valid combinations of format, type, and sized internalformat. . . . Valid combinations of format, type, and unsized internalformat. . Pixel data types. . . . . . . . . . . . . . . . . . . . . . . . . . . . Pixel data formats. . . . . . . . . . . . . . . . . . . . . . . . . . Packed pixel formats. . . . . . . . . . . . . . . . . . . . . . . . . Packed pixel field assignments. . . . . . . . . . . . . . . . . . . .

xii

145 148 149 150 151 154 156

LIST OF TABLES

xiii

8.8

Conversion from RGBA, depth, and stencil pixel components to internal texture components. . . . . . . . . . . . . . . . . . . . . 159 8.9 Effective internal format . . . . . . . . . . . . . . . . . . . . . . 160 8.10 Sized internal color formats. . . . . . . . . . . . . . . . . . . . . 163 8.11 Sized internal depth and stencil formats. . . . . . . . . . . . . . . 164 8.12 ReadPixels format and type used during CopyTex*. . . . . . . . 169 8.13 Valid CopyTexImage source framebuffer/destination texture base internal format combinations. . . . . . . . . . . . . . . . . . . . . 170 8.14 Effective internal format corresponding to floating-point framebuffers171 8.15 Effective internal format corresponding to destination internalformat and linear source buffer component sizes. . . . . . . . . . . . 172 8.16 Effective internal format corresponding to destination internalformat and sRGB source buffer component sizes. . . . . . . . . . . . 172 8.17 Compressed internal formats. . . . . . . . . . . . . . . . . . . . . 179 8.18 Internal formats for buffer textures . . . . . . . . . . . . . . . . . 187 8.19 Texture parameters and their values. . . . . . . . . . . . . . . . . 189 8.20 Selection of cube map images. . . . . . . . . . . . . . . . . . . . 194 8.21 Texel location wrap mode application. . . . . . . . . . . . . . . . 198 8.22 Depth texture comparison functions. . . . . . . . . . . . . . . . . 213 8.23 sRGB texture internal formats. . . . . . . . . . . . . . . . . . . . 214 8.24 Layer numbers for cube map texture faces. . . . . . . . . . . . . . 216 8.25 Mapping of image load, store and atomic texel coordinate components to texel numbers. . . . . . . . . . . . . . . . . . . . . . . . 217 8.26 Supported image unit formats, with equivalent format layout qualifiers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 8.27 Texel sizes, compatibility classes, and pixel format/type combinations for each image format. . . . . . . . . . . . . . . . . . . . . 221 9.1

Framebuffer attachment points. . . . . . . . . . . . . . . . . . . . 239

10.1 Triangles generated by triangle strips with adjacency. . . . . . . . 10.2 Alternative ordering to triangles generated by triangle strips with adjacency (see table 10.1). . . . . . . . . . . . . . . . . . . . . . 10.3 Vertex array sizes (values per vertex) and data types for generic vertex attributes. . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Indirect commands and corresponding indirect buffer targets. . . .

263 264 268 276

11.1 Generic attribute components accessed by attribute variables. . . . 289 11.2 Generic attributes and vector types used by column vectors of matrix variables bound to generic attribute index i. . . . . . . . . . . 290

OpenGL ES 3.2 (May 5, 2022)

LIST OF TABLES

xiv

11.3 Scalar and vector vertex attribute types . . . . . . . . . . . . . . . 290 12.1 Transform feedback modes . . . . . . . . . . . . . . . . . . . . . 340 12.2 Output types for OpenGL ES Shading Language variables . . . . 342 12.3 Provoking vertex selection. . . . . . . . . . . . . . . . . . . . . . 345 14.1 Correspondence of filtered texture components to texture base components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374 15.1 RGB and alpha blend equations. . . . . . . . . . . . . . . . . . . 15.2 Blending functions. . . . . . . . . . . . . . . . . . . . . . . . . . 15.3 Advanced Blend Equations . . . . . . . . . . . . . . . . . . . . . 15.4 Hue-Saturation-Luminosity Advanced Blend Equations . . . . . . 15.5 Buffer selection for a framebuffer object . . . . . . . . . . . . . .

387 388 391 393 397

16.1 PixelStorei parameters. . . . . . . . . . . . . . . . . . . . . . . . 16.2 ReadPixels GL data types and reversed component conversion formulas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.3 Compatible internal formats for copying . . . . . . . . . . . . . . 16.4 Compatible internal formats for CopyImageSubData . . . . . . .

408 411 417 419

18.1 Sources of debug output messages . . . . . . . . . . . . . . . . . 18.2 Types of debug output messages . . . . . . . . . . . . . . . . . . 18.3 Severity levels of messages . . . . . . . . . . . . . . . . . . . . . 18.4 Object namespace identifiers . . . . . . . . . . . . . . . . . . . .

425 426 426 432

19.1 Hint targets and descriptions . . . . . . . . . . . . . . . . . . . . 437 20.1 Internal format targets . . . . . . . . . . . . . . . . . . . . . . . . 443 21.1 State Variable Types . . . . . . . . . . . . . . . . . . . . . . . . . 21.2 Current Values and Associated Data . . . . . . . . . . . . . . . . 21.3 Vertex Array Object State . . . . . . . . . . . . . . . . . . . . . . 21.4 Vertex Array Data (not in vertex array objects) . . . . . . . . . . . 21.5 Buffer Object State . . . . . . . . . . . . . . . . . . . . . . . . . 21.6 Transformation State . . . . . . . . . . . . . . . . . . . . . . . . 21.7 Rasterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.8 Multisampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.9 Textures (selector, state per texture unit) . . . . . . . . . . . . . . 21.10Textures (state per texture object) . . . . . . . . . . . . . . . . . . 21.11Textures (state per texture image) . . . . . . . . . . . . . . . . . . 21.12Textures (state per sampler object) . . . . . . . . . . . . . . . . . OpenGL ES 3.2 (May 5, 2022)

447 448 449 450 451 452 453 454 455 456 457 458

LIST OF TABLES

xv

21.13Pixel Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . 459 21.14Framebuffer Control . . . . . . . . . . . . . . . . . . . . . . . . 460 21.15Framebuffer (state per framebuffer object) . . . . . . . . . . . . . 461 21.16Framebuffer (state per attachment point) . . . . . . . . . . . . . . 462 21.17Renderbuffer (state per renderbuffer object) . . . . . . . . . . . . 463 21.18Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464 21.19Shader Object State . . . . . . . . . . . . . . . . . . . . . . . . . 465 21.20Program Pipeline Object State . . . . . . . . . . . . . . . . . . . 466 21.21Program Object State . . . . . . . . . . . . . . . . . . . . . . . . 467 21.22Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 468 21.23Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 469 21.24Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 470 21.25Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 471 21.26Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 472 21.27Program Interface State . . . . . . . . . . . . . . . . . . . . . . . 473 21.28Program Object Resource State . . . . . . . . . . . . . . . . . . . 474 21.29Program Object Resource State (cont.) . . . . . . . . . . . . . . . 475 21.30Vertex Shader State (not part of program objects) . . . . . . . . . 476 21.31Query Object State . . . . . . . . . . . . . . . . . . . . . . . . . 477 21.32Atomic Counter Buffer Binding State . . . . . . . . . . . . . . . 478 21.33Image State (state per image unit) . . . . . . . . . . . . . . . . . 479 21.34Shader Storage Buffer Binding State . . . . . . . . . . . . . . . . 480 21.35Transform Feedback State . . . . . . . . . . . . . . . . . . . . . 481 21.36Uniform Buffer Binding State . . . . . . . . . . . . . . . . . . . 482 21.37Sync (state per sync object) . . . . . . . . . . . . . . . . . . . . . 483 21.38Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484 21.39Compute Dispatch State . . . . . . . . . . . . . . . . . . . . . . 485 21.40Implementation Dependent Values . . . . . . . . . . . . . . . . . 486 21.41Implementation Dependent Values (cont.) † These queries return the maximum no. of samples for all internal formats required to support multisampled rendering. ‡“ULP” means “Units in the Last Place” or “Unit of Least Precision”. 487

21.42Implementation Dependent Values (cont.) . . . . . . . . . . . . . 21.43Implementation Dependent Version and Extension Support . . . . 21.44Implementation Dependent Vertex Shader Limits . . . . . . . . . 21.45Implementation Dependent Tessellation Shader Limits . . . . . . 21.46Implementation Dependent Tessellation Shader Limits (cont.) . . 21.47Implementation Dependent Geometry Shader Limits . . . . . . . 21.48Implementation Dependent Fragment Shader Limits . . . . . . . . 21.49Implementation Dependent Compute Shader Limits . . . . . . . . OpenGL ES 3.2 (May 5, 2022)

488 489 490 491 492 493 494 495

LIST OF TABLES

xvi

21.50Implementation Dependent Aggregate Shader Limits . . . . . . . 21.51Implementation Dependent Aggregate Shader Limits (cont.) . . . 21.52Implementation Dependent Aggregate Shader Limits (cont.) . . . 21.53Debug Output State . . . . . . . . . . . . . . . . . . . . . . . . . 21.54Implementation Dependent Debug Output State . . . . . . . . . . 21.55Implementation Dependent Transform Feedback Limits . . . . . . 21.56Framebuffer Dependent Values . . . . . . . . . . . . . . . . . . . 21.57Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . .

496 497 498 499 500 501 502 503

C.1 Mapping of OpenGL ES ASTC formats to descriptions. . . . . . . 514 C.2 Mapping of OpenGL ES ETC formats to descriptions. . . . . . . . 515

OpenGL ES 3.2 (May 5, 2022)

Chapter 1

Introduction This document, referred to as the “OpenGL ES Specification” or just “Specification” hereafter, describes the OpenGL ES graphics system: what it is, how it acts, and what is required to implement it. We assume that the reader has at least a rudimentary understanding of computer graphics. This means familiarity with the essentials of compute graphics algorithms and terminology as well as with modern GPUs (Graphic Processing Units). The canonical version of the Specification is available in the official OpenGL ES Registry, located at URL http://www.khronos.org/registry/gles

1.1

Formatting of the OpenGL ES Specification

Starting with version 3.1, the OpenGL ES Specification has undergone major restructuring to describe important concepts and objects in the context of the entire API before describing details of their use in the graphics pipeline, matching similar restructuring of the OpenGL 4.3 Specification.

1.2

What is the OpenGL ES Graphics System?

OpenGL ES (“Open Graphics Library for Embedded Systems”) is an API (Application Programming Interface) to graphics hardware. The API consists of a set of several hundred procedures and functions that allow a programmer to specify the shader programs, objects and operations involved in producing high-quality graphical images, specifically color images of three-dimensional objects. Most of OpenGL ES requires that the graphics hardware contain a framebuffer. Many OpenGL ES calls control drawing geometric objects such as points, lines, 1

1.3. PROGRAMMER’S VIEW OF OPENGL ES

2

and polygons, but the way that some of this drawing occurs (such as when antialiasing or multisampling is in use) relies on the existence of a framebuffer. Some commands explicitly manage the framebuffer.

1.3

Programmer’s View of OpenGL ES

To the programmer, OpenGL ES is a set of commands that allow the specification of shader programs or shaders, data used by shaders, and state controlling aspects of OpenGL ES outside the scope of shaders. Typically the data represent geometry in two or three dimensions and texture images, while the shaders control the geometric processing, rasterization of geometry and the lighting and shading of fragments generated by rasterization, resulting in rendering geometry into the framebuffer. A typical program that uses OpenGL ES begins with calls to open a window into the framebuffer into which the program will draw. Then, calls are made to allocate an OpenGL EScontext and associate it with the window. Once a context is allocated, OpenGL ES commands to define shaders,geometry, and textures are made, followed by commands which draw geometry by transferring specified portions of the geometry to the shaders. Drawing commands specify simple geometric objects such as points, line segments, and polygons, which can be further manipulated by shaders. There are also commands which directly control the framebuffer by reading and writing pixels.

1.4

Implementor’s View of OpenGL ES

To the implementor, OpenGL ES is a set of commands that control the operation of the GPU. Modern GPUs accelerate almost all OpenGL ES operations, storing data and framebuffer images in GPU memory and executing shaders in dedicated GPU processors. However, OpenGL ES may be implemented on less capable GPUs, or even without a GPU, by moving some or all operations into the host CPU. The implementor’s task is to provide a software library on the CPU which implements the OpenGL ES API, while dividing the work for each OpenGL ES command between the CPU and the graphics hardware as appropriate for the capabilities of the GPU. OpenGL ES contains a considerable amount of information including many types of objects representing programmable shaders and the data they consume and generate, as well as other context state controlling non-programmable aspects of OpenGL ES. Most of these objects and state are available to the programmer, who can set, manipulate, and query their values through OpenGL ES commands. Some OpenGL ES 3.2 (May 5, 2022)

1.5. OUR VIEW

3

of it, however, is derived state visible only by the effect it has on how OpenGL ES operates. One of the main goals of this Specification is to describe OpenGL ES objects and context state explicitly, to elucidate how they change in response to OpenGL ES commands, and to indicate what their effects are.

1.5

Our View

We view OpenGL ES as a pipeline having some programmable stages and some state-driven fixed-function stages that are invoked by a set of specific drawing operations. This model should engender a specification that satisfies the needs of both programmers and implementors. It does not, however, necessarily provide a model for implementation. An implementation must produce results conforming to those produced by the specified methods, but there may be ways to carry out a particular computation that are more efficient than the one specified.

1.6

Related APIs

Other APIs related to OpenGL are described below. Most of the specifications for these APIs are available on the Khronos Group websites, although some vendorspecific APIs are documented on that vendor’s developer website.

1.6.1

OpenGL ES Shading Language

The OpenGL ES Specification should be read together with a companion document titled The OpenGL ES Shading Language. The latter document (referred to as the OpenGL ES Shading Language Specification hereafter) defines the syntax and semantics of the programming language used to write shaders (see sections 7). Descriptions of shaders later in this document may include references to concepts and terms (such as shading language variable types) defined in the companion document. OpenGL ES 3.2 implementations are guaranteed to support versions 3.20, 3.10, 3.00 and 1.00 of the OpenGL ES Shading Language. All references to sections of that specification refer to version 3.20. The latest supported version of the shading language may be queried as described in section 20.2. The OpenGL ES Shading Language Specification is available in the OpenGL ES Registry.

OpenGL ES 3.2 (May 5, 2022)

1.6. RELATED APIS

1.6.2

4

WebGL

WebGL is a cross-platform, royalty-free web standard for a low-level 3D graphics API based on OpenGL ES.Developers familiar with OpenGL ES will recognize WebGL as a shader-based API using the OpenGL ES Shading Language, with constructs that are semantically similar to those of the underlying OpenGL ES API. It stays very close to the OpenGL ES specification, with some concessions made for what developers expect out of memory-managed languages such as JavaScript. The WebGL Specification and related documentation are available in the Khronos API Registry.

1.6.3

Window System Bindings

OpenGL ES requires a companion API to create and manage graphics contexts, windows to render into, and other resources beyond the scope of this Specification. There are several such APIs supporting different operating and window systems. The Khronos Native Platform Graphics Interface or “EGL Specification” describes the EGL API for use of OpenGL ES on mobile and embedded devices. EGL implementations may be available supporting OpenGL as well. The EGL Specification is available in the Khronos Extension Registry at URL http://www.khronos.org/registry/egl The EAGL API supports use of OpenGL ES with iOS. EAGL is documented on Apple’s developer website.

1.6.4

OpenCL

OpenCL is an open, royalty-free standard for cross-platform, general-purpose parallel programming of processors found in personal computers, servers, and mobile devices, including GPUs. OpenCL defines interop methods to share OpenCL memory and image objects with corresponding OpenGL ES buffer and texture objects, and to coordinate control of and transfer of data between OpenCL and OpenGL ES. This allows applications to split processing of data between OpenCL and OpenGL ES; for example, by using OpenCL to implement a physics model and then rendering and interacting with the resulting dynamic geometry using OpenGL ES. The OpenCL Specification is available in the Khronos API Registry.

OpenGL ES 3.2 (May 5, 2022)

1.7. FILING BUG REPORTS

1.7

5

Filing Bug Reports

Bug reports on the OpenGL ES and OpenGL ES Shading Language Specifications can be filed in the Khronos Public Bugzilla, located at URL http://www.khronos.org/bugzilla/ Please file bugs against Product: OpenGL ES, Component: Specification, and the appropriate version of the specification. It is best to file bugs against the most recently released versions, since older versions are usually not updated for bugfixes.

OpenGL ES 3.2 (May 5, 2022)

Chapter 2

OpenGL ES Fundamentals This chapter introduces fundamental concepts including the OpenGL ES execution model, API syntax, contexts and threads, numeric representation, context state and state queries, and the different types of objects and shaders. It provides a framework for interpreting more specific descriptions of commands and behavior in the remainder of the Specification.

2.1

OpenGL ES Fundamentals

OpenGL ES (henceforth, the “GL”) is concerned only with processing data in GPU memory, including rendering into a framebuffer and reading values stored in that framebuffer. There is no support for other input or output devices. Programmers must rely on other mechanisms to obtain user input. The GL draws primitives processed by a variety of shader programs and fixedfunction processing units controlled by context state. Each primitive is a point, line segment, or polygon. Context state may be changed independently; the setting of one piece of state does not affect the settings of others (although state and shaders all interact to determine what eventually ends up in the framebuffer). State is set, primitives drawn, and other GL operations described by sending commands in the form of function or procedure calls. Primitives are defined by a group of one or more vertices. A vertex defines a point, an endpoint of a line segment, or a corner of a polygon where two edges meet. Data such as positional coordinates, colors, normals, texture coordinates, etc. are associated with a vertex and each vertex is processed independently, in order, and in the same way. The only exception to this rule is if the group of vertices must be clipped so that the indicated primitive fits within a specified region; in this case vertex data may be modified and new vertices created. The type of clipping 6

2.1. OPENGL ES FUNDAMENTALS

7

depends on which primitive the group of vertices represents. Commands are always processed in the order in which they are received, although there may be an indeterminate delay before the effects of a command are realized. This means, for example, that one primitive must be drawn completely before any subsequent one can affect the framebuffer. It also means that queries and pixel read operations return state consistent with complete execution of all previously invoked GL commands, except where explicitly specified otherwise. In general, the effects of a GL command on either GL modes or the framebuffer must be complete before any subsequent command can have any such effects. In the GL, data binding occurs on call. This means that data passed to a OpenGL ES command are interpreted when that command is received. Even if the command requires a pointer to data, those data are interpreted when the call is made, and any subsequent changes to the data have no effect on the GL (unless the same pointer is used in a subsequent command). The GL provides direct control over the fundamental operations of 3D and 2D graphics. This includes specification of parameters of application-defined shader programs performing transformation, lighting, texturing, and shading operations, as well as built-in functionality such as antialiasing and texture filtering. It does not provide a means for describing or modeling complex geometric objects. In other words, OpenGL ES provides mechanisms to describe how complex geometric objects are to be rendered, rather than mechanisms to describe the complex objects themselves. The model for interpretation of GL commands is client-server. That is, a program (the client) issues commands, and these commands are interpreted and processed by the GL (the server). The server may or may not operate on the same computer or in the same address space as the client. In this sense, the GL is network-transparent. A server may maintain a number of GL contexts, each of which is an encapsulation of current GL state and objects. A client may choose to make any one of these contexts current. Issuing GL commands when the program does not have a current context results in undefined behavior, up to and including program termination. There are two classes of framebuffers: a window system-provided framebuffer associated with a context when the context is made current, and application-created framebuffers. The window system-provided framebuffer is referred to as the default framebuffer. Application-created framebuffers, referred to as framebuffer objects, may be created as desired. A context may be associated with two framebuffers, one for each of reading and drawing operations. The default framebuffer and framebuffer objects are distinguished primarily by the interfaces for configuring and managing their state. The effects of GL commands on the default framebuffer are ultimately conOpenGL ES 3.2 (May 5, 2022)

2.2. COMMAND SYNTAX

8

trolled by the window system, which allocates framebuffer resources, determines which portions of the default framebuffer the GL may access at any given time, and communicates to the GL how those portions are structured. Therefore, there are no GL commands to initialize a GL context or configure the default framebuffer. Similarly, display of framebuffer contents on a physical display device (including the transformation of individual framebuffer values by such techniques as gamma correction) is not addressed by the GL. Allocation and configuration of the default framebuffer occurs outside of the GL in conjunction with the window system, using companion APIs described in section 1.6.3. Allocation and initialization of GL contexts is also done using these companion APIs. GL contexts can typically be associated with different default framebuffers, and some context state is determined at the time this association is performed. It is possible to use a GL context without a default framebuffer, in which case a framebuffer object must be used to perform all rendering. This is useful for applications needing to perform offscreen rendering. OpenGL ES is designed to be run on a range of graphics platforms with varying graphics capabilities and performance. To accommodate this variety, we specify ideal behavior instead of actual behavior for certain GL operations. In cases where deviation from the ideal is allowed, we also specify the rules that an implementation must obey if it is to approximate the ideal behavior usefully. This allowed variation in GL behavior implies that two distinct GL implementations may not agree pixel for pixel when presented with the same input even when run on identical framebuffer configurations. Finally, command names, constants, and types are prefixed in the C language binding to OpenGL ES (by gl, GL_, and GL, respectively), to reduce name clashes with other packages. The prefixes are omitted in this document for clarity.

2.2

Command Syntax

The Specification describes OpenGL ES commands as functions or procedures using ANSI C syntax. Languages such as C++ and Javascript that allow passing of argument type information permit language bindings with simpler declarations and fewer entry points. Various groups of GL commands perform the same operation but differ in how arguments are supplied to them. To conveniently accommodate this variation, we adopt a notation for describing commands and their arguments. GL commands are formed from a name which may be followed, depending on the particular command, by a sequence of characters describing a parameter to the

OpenGL ES 3.2 (May 5, 2022)

2.2. COMMAND SYNTAX

9

Type Descriptor i i64 f ui

Corresponding GL Type int int64 float uint

Table 2.1: Correspondence of command suffix type descriptors to GL argument types. Refer to table 2.2 for definitions of the GL types.

command. If present, a digit indicates the required length (number of values) of the indicated type. Next, a string of characters making up one of the type descriptors from table 2.1 indicates the specific size and data type of parameter values. A final v character, if present, indicates that the command takes a pointer to an array (a vector) of values rather than a series of individual arguments. Two specific examples are: void Uniform4f( int location, float v0, float v1, float v2, float v3 ); and void GetFloatv( enum pname, float *data ); In general, a command declaration has the form rtype Name{1234}{ i i64 f ui }{v} ( [args ,] T arg1 , . . . , T argN [, args] ); rtype is the return type of the function. The braces ({}) enclose a series of type descriptors (see table 2.1), of which one is selected.  indicates no type descriptor. The arguments enclosed in brackets ([args ,] and [, args]) may or may not be present. The N arguments arg1 through argN have type T, which corresponds to one of the type descriptors indicated in table 2.1 (if there are no letters, then the arguments’ type is given explicitly). If the final character is not v, then N is given by the digit 1, 2, 3, or 4 (if there is no digit, then the number of arguments is fixed). If the final character is v, then only arg1 is present and it is an array of N values of the indicated type. For example, void Uniform{1234}{if}( int location, T value ); OpenGL ES 3.2 (May 5, 2022)

2.2. COMMAND SYNTAX

10

indicates the eight declarations void Uniform1i( int location, int value ); void Uniform1f( int location, float value ); void Uniform2i( int location, int v0, int v1 ); void Uniform2f( int location, float v0, float v1 ); void Uniform3i( int location, int v0, int v1, int v2 ); void Uniform3f( int location, float v0, float v1, float v2 ); void Uniform4i( int location, int v0, int v1, int v2, int v3 ); void Uniform4f( int location, float v0, float v1, float v2, float v3 ); Arguments whose type is fixed (i.e. not indicated by a suffix on the command) are of one of the GL data types summarized in table 2.2, or pointers to one of these types1 . Since many GL operations represent bitfields within these types, transfer blocks of data in these types to graphics hardware which uses the same data types, or otherwise requires these sizes, it is not possible to implement the GL API on an architecture which cannot satisfy the exact bit width requirements in table 2.2.

2.2.1

Data Conversion For State-Setting Commands

Many GL commands specify a value or values to which GL state of a specific type (boolean, enum, integer, or floating-point) is to be set. When multiple versions of such a command exist, using the type descriptor syntax described above, any such version may be used to set the state value. When state values are specified using a different parameter type than the actual type of that state, data conversions are performed as follows: • When the type of internal state is boolean, zero integer or floating-point values are converted to FALSE and non-zero values are converted to TRUE. • When the type of internal state is integer or enum, boolean values of FALSE and TRUE are converted to 0 and 1, respectively. Floating-point values are rounded to the nearest integer. If the resulting value is so large in magnitude that it cannot be represented by the internal state variable, the internal state value is undefined. 1 Note that OpenGL ES 3.x uses float where OpenGL ES 2.0 used clampf. Clamping is now explicitly specified to occur only where and when appropriate, retaining proper clamping in conjunction with fixed-point framebuffers. Because clampf and float are both defined as the same floating-point type, this change should not introduce compatibility obstacles.

OpenGL ES 3.2 (May 5, 2022)

2.2. COMMAND SYNTAX

GL Type

11

Description

boolean byte

Bit Width 8 8

ubyte char short

8 8 16

ushort int

16 32

uint int64

32 64

uint64 fixed

64 32

sizei enum intptr

32 32 ptrbits

sizeiptr sync bitfield half

ptrbits ptrbits 32 16

float clampf

32 32

Boolean Signed two’s complement binary integer Unsigned binary integer Characters making up strings Signed two’s complement binary integer Unsigned binary integer Signed two’s complement binary integer Unsigned binary integer Signed two’s complement binary integer Unsigned binary integer Signed two’s complement 16.16 scaled integer Non-negative binary integer size Enumerated binary integer value Signed two’s complement binary integer Non-negative binary integer size Sync object handle (see section 4.1) Bit field Half-precision floating-point value encoded in an unsigned scalar Floating-point value Floating-point value clamped to [0, 1]

Table 2.2: GL data types. GL types are not C types. Thus, for example, GL type int is referred to as GLint outside this document, and is not necessarily equivalent to the C type int. An implementation must use exactly the number of bits indicated in the table to represent a GL type. ptrbits is the number of bits required to represent a pointer type; in other words, types intptr, sizeiptr, and sync must be sufficiently large as to store any address.

OpenGL ES 3.2 (May 5, 2022)

2.2. COMMAND SYNTAX

12

• When the type of internal state is floating-point, boolean values of FALSE and TRUE are converted to 0.0 and 1.0, respectively. Integer values are converted to floating-point. For commands taking arrays of the specified type, these conversions are performed for each element of the passed array. Each command following these conversion rules refers back to this section. Some commands have additional conversion rules specific to certain state values and data types, which are described following the reference. Validation of values performed by state-setting commands is performed after conversion, unless specified otherwise for a specific command.

2.2.2

Data Conversions For State Query Commands

Query commands (commands whose name begins with Get) return a value or values to which GL state has been set. Some of these commands exist in multiple versions returning different data types. When a query command is issued that returns data types different from the actual type of that state, data conversions are performed as follows. If more than one step is applicable, all relevant steps are applied in the following order: • If a command returning boolean data is called, such as GetBooleanv, a floating-point or integer value converts to FALSE if and only if it is zero. Otherwise it converts to TRUE. • If a command returning unsigned integer data is called, such as GetSamplerParameterIuiv, negative values are clamped to zero. • If a command returning integer data is called, such as GetIntegerv or GetInteger64v, a boolean value of TRUE or FALSE is interpreted as one or zero, respectively. A floating-point value is rounded to the nearest integer, unless the value is an RGBA color component, a DepthRangef value, or a depth buffer clear value. In these cases, the query command converts the floatingpoint value to an integer according to the INT entry of table 16.2; a value not in [−1, 1] converts to an undefined value. • If a command returning floating-point data is called, such as GetFloatv, a boolean value of TRUE or FALSE is interpreted as 1.0 or 0.0, respectively. An integer value is coerced to floating-point. Following these steps, if a value is so large in magnitude that it cannot be represented by the returned data type, then the nearest value representable using that type is returned. OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

13

When querying bitmasks (such as SAMPLE_MASK_VALUE or STENCIL_WRITEMASK) with GetIntegerv, the mask value is treated as a signed integer, so that mask values with the high bit set will not be clamped when returned as signed integers. Unless otherwise indicated, multi-valued state variables return their multiple values in the same order as they are given as arguments to the commands that set them. For instance, the two DepthRangef parameters are returned in the order n followed by f. Most texture state variables are qualified by the value of ACTIVE_TEXTURE to determine which server texture state vector is queried. Table 21.9 indicates those state variables which are qualified by ACTIVE_TEXTURE during state queries. Vertex array state variables are qualified by the value of VERTEX_ARRAY_BINDING to determine which vertex array object is queried. Table 21.3 defines the set of state stored in a vertex array object.

2.3

Command Execution

Most of the Specification discusses the behavior of a single context bound to a single CPU thread. It is also possible for multiple contexts to share GL objects and for each such context to be bound to a different thread. This section introduces concepts related to GL command execution including error reporting, command queue flushing, and synchronization between command streams. Using these tools can increase performance and utilization of the GPU by separating loosely related tasks into different contexts. Methods to create, manage, and destroy CPU threads are defined by the host CPU operating system and are not described in the Specification. Binding of GL contexts to CPU threads is controlled through a window system binding layer such as those described in section 1.6.3.

2.3.1

Errors

The GL detects only a subset of those conditions that could be considered errors. This is because in many cases error checking would adversely impact the performance of an error-free program. The command enum GetError( void ); is used to obtain error information. Each detectable error is assigned a numeric code. When an error is detected, a flag is set and the code is recorded. Further OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

14

errors, if they occur, do not affect this recorded code. When GetError is called, the code is returned and the flag is cleared, so that a further error will again record its code. If a call to GetError returns NO_ERROR, then there has been no detectable error since the last call to GetError (or since the GL was initialized). To allow for distributed implementations, there may be several flag-code pairs. In this case, after a call to GetError returns a value other than NO_ERROR each subsequent call returns the non-zero code of a distinct flag-code pair (in unspecified order), until all non-NO_ERROR codes have been returned. When there are no more non-NO_ERROR error codes, all flags are reset. This scheme requires some positive number of pairs of a flag bit and an integer. The initial state of all flags is cleared and the initial value of all codes is NO_ERROR. Table 2.3 summarizes GL errors. Currently, when an error flag is set, results of GL operation are undefined only if an OUT_OF_MEMORY error has occurred. In other cases, there are no side effects unless otherwise noted; the command which generates the error is ignored so that it has no effect on GL state or framebuffer contents. Except as otherwise noted, if the generating command returns a value, it returns zero. If the generating command modifies values through a pointer argument, no change is made to these values. These error semantics apply only to GL errors, not to system errors such as memory access errors. This behavior is the current behavior; the action of the GL in the presence of errors is subject to change, and extensions to OpenGL ES may define behavior currently considered as an error. Several error generation conditions are implicit in the description of every GL command: • If the GL context has been reset as a result of previous GL command, or if the context is reset as a side effect of execution of a command, a CONTEXT_LOST error is generated. • If a command that requires an enumerated value is passed a symbolic constant that is not one of those specified as allowable for that command, an INVALID_ENUM error is generated. This is the case even if the argument is a pointer to a symbolic constant, if the value pointed to is not allowable for the given command. • If a negative number is provided where an argument of type sizei or sizeiptr is specified, an INVALID_VALUE error is generated. • If memory is exhausted as a side effect of the execution of a command, an OUT_OF_MEMORY error may be generated.

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION Error

Description

CONTEXT_LOST

Context has been lost and reset by the driver

INVALID_ENUM INVALID_VALUE INVALID_OPERATION INVALID_FRAMEBUFFER_OPERATION OUT_OF_MEMORY

enum argument out of range Numeric argument out of range Operation illegal in current state Framebuffer object is not complete

STACK_OVERFLOW STACK_UNDERFLOW

Not enough memory left to execute command Command would cause a stack overflow Command would cause a stack underflow

15 Offending command ignored? Except as noted for specific commands Yes Yes Yes Yes

Unknown Yes Yes

Table 2.3: Summary of GL errors The Specification attempts to explicitly describe these implicit error conditions (with the exception of CONTEXT_LOST2 and OUT_OF_MEMORY3 ) wherever they apply. However, they apply even if not explicitly described, unless a specific command describes different behavior. For example, certain commands use a sizei parameter to indicate the length of a string, and also use negative values of the parameter to indicate a null-terminated string. These commands do not generate an INVALID_VALUE error, because they explicitly describe different behavior. Otherwise, errors are generated only for conditions that are explicitly described in this specification. When a command could potentially generate several different errors (for example, when it is passed separate enum and numeric parameters which are both out of range), the GL implementation may choose to generate any of the applicable errors. Errors based solely on one or more argument values to a command must be 2

CONTEXT_LOST is not described because it can potentially be generated by almost all GL commands, and occurs for reasons not directly related to the affected commands. 3 OUT_OF_MEMORY is not described because it can potentially be generated by any GL command, even those which do not explicitly allocate GPU memory.

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

16

detected before any processing based on current state4 . When an error is generated, the GL may also generate a debug output message describing its cause (see chapter 18). The message has source DEBUG_SOURCE_API, type DEBUG_TYPE_ERROR, and an implementation-dependent ID. Most commands include a complete summary of errors at the end of their description, including even the implicit errors described above. Such error summaries are set in a distinct style, like this sentence. In some cases, however, errors may be generated for a single command for reasons not directly related to that command. One such example is that deferred processing for shader programs may result in link errors detected only when attempting to draw primitives using vertex specification commands. In such cases, errors generated by a command may be described elsewhere in the specification than the command itself.

2.3.2

Graphics Reset Recovery

Certain events can result in a reset of the GL context. After such an event, it is referred to as a lost context and is unusable for almost all purposes. Recovery requires creating a new context and recreating all relevant state from the lost context. The current status of the graphics reset state is returned by enum GetGraphicsResetStatus( void ); The value returned indicates if the GL context has been in a reset state at any point since the last call to GetGraphicsResetStatus: • NO_ERROR indicates that the GL context has not been in a reset state since the last call. • GUILTY_CONTEXT_RESET indicates that a reset has been detected that is attributable to the current GL context. • INNOCENT_CONTEXT_RESET indicates a reset has been detected that is not attributable to the current GL context. • UNKNOWN_CONTEXT_RESET indicates a detected graphics reset whose cause is unknown. 4

This ensures consistent behavior for commands including language which ignores certain parameters under some conditions, such as glBlitFramebuffer treatment of mask and filter.

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

17

If a reset status other than NO_ERROR is returned and subsequent calls return NO_ERROR, the context reset was encountered and completed. If a reset status is repeatedly returned, the context may be in the process of resetting. Reset notification behavior is determined at context creation time, and may be queried by calling GetIntegerv with pname RESET_NOTIFICATION_STRATEGY. If the reset notification behavior is NO_RESET_NOTIFICATION, then the implementation will never deliver notification of reset events, and GetGraphicsResetStatus will always return NO_ERROR5 . If the behavior is LOSE_CONTEXT_ON_RESET, a graphics reset will result in a lost context and require creating a new context as described above. In this case GetGraphicsResetStatus may return any of the values described above. If a graphics reset notification occurs in a context, a notification must also occur in all other contexts which share objects with that context6 . After a graphics reset has occurred on a context, subsequent GL commands on that context (or any context which shares with that context) will generate a CONTEXT_LOST error. Such commands will not have side effects (in particular, they will not modify memory passed by pointer for query results), and may not block indefinitely or cause termination of the application. Exceptions to this behavior include: • GetError and GetGraphicsResetStatus behave normally following a graphics reset, so that the application can determine a reset has occurred, and when it is safe to destroy and re-create the context. • Any commands which might cause a polling application to block indefinitely will generate a CONTEXT_LOST error, but will also return a value indicating completion to the application. Such commands include: – GetSynciv with pname SYNC_STATUS ignores the other parameters and returns SIGNALED in values. – GetQueryObjectuiv with pname QUERY_RESULT_AVAILABLE ignores the other parameters and returns TRUE in params.

2.3.3

Flush and Finish

Implementations may buffer multiple commands in a command queue before sending them to the GL server for execution. This may happen in places such as the 5 In this case, it is recommended that implementations should not allow loss of context state no matter what events occur. However, this is only a recommendation, and cannot be relied upon by applications. 6 The values returned by GetGraphicsResetStatus in the different contexts may differ.

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

18

network stack (for network transparent implementations), CPU code executing as part of the GL client or the GL server, or internally to the GPU hardware. Coarse control over command queues is available using the command void Flush( void ); which causes all previously issued GL commands to complete in finite time7 (although such commands may still be executing when Flush returns). The command void Finish( void ); forces all previous GL commands to complete. Finish does not return until all effects from previously issued commands on GL client and server state and the framebuffer are fully realized. Finer control over command execution can be expressed using fence commands and sync objects, as discussed in section 4.1.

2.3.4

Numeric Representation and Computation

The GL must perform a number of floating-point operations during the course of its operation. Implementations normally perform computations in floating-point, and must meet the range and precision requirements defined in section 2.3.4.1 below. These requirements only apply to computations performed in GL operations outside of shader execution, such as texture image specification and sampling, and per-fragment operations. Range and precision requirements during shader execution differ and are specified by the OpenGL ES Shading Language Specification. In some cases, the representation and/or precision of operations is implicitly limited by the specified format of vertex, texture, or renderbuffer data consumed by the GL. Specific floating-point formats are described later in this section. 7

Historically, use of Flush has had negative performance implications for some implementations, and subsequently it does not universally operate as described here. One exception to this is when the default framebuffer is bound, and it is single-buffered; in this case, flush behaves as expected. Other references to the flush operation in the specification, such as that in section 4.1.2, will behave as expected. Waiting on a fence sync object with SYNC_FLUSH_COMMANDS_BIT is thus recommended as a way to perform a guaranteed flush.

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION 2.3.4.1

19

Floating-Point Computation

We do not specify how floating-point numbers are to be represented, or the details of how operations on them are performed. We require simply that numbers’ floating-point parts contain enough bits and that their exponent fields are large enough so that individual results of floating-point operations are accurate to about 1 part in 105 . The maximum representable magnitude for all floating-point values must be at least 232 . x · 0 = 0 · x = 0 for any non-infinite and non-NaN x. 1 · x = x · 1 = x. x + 0 = 0 + x = x. 00 = 1. (Occasionally further requirements will be specified.) Most single-precision floating-point formats meet these requirements. The special values Inf and −Inf encode values with magnitudes too large to be represented; the special value NaN encodes “Not A Number” values resulting from undefined arithmetic operations such as 00 . Implementations are permitted, but not required, to support Inf s and NaN s in their floating-point computations. Any representable floating-point value is legal as input to a GL command that requires floating-point data. The result of providing a value that is not a floatingpoint number to such a command is unspecified, but must not lead to GL interruption or termination. In IEEE arithmetic, for example, providing a negative zero or a denormalized number to a GL command yields predictable results, while providing a NaN or an infinity yields unspecified results. 2.3.4.2

16-Bit Floating-Point Numbers

A 16-bit floating-point number has a 1-bit sign (S), a 5-bit exponent (E), and a 10-bit mantissa (M ). The value V of a 16-bit floating-point number is determined by the following:  S  E = 0, M = 0 (−1) × 0.0,   M S −14   × 210 , E = 0, M 6= 0 (−1) × 2
M S E−15 V = (−1) × 2 × 1 + 210 , 0 < E < 31   S  (−1) × Inf , E = 31, M = 0     NaN , E = 31, M 6= 0 If the floating-point number is interpreted as an unsigned 16-bit integer N , then

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

20



 mod 65536 S= 32768   N mod 32768 E= 1024 M = N mod 1024. N

Any representable 16-bit floating-point value is legal as input to a GL command that accepts 16-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN ) to such a command is unspecified, but must not lead to GL interruption or termination. Providing a denormalized number or negative zero to GL must yield predictable results, whereby the value is either preserved or forced to positive or negative zero. 2.3.4.3

Unsigned 11-Bit Floating-Point Numbers

An unsigned 11-bit floating-point number has no sign bit, a 5-bit exponent (E), and a 6-bit mantissa (M ). The value V of an unsigned 11-bit floating-point number is determined by the following:   0.0, E = 0, M = 0    M −14   E = 0, M 6= 0 × 64 , 2
M E−15 V = 2 × 1 + 64 , 0 < E < 31    Inf , E = 31, M = 0     NaN , E = 31, M 6= 0 If the floating-point number is interpreted as an unsigned 11-bit integer N , then 

 N E= 64 M = N mod 64. When a floating-point value is converted to an unsigned 11-bit floating-point representation, finite values are rounded to the closest representable finite value. While less accurate, implementations are allowed to always round in the direction of zero. This means negative values are converted to zero. Likewise, finite positive values greater than 65024 (the maximum finite representable unsigned 11-bit floating-point value) are converted to 65024. Additionally: negative infinity is converted to zero; positive infinity is converted to positive infinity; and both positive and negative NaN are converted to positive NaN . OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION

21

Any representable unsigned 11-bit floating-point value is legal as input to a GL command that accepts 11-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN ) to such a command is unspecified, but must not lead to GL interruption or termination. Providing a denormalized number to GL must yield predictable results, whereby the value is either preserved or forced to zero. 2.3.4.4

Unsigned 10-Bit Floating-Point Numbers

An unsigned 10-bit floating-point number has no sign bit, a 5-bit exponent (E), and a 5-bit mantissa (M ). The value V of an unsigned 10-bit floating-point number is determined by the following:   0.0, E = 0, M = 0    M  2−14 × 32 , E = 0, M 6= 0 
V = 2E−15 × 1 + M 32 , 0 < E < 31    Inf , E = 31, M = 0     NaN , E = 31, M 6= 0 If the floating-point number is interpreted as an unsigned 10-bit integer N , then  N E= 32 M = N mod 32. 

When a floating-point value is converted to an unsigned 10-bit floating-point representation, finite values are rounded to the closest representable finite value. While less accurate, implementations are allowed to always round in the direction of zero. This means negative values are converted to zero. Likewise, finite positive values greater than 64512 (the maximum finite representable unsigned 10-bit floating-point value) are converted to 64512. Additionally: negative infinity is converted to zero; positive infinity is converted to positive infinity; and both positive and negative NaN are converted to positive NaN . Any representable unsigned 10-bit floating-point value is legal as input to a GL command that accepts 10-bit floating-point data. The result of providing a value that is not a floating-point number (such as Inf or NaN ) to such a command is unspecified, but must not lead to GL interruption or termination. Providing a denormalized number to GL must yield predictable results, whereby the value is either preserved or forced to zero. OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION 2.3.4.5

22

Fixed-Point Computation

Vertex attributes may be specified using a 32-bit two’s complement signed representation with 16 bits to the right of the binary point (fraction bits). 2.3.4.6

General Requirements

Some calculations require division. In such cases (including implied divisions required by vector normalizations), a division by zero produces an unspecified result but must not lead to GL interruption or termination.

2.3.5

Fixed-Point Data Conversions

When generic vertex attributes and pixel color or depth components are represented as integers, they are often (but not always) considered to be normalized. Normalized integer values are treated specially when being converted to and from floating-point values, and are usually referred to as normalized fixed-point. Such values are always either signed or unsigned. In the remainder of this section, b denotes the bit width of the fixed-point integer representation. When the integer is one of the types defined in table 2.2, b is the minimum required bit width of that type. When the integer is a texture or renderbuffer color or depth component (see section 8.5), b is the number of bits allocated to that component in the internal format of the texture or renderbuffer. When the integer is a framebuffer color or depth component (see section 9), b is the number of bits allocated to that component in the framebuffer. The signed and unsigned fixed-point representations are assumed to be b-bit binary two’s-complement integers and binary unsigned integers, respectively. All the conversions described below are performed as defined, even if the implemented range of an integer data type is greater than the minimum required range. 2.3.5.1

Conversion from Normalized Fixed-Point to Floating-Point

Unsigned normalized fixed-point integers represent numbers in the range [0, 1]. The conversion from an unsigned normalized fixed-point value c to the corresponding floating-point value f is defined as f=

c . 2b − 1

(2.1)

Signed normalized fixed-point integers represent numbers in the range [−1, 1]. The conversion from a signed normalized fixed-point value c to the corresponding

OpenGL ES 3.2 (May 5, 2022)

2.3. COMMAND EXECUTION floating-point value f is performed using   c f = max , −1.0 . 2b−1 − 1

23

(2.2)

Only the range [−2b−1 + 1, 2b−1 − 1] is used to represent signed fixed-point values in the range [−1, 1]. For example, if b = 8, then the integer value −127 corresponds to −1.0 and the value 127 corresponds to 1.0. Note that while zero can be exactly expressed in this representation, one value (−128 in the example) is outside the representable range, and must be clamped before use.This equation is used everywhere that signed normalized fixed-point values are converted to floating-point, including for all signed normalized fixed-point parameters in GL commands, such as vertex attribute values8 , as well as for texture sampling or framebuffer values used for blending. 2.3.5.2

Conversion from Floating-Point to Normalized Fixed-Point

The conversion from a floating-point value f to the corresponding unsigned normalized fixed-point value c is defined by first clamping f to the range [0, 1], then computing f 0 = convert f loat uint(f × (2b − 1), b)

(2.3)

where convert f loat uint(r, b) returns one of the two unsigned binary integer values with exactly b bits which are closest to the floating-point value r (where rounding to nearest is preferred). The conversion from a floating-point value f to the corresponding signed normalized fixed-point value c is performed by clamping f to the range [−1, 1], then computing: f 0 = convert f loat int(f × (2b−1 − 1), b)

(2.4)

where convert f loat int(r, b) returns one of the two signed two’s-complement binary integer values with exactly b bits which are closest to the floating-point value r (where rounding to nearest is preferred). This equation is used everywhere that floating-point values are converted to signed normalized fixed-point, including when querying floating-point state (see 8 This is a behavior change in OpenGL ES 3.0. In previous versions, a different conversion for signed normalized values was used in which −128 mapped to −1.0, 127 mapped to 1.0, and 0.0 was not exactly representable.

OpenGL ES 3.2 (May 5, 2022)

2.4. RENDERING COMMANDS

24

section 20) and returning integers9 , as well as for specifying signed normalized texture or framebuffer values using floating-point.

2.4

Rendering Commands

GL commands performing rendering into a framebuffer are called rendering commands, and include the drawing commands *Draw* (see section 10.5), as well as these additional commands: • BlitFramebuffer (see section 16.2.1) • Clear (see section 15.2.3) • ClearBuffer* (see section 15.2.3.1) • DispatchCompute* (see section 17)

2.5

Context State

Context state is state that belongs to the GL context as a whole, rather than to instances of the different object types described in section 2.6. Context state controls fixed-function stages of the GPU, such as clipping, primitive rasterization, and framebuffer clears, and also specifies bindings of objects to the context specifying which objects are used during command execution. The Specification describes all visible context state variables and describes how each one can be changed. State variables are grouped somewhat arbitrarily by their function. Although we describe operations that the GL performs on the framebuffer, the framebuffer is not a part of GL state. There are two types of context state. Server state resides in the GL server; the majority of GL state falls into this category. Client state resides in the GL client. Unless otherwise specified, all state is server state; client state is specifically identified. Each instance of a context includes a complete set of server state; each connection from a client to a server also includes a complete set of client state. While an implementation of OpenGL ES may be hardware dependent, the Specification is independent of any specific hardware on which it is implemented. We are concerned with the state of graphics hardware only when it corresponds precisely to GL state. 9 This is a behavior change in OpenGL ES 3.0. In previous versions, a different conversion for signed normalized values was used in which −1.0 mapped to −128, 1.0 mapped to 127, and 0.0 was not exactly representable.

OpenGL ES 3.2 (May 5, 2022)

2.6. OBJECTS AND THE OBJECT MODEL

2.5.1

25

Generic Context State Queries

Context state queries are described in detail in chapter 20.

2.6

Objects and the Object Model

Many types of objects are defined in the remainder of the Specification. Applications may create, modify, query, and destroy many instances of each of these object types, limited in most cases only by available graphics memory. Specific instances of different object types are bound to a context. The set of bound objects define the shaders which are invoked by GL drawing operations; specify the buffer data, texture image, and framebuffer memory that is accessed by shaders and directly by GL commands; and contain the state used by other operations such as fence synchronization and timer queries. Each object type corresponds to a distinct set of commands which manage objects of that type. However, there is an object model describing how most types of objects are managed, described below. Exceptions to the object model for specific object types are described later in the Specification together with those object types. Following the description of the object model, each type of object is briefly described below, together with forward references to full descriptions of that object type in later chapters of the Specification. Objects are described in an order corresponding to the structure of the remainder of the Specification.

2.6.1

Object Management

2.6.1.1

Name Spaces, Name Generation, and Object Creation

Each object type has a corresponding name space. Names of objects are represented by unsigned integers of type uint. The name zero is reserved by the GL; for some object types, zero names a default object of that type, and in others zero will never correspond to an actual instance of that object type. Names of most types of objects are created by generating unused names using commands starting with Gen followed by the object type. For example, the command GenBuffers returns one or more previously unused buffer object names. Generated names are marked by the GL as used, for the purpose of name generation only. Object names marked in this fashion will not be returned by additional calls to generate names of the same type until the names are marked unused again by deleting them (see below).

OpenGL ES 3.2 (May 5, 2022)

2.6. OBJECTS AND THE OBJECT MODEL

26

Generated names do not initially correspond to an instance of an object. Objects with generated names are created by binding a generated name to the context. For example, a buffer object is created by calling the command BindBuffer with a name returned by GenBuffers, which allocates resources for the buffer object and its state, and associate the name with that object. Sampler objects may also be created by commands in addition to BindSampler, as described in section 8.2. A few types of objects are created by commands which return the name of the new object at the same time they create the object. Examples include CreateProgram for program objects and FenceSync for fence sync objects. 2.6.1.2

Name Deletion and Object Deletion

Objects are deleted by calling deletion commands specific to that object type. For example, the command DeleteBuffers is passed an array of buffer object names to delete. After an object is deleted it has no contents, and its name is once again marked unused for the purpose of name generation. If names are deleted that do not correspond to an object, but have been marked for the purpose of name generation, such names are marked as unused again. If unused and unmarked names are deleted they are silently ignored, as is the name zero. If an object is deleted while it is currently in use by a GL context, its name is immediately marked as unused, and some types of objects are automatically unbound from binding points in the current context, as described in section 5.1.2. However, the actual underlying object is not deleted until it is no longer in use. This situation is discussed in more detail in section 5.1.3. 2.6.1.3

Shared Object State

It is possible for groups of contexts to share some server state. Enabling such sharing between contexts is done through window system binding APIs such as those described in section 1.6.3. These APIs are responsible for creation and management of contexts, and are not discussed further here. More detailed discussion of the behavior of shared objects is included in chapter 5. Except as defined below for specific object types, all state in a context is specific to that context only.

2.6.2

Buffer Objects

The GL uses many types of data supplied by the client. Some of this data must be stored in server memory, and it is desirable to store other types of frequently used client data, such as vertex array and pixel data, in server memory for performance reasons, even if the option to store it in client memory exists.

OpenGL ES 3.2 (May 5, 2022)

2.6. OBJECTS AND THE OBJECT MODEL

27

Buffer objects contain a data store holding a fixed-sized allocation of server memory, and provide a mechanism to allocate, initialize, read from, and write to such memory. Buffer objects may be shared. They are described in detail in chapter 6.

2.6.3

Shader Objects

The source and/or binary code representing part or all of a shader program that is executed by one of the programmable stages defined by the GL (such as a vertex or fragment shader) is encapsulated in one or more shader objects. Shader objects may be shared. They are described in detail in chapter 7.

2.6.4

Program Objects

Shader objects that are to be used by one or more of the programmable stages of the GL are linked together to form a program object. The shader programs that are executed by these programmable stages are called executables. All information necessary for defining each executable is encapsulated in a program object. Program objects may be shared. They are described in detail in chapter 7.

2.6.5

Program Pipeline Objects

Program pipeline objects contain a separate program object binding point for each programmable stage. They allow a primitive to be processed by independent programs in each programmable stage, instead of requiring a single program object for each combination of shader operations. They allow greater flexibility when combining different shaders in various ways, without requiring a program object for each such combination. Program pipeline objects are container objects including references to program objects, and are not shared. They are described in detail in chapter 7.

2.6.6

Texture Objects

Texture objects or textures include data store containing a collection of texture images built from arrays of image elements referred to as texels. There are many types of texture objects varying by dimensionality and structure; the different texture types are described in detail in the introduction to chapter 8. Texture objects also include state describing the image parameters of the texture images, and state describing how sampling is performed when a shader accesses a texture.

OpenGL ES 3.2 (May 5, 2022)

2.6. OBJECTS AND THE OBJECT MODEL

28

Shaders may sample a texture at a location indicated by specified texture coordinates, with details of sampling determined by the sampler state of the texture. The resulting texture samples are typically used to modify a fragment’s color, in order to map an image onto a geometric primitive being drawn, but may be used for any purpose in a shader. Texture objects may be shared. They are described in detail in chapter 8.

2.6.7

Sampler Objects

Sampler objects contain the subset of texture object state controlling how sampling is performed when a shader accesses a texture. Sampler and texture objects may be bound together so that the sampler object state is used by shaders when sampling the texture, overriding equivalent state in the texture object. Separating texture image data from the method of sampling that data allows reuse of the same sampler state with many different textures without needing to set the sampler state in each texture. Sampler objects may be shared. They are described in detail in chapter 8.

2.6.8

Renderbuffer Objects

Renderbuffer objects contain a single image in a format which can be rendered to. Renderbuffer objects are attached to framebuffer objects (see below) when performing off-screen rendering. Renderbuffer objects may be shared. They are described in detail in chapter 9.

2.6.9

Framebuffer Objects

Framebuffer objects encapsulate the state of a framebuffer, including a collection of color, depth, and stencil buffers. Each such buffer is represented by a renderbuffer object or texture object attached to the framebuffer object. Framebuffer objects are container objects including references to renderbuffer and/or texture objects, and are not shared. They are described in detail in chapter 9.

2.6.10

Vertex Array Objects

Vertex array objects represent a collection of sets of vertex attributes. Each set is stored as an array in a buffer object data store, with each element of the array having a specified format and component count. The attributes of the currently bound vertex array object are used as inputs to the vertex shader when executing drawing commands.

OpenGL ES 3.2 (May 5, 2022)

2.6. OBJECTS AND THE OBJECT MODEL

29

Vertex array objects are container objects including references to buffer objects, and are not shared. They are described in detail in chapter 10.

2.6.11

Transform Feedback Objects

Transform feedback objects are used to capture attributes of the vertices of transformed primitives passed to the transform feedback stage when transform feedback mode is active. They include state required for transform feedback together with references to buffer objects in which attributes are captured. Transform feedback objects are container objects including references to buffer objects, and are not shared. They are described in detail in section 12.2.1.

2.6.12

Query Objects

Query objects return information about the processing of a sequence of GL commands, such as the number of primitives processed by drawing commands; the number of primitives written to transform feedback buffers; the number of samples that pass the depth test during fragment processing; and the amount of time required to process commands. Query objects are not shared. They are described in detail in section 4.2.

2.6.13

Sync Objects

A sync object acts as a synchronization primitive – a representation of events whose completion status can be tested or waited upon. Sync objects may be used for synchronization with operations occurring in the GL state machine or in the graphics pipeline, and for synchronizing between multiple graphics contexts, among other purposes. Sync objects may be shared. They are described in detail in section 4.1.

OpenGL ES 3.2 (May 5, 2022)

Chapter 3

Dataflow Model Figure 3.1 shows a block diagram of the GL. Some commands specify geometric objects to be drawn while others specify state controlling how objects are handled by the various stages, or specify data contained in textures and buffer objects. Commands are effectively sent through a processing pipeline. Different stages of the pipeline use data contained in different types of buffer objects. The first stage assembles vertices to form geometric primitives such as points, line segments, and polygons. In the next stage vertices may be transformed, followed by assembly into geometric primitives. Tessellation and geometry shaders may then generate multiple primitives from single input primitives. Optionally, the results of these pipeline stages may be fed back into buffer objects using transform feedback. The final resulting primitives are clipped to a clip volume in preparation for the next stage, rasterization. The rasterizer produces a series of framebuffer addresses and values using a two-dimensional description of a point, line segment, or polygon. Each fragment so produced is fed to the next stage that performs operations on individual fragments before they finally alter the framebuffer. These operations include conditional updates into the framebuffer based on incoming and previously stored depth values (to effect depth buffering), blending of incoming fragment colors with stored colors, as well as masking. Pixels may also be read back from the framebuffer or copied from one portion of the framebuffer to another. These transfers may include some type of decoding or encoding. Finally, compute shaders which may read from and write to buffer objects may be executed independently of the pipeline shown in figure 3.1. This ordering is meant only as a tool for describing the GL, not as a strict rule of how the GL is implemented, and we present it only as a means to organize the

30

31 various operations of the GL.

OpenGL ES 3.2 (May 5, 2022)

Vertex Buffer Object b

Figure 3.1. Block diagram of the OpenGL ES pipeline.

OpenGL ES 3.2 (May 5, 2022) t – Texture Binding

b – Buffer Binding

Programmable Stage

Fixed Function Stage

Legend

Transform Feedback Buffer b

Tessellation Control Shader

Pixel Pack

Pixel Operations

Per-Fragment Operations Framebuffer

Pixel Assembly

From Application

Uniform Block b

Texture Fetch t/b

Shader Storage b

Atomic Counter b

Image Load / Store t/b

Dispatch Indirect Buffer b

Fragment Shader

Rasterization

Transform Feedback

Geometry Shader

Tessellation Eval. Shader

Tessellation Primitive Gen.

Vertex Shader

Vertex Puller

Draw Indirect Buffer b

Element Array Buffer b

From Application

Pixel Pack Buffer b

Texture Image t

Pixel Unpack Buffer b

Compute Shader

Dispatch

From Application

32

Chapter 4

Event Model 4.1

Sync Objects and Fences

A sync object acts as a synchronization primitive – a representation of events whose completion status can be tested or waited upon. Sync objects may be used for synchronization with operations occurring in the GL state machine or in the graphics pipeline, and for synchronizing between multiple graphics contexts, among other purposes. Sync objects have a status value with two possible states: signaled and unsignaled. Events are associated with a sync object. When a sync object is created, its status is set to unsignaled. When the associated event occurs, the sync object is signaled (its status is set to signaled). The GL may be asked to wait for a sync object to become signaled. Initially, only one specific type of sync object is defined: the fence sync object, whose associated event is triggered by a fence command placed in the GL command stream. Fence sync objects are used to wait for partial completion of the GL command stream, as a more flexible form of Finish. The command sync FenceSync( enum condition, bitfield flags ); creates a new fence sync object, inserts a fence command in the GL command stream and associates it with that sync object, and returns a non-zero name corresponding to the sync object. When the specified condition of the sync object is satisfied by the fence command, the sync object is signaled by the GL, causing any ClientWaitSync or WaitSync commands (see below) blocking on sync to unblock. No other state is affected by FenceSync or by execution of the associated fence command. 33

4.1. SYNC OBJECTS AND FENCES

34

Property Name

Property Value

OBJECT_TYPE SYNC_CONDITION SYNC_STATUS SYNC_FLAGS

SYNC_FENCE

condition UNSIGNALED

flags

Table 4.1: Initial properties of a sync object created with FenceSync. condition must be SYNC_GPU_COMMANDS_COMPLETE. This condition is satisfied by completion of the fence command corresponding to the sync object and all preceding commands in the same command stream. The sync object will not be signaled until all effects from these commands on GL client and server state and the framebuffer are fully realized. Note that completion of the fence command occurs once the state of the corresponding sync object has been changed, but commands waiting on that sync object may not be unblocked until some time after the fence command completes. flags must be zero. Each sync object contains a number of properties which determine the state of the object and the behavior of any commands associated with it. Each property has a property name and property value. The initial property values for a sync object created by FenceSync are shown in table 4.1. Properties of a sync object may be queried with GetSynciv (see section 4.1.3). The SYNC_STATUS property will be changed to SIGNALED when condition is satisfied. Errors If FenceSync fails to create a sync object, zero will be returned and a GL error is generated. An INVALID_ENUM error is generated if condition is not SYNC_GPU_COMMANDS_COMPLETE. An INVALID_VALUE error is generated if flags is not zero. A sync object can be deleted by passing its name to the command void DeleteSync( sync sync ); If the fence command corresponding to the specified sync object has completed, or if no ClientWaitSync or WaitSync commands are blocking on sync, the object is deleted immediately. Otherwise, sync is flagged for deletion and will be OpenGL ES 3.2 (May 5, 2022)

4.1. SYNC OBJECTS AND FENCES

35

deleted when it is no longer associated with any fence command and is no longer blocking any ClientWaitSync or WaitSync command. In either case, after returning from DeleteSync the sync name is invalid and can no longer be used to refer to the sync object. DeleteSync will silently ignore a sync value of zero. Errors An INVALID_VALUE error is generated if sync is neither zero nor the name of a sync object.

4.1.1

Waiting for Sync Objects

The command enum ClientWaitSync( sync sync, bitfield flags, uint64 timeout ); causes the GL to block, and will not return until the sync object sync is signaled, or until the specified timeout period expires. timeout is in units of nanoseconds. timeout is adjusted to the closest value allowed by the implementation-dependent timeout accuracy, which may be substantially longer than one nanosecond, and may be longer than the requested period. If sync is signaled at the time ClientWaitSync is called, then ClientWaitSync returns immediately. If sync is unsignaled at the time ClientWaitSync is called, then ClientWaitSync will block and will wait up to timeout nanoseconds for sync to become signaled. flags controls command flushing behavior, and may be SYNC_FLUSH_COMMANDS_BIT, as discussed in section 4.1.2. ClientWaitSync returns one of four status values. A return value of ALREADY_SIGNALED indicates that sync was signaled at the time ClientWaitSync was called. ALREADY_SIGNALED will always be returned if sync was signaled, even if the value of timeout is zero. A return value of TIMEOUT_EXPIRED indicates that the specified timeout period expired before sync was signaled. A return value of CONDITION_SATISFIED indicates that sync was signaled before the timeout expired. Finally, if an error occurs, in addition to generating a GL error as specified below, ClientWaitSync immediately returns WAIT_FAILED without blocking. If the value of timeout is zero, then ClientWaitSync does not block, but simply tests the current state of sync. TIMEOUT_EXPIRED will be returned in this case if sync is not signaled, even though no actual wait was performed.

OpenGL ES 3.2 (May 5, 2022)

4.1. SYNC OBJECTS AND FENCES

36

Errors An INVALID_VALUE error is generated if sync is not the name of a sync object. An INVALID_VALUE error is generated if flags contains any bits other than SYNC_FLUSH_COMMANDS_BIT. The command void WaitSync( sync sync, bitfield flags, uint64 timeout ); is similar to ClientWaitSync, but instead of blocking and not returning to the application until sync is signaled, WaitSync returns immediately, instead causing the GL server to block1 until sync is signaled2 . sync has the same meaning as for ClientWaitSync. timeout must currently be the special value TIMEOUT_IGNORED, and is not used. Instead, WaitSync will always wait no longer than an implementationdependent timeout. The duration of this timeout in nanoseconds may be queried by calling GetInteger64v with the symbolic constant MAX_SERVER_WAIT_TIMEOUT. There is currently no way to determine whether WaitSync unblocked because the timeout expired or because the sync object being waited on was signaled. flags must be zero. If an error occurs, WaitSync generates a GL error as specified below, and does not cause the GL server to block. Errors An INVALID_VALUE error is generated if sync is not the name of a sync object. An INVALID_VALUE error is generated if timeout is not TIMEOUT_IGNORED or flags is not zeroa . a flags and timeout are placeholders for anticipated future extensions of sync object capabilities. They must have these reserved values in order that existing code calling WaitSync operate properly in the presence of such extensions. 1

The GL server may choose to wait either in the CPU executing server-side code, or in the GPU hardware if it supports this operation. 2 WaitSync allows applications to continue to queue commands from the client in anticipation of the sync being signaled, increasing client-server parallelism.

OpenGL ES 3.2 (May 5, 2022)

4.1. SYNC OBJECTS AND FENCES 4.1.1.1

37

Multiple Waiters

It is possible for both the GL client to be blocked on a sync object in a ClientWaitSync command, the GL server to be blocked as the result of a previous WaitSync command, and for additional WaitSync commands to be queued in the GL server, all for a single sync object. When such a sync object is signaled in this situation, the client will be unblocked, the server will be unblocked, and all such queued WaitSync commands will continue immediately when they are reached. See section 5.2 for more information about blocking on a sync object in multiple GL contexts.

4.1.2

Signaling

A fence sync object enters the signaled state only once the corresponding fence command has completed and signaled the sync object. If the sync object being blocked upon will not be signaled in finite time (for example, by an associated fence command issued previously, but not yet flushed to the graphics pipeline), then ClientWaitSync may hang forever. To help prevent this behavior3 , if ClientWaitSync is called and all of the following are true: • the SYNC_FLUSH_COMMANDS_BIT bit is set in flags, • sync is unsignaled when ClientWaitSync is called, • and the calls to ClientWaitSync and FenceSync were issued from the same context, then the GL will behave as if the equivalent of Flush were inserted immediately after the creation of sync. If a sync object is marked for deletion while a client is blocking on that object in a ClientWaitSync command, or a GL server is blocking on that object as a result of a prior WaitSync command, deletion is deferred until the sync object is signaled and all blocked GL clients and servers are unblocked. Additional constraints on the use of sync objects are discussed in chapter 5. State must be maintained to indicate which sync object names are currently in use. The state required for each sync object in use is an integer for the specific type, an integer for the condition, and a bit indicating whether the object is signaled 3

The simple flushing behavior defined by SYNC_FLUSH_COMMANDS_BIT will not help when waiting for a fence command issued in another context’s command stream to complete. Applications which block on a fence sync object must take additional steps to assure that the context from which the corresponding fence command was issued has flushed that command to the graphics pipeline.

OpenGL ES 3.2 (May 5, 2022)

4.1. SYNC OBJECTS AND FENCES

38

or unsignaled. The initial values of sync object state are defined as specified by FenceSync.

4.1.3

Sync Object Queries

Properties of sync objects may be queried using the command void GetSynciv( sync sync, enum pname, sizei count, sizei *length, int *values ); The value or values being queried are returned in the parameters length and values. On success, GetSynciv replaces up to count integers in values with the corresponding property values of the object being queried. The actual number of integers replaced is returned in *length. If length is NULL, no length is returned. If pname is OBJECT_TYPE, a single value representing the specific type of the sync object is placed in values. The only type supported is SYNC_FENCE. If pname is SYNC_STATUS, a single value representing the status of the sync object (SIGNALED or UNSIGNALED) is placed in values. If pname is SYNC_CONDITION, a single value representing the condition of the sync object is placed in values. The only condition supported is SYNC_GPU_COMMANDS_COMPLETE. If pname is SYNC_FLAGS, a single value representing the flags with which the sync object was created is placed in values. No flags are currently supported. Errors An INVALID_VALUE error is generated if sync is not the name of a sync object. An INVALID_ENUM error is generated if pname is not one of the values described above. An INVALID_VALUE error is generated if count is negative. The command boolean IsSync( sync sync ); returns TRUE if sync is the name of a sync object. If sync is not the name of a sync object, or if an error condition occurs, IsSync returns FALSE (note that zero is not the name of a sync object). Sync object names immediately become invalid after calling DeleteSync, as discussed in sections 4.1 and 5.2, but the underlying sync object will not be deleted OpenGL ES 3.2 (May 5, 2022)

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES Query target PRIMITIVES_GENERATED TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN ANY_SAMPLES_PASSED, ANY_SAMPLES_PASSED_CONSERVATIVE

39

Information Returned Number of primitives processed by the GL (see section 12.3) Number of primitives written to one or more buffer objects. There may be at most one active query of this type (see section 12.3). Boolean value set to TRUE when any fragments or samples pass the depth test. There may be at most one active query of this type (see section 15.1.4).

Table 4.2: Asynchronous query targets

until it is no longer associated with any fence command and no longer blocking any *WaitSync command.

4.2

Query Objects and Asynchronous Queries

Asynchronous queries provide a mechanism to return information about the processing of a sequence of GL commands. Query types supported by the GL are summarized in table 4.2. The results of asynchronous queries are not returned by the GL immediately after the completion of the last command in the set; subsequent commands can be processed while the query results are not complete. When available, the query results are stored in an associated query object. The commands described in section 4.2.1 provide mechanisms to determine when query results are available and return the actual results of the query. The name space for query objects is the unsigned integers, with zero reserved by the GL. The command void GenQueries( sizei n, uint *ids ); returns n previously unused query object names in ids. These names are marked as used, for the purposes of GenQueries only, but no object is associated with them until the first time they are used by BeginQuery. Errors An INVALID_VALUE error is generated if n is negative.

OpenGL ES 3.2 (May 5, 2022)

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

40

Query objects are deleted by calling void DeleteQueries( sizei n, const uint *ids ); ids contains n names of query objects to be deleted. After a query object is deleted, its name is again unused. If an active query object is deleted its name immediately becomes unused, but the underlying object is not deleted until it is no longer active (see section 5.1). Unused names in ids that have been marked as used for the purposes of GenQueries are marked as unused again. Unused names in ids are silently ignored, as is the value zero. Errors An INVALID_VALUE error is generated if n is negative. Each type of query supported by the GL has an active query object name. If an active query object name is non-zero, the GL is currently tracking the corresponding information, and the query results will be written into that query object. If an active query object name is zero, no such information is being tracked. A query object may be created and made active with the command void BeginQuery( enum target, uint id ); target indicates the type of query to be performed. The valid values of target are discussed in more detail in subsequent sections. BeginQuery sets the active query object name for target and index to id. If id is an unused query object name, the name is marked as used and associated with a new query object of the type specified by target. Otherwise id must be the name of an existing query object of that type. Note that occlusion query objects specified by either of the two targets ANY_SAMPLES_PASSED or ANY_SAMPLES_PASSED_CONSERVATIVE may be reused for either target in future queries. The state of the query object named id, whether newly created or not, is that the result is marked unavailable (the value of QUERY_RESULT_AVAILABLE for the query object is FALSE), and the result value (the value of QUERY_RESULT) is zero. Errors An INVALID_ENUM error is generated if target is not one of the valid targets listed in table 4.2 An INVALID_OPERATION error is generated if id is not a name returned from a previous call to GenQueries, or if such a name has since been deleted

OpenGL ES 3.2 (May 5, 2022)

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

41

with DeleteQueries. An INVALID_OPERATION error is generated if id is any of: • zero • the name of an existing query object whose type does not match target • an active query object name for any target. An INVALID_OPERATION error is generated if the active query object name for target is non-zero (for the targets ANY_SAMPLES_PASSED and ANY_SAMPLES_PASSED_CONSERVATIVE, if the active query for either target is non-zero). The command void EndQuery( enum target ); marks the end of the sequence of commands to be tracked for the active query specified by target. The corresponding active query object is updated to indicate that query results are not available, and the active query object name for target is reset to zero. When the commands issued prior to EndQuery have completed and a final query result is available, the query object active when EndQuery was called is updated to contain the query result and to indicate that the query result is available. target has the same meaning as for BeginQuery. Errors An INVALID_ENUM error is generated if target is not one of the valid targets listed in table 4.2 An INVALID_OPERATION error is generated if the active query object name for target is zero. Query objects contain two pieces of state: a single bit indicating whether a query result is available, and an integer containing the query result value. The number of bits, n, used to represent the query result depends on the query type as described in section 4.2.1. In the initial state of a query object, the result is not available (the flag is FALSE), and the result value is zero. If the query result overflows (exceeds the value 2n − 1), its value becomes undefined. It is recommended, but not required, that implementations handle this overflow case by saturating at 2n − 1 and incrementing no further. OpenGL ES 3.2 (May 5, 2022)

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

42

The necessary state for each possible active query target is an unsigned integer holding the active query object name (zero if no query object is active), and any state necessary to keep the current results of an asynchronous query in progress. Only a single type of occlusion query can be active at one time, so the required state for occlusion queries is shared.

4.2.1

Query Object Queries

The number of bits required to represent query results cannot be queried, but must be at least 1 bit for query targets ANY_SAMPLES_PASSED and ANY_SAMPLES_PASSED_CONSERVATIVE, and at least 32 bits for query target TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN. The command boolean IsQuery( uint id ); returns TRUE if id is the name of a query object. If id is zero, or if id is a non-zero value that is not the name of a query object, IsQuery returns FALSE. Information about an active query object can be queried with the command void GetQueryiv( enum target, enum pname, int *params ); target specifies the active query, and has the same meaning as for BeginQuery. If pname is CURRENT_QUERY, the name of the currently active query object for target, or zero if no query is active, will be placed in params. Errors An INVALID_ENUM error is generated if target is not one of the valid targets listed in table 4.2 An INVALID_ENUM error is generated if pname is not CURRENT_QUERY. The state of a query object can be queried with the commands void GetQueryObjectuiv( uint id, enum pname, uint *params ); id is the name of a query object. There may be an indeterminate delay before a query object’s result value is available. If pname is QUERY_RESULT_AVAILABLE, FALSE is returned if such a delay would be required; otherwise TRUE is returned. It must always be true that OpenGL ES 3.2 (May 5, 2022)

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

43

if any query object returns a result available of TRUE, all queries of the same type issued prior to that query must also return TRUE. Repeatedly querying QUERY_RESULT_AVAILABLE for any given query object is guaranteed to return TRUE eventually4 . If pname is QUERY_RESULT, then the query object’s result value is returned as a single integer in params. If the value is so large in magnitude that it cannot be represented with the requested type, then the nearest value representable using the requested type is returned. Querying QUERY_RESULT for any given query object forces that query to complete within a finite amount of time. If multiple queries are issued using the same object name prior to calling GetQueryObject*, the result and availability information returned will always be from the last query issued. The results from any queries before the last one will be lost if they are not retrieved before starting a new query on the same target and id. Errors An INVALID_OPERATION error is generated if id is not the name of a query object, or if the query object named by id is currently active. An INVALID_ENUM error is generated if pname is not QUERY_RESULT or QUERY_RESULT_AVAILABLE.

4

Note that multiple queries to the same occlusion object may result in a significant performance loss. For better performance it is recommended to wait N frames before querying this state. N is implementation-dependent but is generally between one and three.

OpenGL ES 3.2 (May 5, 2022)

Chapter 5

Shared Objects and Multiple Contexts This chapter describes special considerations for objects shared between multiple OpenGL ES contexts, including deletion behavior and how changes to shared objects are propagated between contexts. Objects that can be shared between contexts include buffer objects, program and shader objects, renderbuffer objects, sampler objects, sync objects, and texture objects (except for the texture objects named zero). Objects which contain references to other objects include framebuffer, program pipeline, transform feedback, and vertex array objects. Such objects are called container objects and are not shared. Implementations may allow sharing between contexts implementing different OpenGL ES versions. However, implementation-dependent behavior may result when aspects and/or behaviors of such shared objects do not apply to, and/or are not described by more than one version or profile.

5.1

Object Deletion Behavior

5.1.1

Side Effects of Shared Context Destruction

The share list is the group of all contexts which share objects. If a shared object is not explicitly deleted, then destruction of any individual context has no effect on that object unless it is the only remaining context in the share list. Once the last context on the share list is destroyed, all shared objects, and all other resources allocated for that context or share list, will be deleted and reclaimed by the implementation as soon as possible.

44

5.1. OBJECT DELETION BEHAVIOR

5.1.2

45

Automatic Unbinding of Deleted Objects

When a buffer, texture, transform feedback or renderbuffer object is successfully deleted, it is unbound from any bind points it is bound to in the current context, and detached from any attachments of container objects that are bound to the current context, as described for DeleteBuffers, DeleteTextures, DeleteTransformFeedbacks and DeleteRenderbuffers. If the object binding was established with other related state (such as a buffer range in BindBufferRange or selected level and layer information in FramebufferTexture or BindImageTexture), all such related state are restored to default values by the automatic unbind. Bind points in other contexts are not affected. Attachments to unbound container objects, such as deletion of a buffer attached to a vertex array object which is not bound to the context, are not affected and continue to act as references on the deleted object, as described in the following section.

5.1.3

Deleted Object and Object Name Lifetimes

When a buffer, query, renderbuffer, sampler, sync, or texture object is deleted, its name immediately becomes invalid (e.g. is marked unused), but the underlying object will not be deleted until it is no longer in use. A buffer, renderbuffer,sampler, or texture object is in use if any of the following conditions are satisfied: • the object is attached to any container object (such as a buffer object attached to a vertex array object, or a renderbuffer or texture attached to a framebuffer object) • the object is bound to a context bind point in any context A sync object is in use while there is a corresponding fence command which has not yet completed and signaled the sync object, or while there are any GL clients and/or servers blocked on the sync object as a result of ClientWaitSync or WaitSync commands. Query objects are in use so long as they are active, as described in section 4.2. When a shader object or program object is deleted, it is flagged for deletion, but its name remains valid until the underlying object can be deleted because it is no longer in use. A shader object is in use while it is attached to any program object. A program object is in use while it is attached to any program pipeline object or is a current program in any context. Caution should be taken when deleting an object attached to a container object, or a shared object bound in multiple contexts. Following its deletion, the object’s OpenGL ES 3.2 (May 5, 2022)

5.2. SYNC OBJECTS AND MULTIPLE CONTEXTS

46

name may be returned by Gen* commands, even though the underlying object state and data may still be referred to by container objects, or in use by contexts other than the one in which the object was deleted. Such a container or other context may continue using the object, and may still contain state identifying its name as being currently bound, until such time as the container object is deleted, the attachment point of the container object is changed to refer to another object, or another attempt to bind or attach the name is made in that context. Since the name is marked unused, binding the name will create a new object with the same name, and attaching the name will generate an error. The underlying storage backing a deleted object will not be reclaimed by the GL until all references to the object from container object attachment points or context binding points are removed.

5.2

Sync Objects and Multiple Contexts

When multiple GL clients and/or servers are blocked on a single sync object and that sync object is signalled, all such blocks are released. The order in which blocks are released is implementation-dependent.

5.3

Propagating Changes to Objects

GL objects contain two types of information, data and state. Collectively these are referred to below as the contents of an object. For the purposes of propagating changes to object contents as described below, data and state are treated consistently. Data is information the GL implementation does not have to inspect, and does not have an operational effect. Currently, data consists of: • Pixels in the framebuffer. • The contents of the data stores of buffer objects, renderbuffers, and textures. State determines the configuration of the rendering pipeline, and the GL implementation does have to inspect it. In hardware-accelerated GL implementations, state typically lives in GPU registers, while data typically lives in GPU memory. When the contents of an object T are changed, such changes are not always immediately visible, and do not always immediately affect GL operations involving that object. Changes may occur via any of the following means:

OpenGL ES 3.2 (May 5, 2022)

5.3. PROPAGATING CHANGES TO OBJECTS

47

• State-setting commands, such as TexParameter. • Data-setting commands, such as TexSubImage* or BufferSubData. • Data-setting through rendering to renderbuffers or textures attached to a framebuffer object. • Data-setting through transform feedback operations followed by an EndTransformFeedback command. • Commands that affect both state and data, such as TexImage* and BufferData. • Changes to mapped buffer data followed by a command such as UnmapBuffer or FlushMappedBufferRange. • Rendering commands that trigger shader invocations, where the shader performs image or buffer variable stores or atomic operations, or built-in atomic counter functions. When T is a texture, the contents of T are construed to include the contents of the data store of T.

5.3.1

Determining Completion of Changes to an object

The contents of an object T are considered to have been changed once a command such as described in section 5.3 has completed. Completion of a command1 may be determined either by calling Finish, or by calling FenceSync and executing a WaitSync command on the associated sync object. The second method does not require a round trip to the GL server and may be more efficient, particularly when changes to T in one context must be known to have completed before executing commands dependent on those changes in another context. In cases where a feedback loop has been established (see sections 8.6.1, 8.14.2.1, and 9.3, as well as the discussion of rule 1 below in section 5.3.3) the resulting contents of an object may be undefined.

5.3.2

Definitions

In the remainder of this section, the following terminology is used: 1 The GL already specifies that a single context processes commands in the order they are received. This means that a change to an object in a context at time t must be completed by the time a command issued in the same context at time t + 1 uses the result of that change.

OpenGL ES 3.2 (May 5, 2022)

5.3. PROPAGATING CHANGES TO OBJECTS

48

• An object T is directly attached to the current context if it has been bound to one of the context binding points. Examples include but are not limited to bound textures, bound framebuffers, bound vertex arrays, and current programs. • T is indirectly attached to the current context if it is attached to another object C, referred to as a container object, and C is itself directly or indirectly attached. Examples include but are not limited to renderbuffers or textures attached to framebuffers; buffers attached to vertex arrays; and shaders attached to programs. • An object T which is directly attached to the current context may be reattached by re-binding T at the same bind point. An object T which is indirectly attached to the current context may be re-attached by re-attaching the container object C to which T is attached. Corollary: re-binding C to the current context re-attaches C and its hierarchy of contained objects.

5.3.3

Rules

The following rules must be obeyed by all GL implementations: Rule 1 If the contents of an object T are changed in the current context while T is directly or indirectly attached, then all operations on T will use the new contents in the current context. Note: The intent of this rule is to address changes in a single context only. The multi-context case is handled by the other rules. Note: “Updates” via rendering or transform feedback are treated consistently with update via GL commands. Once EndTransformFeedback has been issued, any subsequent command in the same context that uses the results of the transform feedback operation will see the results. If a feedback loop is setup between rendering and transform feedback (see section 11.1.2.1), results will be undefined. Rule 2 While a container object C is bound, any changes made to the contents of C’s attachments in the current context are guaranteed to be seen. To guarantee seeing changes made in another context to objects attached to C, such changes must be completed in that other context (see section 5.3.1) prior to C being bound. Changes made in another context but not determined to have completed as described in section 5.3.1, or after C is bound in the current context, are not guaranteed to be seen.

OpenGL ES 3.2 (May 5, 2022)

5.3. PROPAGATING CHANGES TO OBJECTS

49

Rule 3 Changes to the contents of shared objects are not automatically propagated between contexts. If the contents of a shared object T are changed in a context other than the current context, and T is already directly or indirectly attached to the current context, any operations on the current context involving T via those attachments are not guaranteed to use its new contents. Rule 4 If the contents of an object T are changed in a context other than the current context, T must be attached or re-attached to at least one binding point in the current context, or at least one attachment point of a currently bound container object C, in order to guarantee that the new contents of T are visible in the current context. Note: “Attached or re-attached” means either attaching an object to a binding point it wasn’t already attached to, or attaching an object again to a binding point it was already attached to. Example: If a texture image is bound to multiple texture bind points and the texture is changed in another context, re-binding the texture at any one of the texture bind points is sufficient to cause the changes to be visible at all texture bind points.

OpenGL ES 3.2 (May 5, 2022)

Chapter 6

Buffer Objects Buffer objects contain a data store holding a fixed-sized allocation of server memory. This chapter specifies commands to create, manage, and destroy buffer objects. Specific types of buffer objects and their uses are briefly described together with references to their full specification. The name space for buffer objects is the unsigned integers, with zero reserved by the GL. The command void GenBuffers( sizei n, uint *buffers ); returns n previously unused buffer object names in buffers. These names are marked as used, for the purposes of GenBuffers only, but they acquire buffer state only when they are first bound with BindBuffer (see below), just as if they were unused. Errors An INVALID_VALUE error is generated if n is negative. Buffer objects are deleted by calling void DeleteBuffers( sizei n, const uint *buffers ); buffers contains n names of buffer objects to be deleted. After a buffer object is deleted it has no contents, and its name is again unused. If any portion of a buffer object being deleted is mapped in the current context or any context current to another thread, it is as though UnmapBuffer (see section 6.3.1) is executed in each such context prior to deleting the data store of the buffer. 50

6.1. CREATING AND BINDING BUFFER OBJECTS

51

Unused names in buffers that have been marked as used for the purposes of GenBuffers are marked as unused again. Unused names in buffers are silently ignored, as is the value zero. Errors An INVALID_VALUE error is generated if n is negative. The command boolean IsBuffer( uint buffer ); returns TRUE if buffer is the name of a buffer object. If buffer is zero, or if buffer is a non-zero value that is not the name of a buffer object, IsBuffer returns FALSE.

6.1

Creating and Binding Buffer Objects

A buffer object is created by binding an unused name to a buffer target. The binding is effected by calling void BindBuffer( enum target, uint buffer ); target must be one of the targets listed in table 6.1. If the buffer object named buffer has not been previously bound, or has been deleted since the last binding, the GL creates a new state vector, initialized with a zero-sized memory buffer and comprising all the state and with the same initial values listed in table 6.2. Buffer objects created by binding an unused name to any of the valid targets are formally equivalent. BindBuffer may also be used to bind an existing buffer object. If the bind is successful no change is made to the state of the newly bound buffer object, and any previous binding to target is broken. While a buffer object is bound, GL operations on the target to which it is bound affect the bound buffer object, and queries of the target to which a buffer object is bound return state from the bound object. Operations on the target also affect any other bindings of that object. If a buffer object is deleted while it is bound, all bindings to that object in the current context (i.e. in the thread that called DeleteBuffers) are reset to zero. Bindings to that buffer in other contexts are not affected, and the deleted buffer may continue to be used at any places it remains bound or attached, as described in appendix 5.1. Initially, each buffer object target is bound to zero. OpenGL ES 3.2 (May 5, 2022)

6.1. CREATING AND BINDING BUFFER OBJECTS

52

Target name

Described in section(s) 10.3.7 7.7 6.5 6.5 10.3.9 10.3.9 10.3.8 16.1, 20 8.4 7.8 8.9 12.2 7.6.2

ARRAY_BUFFER ATOMIC_COUNTER_BUFFER COPY_READ_BUFFER COPY_WRITE_BUFFER DISPATCH_INDIRECT_BUFFER DRAW_INDIRECT_BUFFER ELEMENT_ARRAY_BUFFER PIXEL_PACK_BUFFER PIXEL_UNPACK_BUFFER SHADER_STORAGE_BUFFER TEXTURE_BUFFER TRANSFORM_FEEDBACK_BUFFER UNIFORM_BUFFER

Purpose Vertex attributes Atomic counter storage Buffer copy source Buffer copy destination Indirect compute dispatch commands Indirect command arguments Vertex array indices Pixel read target Texture data source Read-write storage for shaders Texture data buffer Transform feedback buffer Uniform block storage

Table 6.1: Buffer object binding targets.

Name BUFFER_SIZE BUFFER_USAGE

Type int64 enum

BUFFER_ACCESS_FLAGS BUFFER_MAPPED BUFFER_MAP_POINTER BUFFER_MAP_OFFSET BUFFER_MAP_LENGTH

int boolean void* int64 int64

Initial Value 0 STATIC_DRAW

Legal Values any non-negative integer STREAM_DRAW, STREAM_READ, STREAM_COPY, STATIC_DRAW, STATIC_READ, STATIC_COPY, DYNAMIC_DRAW, DYNAMIC_READ, DYNAMIC_COPY

0 FALSE NULL

0 0

See section 6.3 TRUE, FALSE address any non-negative integer any non-negative integer

Table 6.2: Buffer object parameters and their values.

OpenGL ES 3.2 (May 5, 2022)

6.1. CREATING AND BINDING BUFFER OBJECTS

53

Errors An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. There is no buffer object corresponding to the name zero, so client attempts to modify or query buffer object state for a target bound to zero generate an INVALID_OPERATION error.

6.1.1

Binding Buffer Objects to Indexed Targets

Buffer objects may be created and bound to indexed targets by calling one of the commands void BindBufferRange( enum target, uint index, uint buffer, intptr offset, sizeiptr size ); void BindBufferBase( enum target, uint index, uint buffer ); ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BUFFER, TRANSFORM_FEEDBACK_BUFFER or UNIFORM_BUFFER. Ad-

target

must

be

ditional language specific to each target is included in sections referred to for each target in table 6.1. Each target represents an indexed array of buffer object binding points, as well as a single general binding point that can be used by other buffer object manipulation functions, such as BindBuffer or MapBufferRange. Both commands bind the buffer object named by buffer to both the general binding point, and to the binding point in the array given by index. If the binds are successful no change is made to the state of the bound buffer object, and any previous bindings to the general binding point or to the binding point in the array are broken. If the buffer object named buffer has not been previously bound, or has been deleted since the last binding, the GL creates a new state vector, initialized with a zero-sized memory buffer and comprising all the state and with the same initial values listed in table 6.2. For BindBufferRange, offset specifies a starting offset into the buffer object buffer, and size specifies the amount of data that can be read from or written to the buffer object while used as an indexed target. Both offset and size are in basic machine units. BindBufferBase binds the entire buffer, even when the size of the buffer is changed after the binding is established. The starting offset is zero, and the amount of data that can be read from or written to the buffer is determined by the size of the bound buffer at the time the binding is used. OpenGL ES 3.2 (May 5, 2022)

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

54

Regardless of the size specified with BindBufferRange, the GL will never read or write beyond the end of a bound buffer. In some cases this constraint may result in visibly different behavior when a buffer overflow would otherwise result, such as described for transform feedback operations in section 12.2.2. Errors An INVALID_ENUM error is generated if target is not one of the targets listed above. An INVALID_VALUE error is generated if index is greater than or equal to the number of target-specific indexed binding points, as described in section 6.6.1. An INVALID_VALUE error is generated by BindBufferRange if buffer is non-zero and offset is negative. An INVALID_VALUE error is generated by BindBufferRange if buffer is non-zero and size is less than or equal to zero. An INVALID_VALUE error is generated by BindBufferRange if buffer is non-zero and offset or size do not respectively satisfy the constraints described for those parameters for the specified target, as described in section 6.6.1.

6.2

Creating and Modifying Buffer Object Data Stores

The data store of a buffer object is created and initialized by calling void BufferData( enum target, sizeiptr size, const void *data, enum usage ); with target set to one of the targets listed in table 6.1, size set to the size of the data store in basic machine units, and data pointing to the source data in client memory. If data is non-NULL, then the source data is copied to the buffer object’s data store. If data is NULL, then the contents of the buffer object’s data store are undefined. usage is specified as one of nine enumerated values, indicating the expected application usage pattern of the data store. In the following descriptions, a buffer’s data store is sourced when it is read from as a result of GL commands which specify images, or invoke shaders accessing buffer data as a result of drawing commands or compute shader dispatch. The values are: STREAM_DRAW The data store contents will be specified once by the application,

and sourced at most a few times. OpenGL ES 3.2 (May 5, 2022)

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES Name BUFFER_SIZE BUFFER_USAGE BUFFER_ACCESS_FLAGS BUFFER_MAPPED BUFFER_MAP_POINTER BUFFER_MAP_OFFSET BUFFER_MAP_LENGTH

55

Value size usage 0 FALSE NULL

0 0

Table 6.3: Buffer object initial state.

STREAM_READ The data store contents will be specified once by reading data from

the GL, and queried at most a few times by the application. STREAM_COPY The data store contents will be specified once by reading data from

the GL, and sourced at most a few times STATIC_DRAW The data store contents will be specified once by the application,

and sourced many times. STATIC_READ The data store contents will be specified once by reading data from

the GL, and queried many times by the application. STATIC_COPY The data store contents will be specified once by reading data from

the GL, and sourced many times. DYNAMIC_DRAW The data store contents will be respecified repeatedly by the ap-

plication, and sourced many times. DYNAMIC_READ The data store contents will be respecified repeatedly by reading

data from the GL, and queried many times by the application. DYNAMIC_COPY The data store contents will be respecified repeatedly by reading

data from the GL, and sourced many times. usage is provided as a performance hint only. The specified usage value does not constrain the actual usage pattern of the data store. BufferData deletes any existing data store, and sets the values of the buffer object’s state variables as shown in table 6.3. If any portion of the buffer object is mapped in the current context or any context current to another thread, it is as though UnmapBuffer (see section 6.3.1) is executed in each such context prior to deleting the existing data store. OpenGL ES 3.2 (May 5, 2022)

6.3. MAPPING AND UNMAPPING BUFFER DATA

56

Clients must align data elements consistently with the requirements of the client platform, with an additional base-level requirement that an offset within a buffer to a datum comprising N basic machine units be a multiple of N . Errors An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_VALUE error is generated if size is negative. An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. An INVALID_ENUM error is generated if usage is not one of the nine usages described above. To modify some or all of the data contained in a buffer object’s data store, the client may use the command void BufferSubData( enum target, intptr offset, sizeiptr size, const void *data ); with target set to one of the targets listed in table 6.1. offset and size indicate the range of data in the buffer object that is to be replaced, in terms of basic machine units. data specifies a region of client memory size basic machine units in length, containing the data that replace the specified buffer range. Errors An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. An INVALID_VALUE error is generated if offset or size is negative, or if offset + size is greater than the value of BUFFER_SIZE for the buffer bound to target. An INVALID_OPERATION error is generated if any part of the specified buffer range is mapped with MapBufferRange (see section 6.3).

6.3

Mapping and Unmapping Buffer Data

All or part of the data store of a buffer object may be mapped into the client’s address space by calling

OpenGL ES 3.2 (May 5, 2022)

6.3. MAPPING AND UNMAPPING BUFFER DATA

57

void *MapBufferRange( enum target, intptr offset, sizeiptr length, bitfield access ); with target set to one of the targets listed in table 6.1. offset and length indicate the range of data in the buffer object that is to be mapped, in terms of basic machine units. access is a bitfield containing flags which describe the requested mapping. These flags are described below. If no error occurs, a pointer to the beginning of the mapped range is returned once all pending operations on that buffer have completed, and may be used to modify and/or query the corresponding range of the buffer, according to the following flag bits set in access: • MAP_READ_BIT indicates that the returned pointer may be used to read buffer object data. No GL error is generated if the pointer is used to query a mapping which excludes this flag, but the result is undefined and system errors (possibly including program termination) may occur. • MAP_WRITE_BIT indicates that the returned pointer may be used to modify buffer object data. No GL error is generated if the pointer is used to modify a mapping which excludes this flag, but the result is undefined and system errors (possibly including program termination) may occur. Pointer values returned by MapBufferRange may not be passed as parameter values to GL commands. For example, they may not be used to specify array pointers, or to specify or query pixel or texture image data; such actions produce undefined results, although implementations may not check for such behavior for performance reasons. Mappings to the data stores of buffer objects may have nonstandard performance characteristics. For example, such mappings may be marked as uncacheable regions of memory, and in such cases reading from them may be very slow. To ensure optimal performance, the client should use the mapping in a fashion consistent with the values of BUFFER_USAGE and access. Using a mapping in a fashion inconsistent with these values is liable to be multiple orders of magnitude slower than using normal memory. The following optional flag bits in access may be used to modify the mapping: • MAP_INVALIDATE_RANGE_BIT indicates that the previous contents of the specified range may be discarded. Data within this range are undefined with the exception of subsequently written data. No GL error is generated if subsequent GL operations access unwritten data, but the result is undefined and system errors (possibly including program termination) may occur. This flag may not be used in combination with MAP_READ_BIT. OpenGL ES 3.2 (May 5, 2022)

6.3. MAPPING AND UNMAPPING BUFFER DATA Name BUFFER_ACCESS_FLAGS BUFFER_MAPPED BUFFER_MAP_POINTER BUFFER_MAP_OFFSET BUFFER_MAP_LENGTH

58

Value access TRUE

pointer to the data store offset length

Table 6.4: Buffer object state set by MapBufferRange.

• MAP_INVALIDATE_BUFFER_BIT indicates that the previous contents of the entire buffer may be discarded. Data within the entire buffer are undefined with the exception of subsequently written data. No GL error is generated if subsequent GL operations access unwritten data, but the result is undefined and system errors (possibly including program termination) may occur. This flag may not be used in combination with MAP_READ_BIT. • MAP_FLUSH_EXPLICIT_BIT indicates that one or more discrete subranges of the mapping may be modified. When this flag is set, modifications to each subrange must be explicitly flushed by calling FlushMappedBufferRange. No GL error is set if a subrange of the mapping is modified and not flushed, but data within the corresponding subrange of the buffer are undefined. This flag may only be used in conjunction with MAP_WRITE_BIT. When this option is selected, flushing is strictly limited to regions that are explicitly indicated with calls to FlushMappedBufferRange prior to unmap; if this option is not selected UnmapBuffer will automatically flush the entire mapped range when called. • MAP_UNSYNCHRONIZED_BIT indicates that the GL should not attempt to synchronize pending operations on the buffer prior to returning from MapBufferRange. No GL error is generated if pending operations which source or modify the buffer overlap the mapped region, but the result of such previous and any subsequent operations is undefined. If MAP_UNSYNCHRONIZED_BIT is used in conjunction with either MAP_INVALIDATE_BUFFER_BIT or MAP_INVALIDATE_RANGE_BIT, the GL may discard previous contents without attempting to synchronize pending operations on the buffer. A successful MapBufferRange sets buffer object state values as shown in table 6.4. OpenGL ES 3.2 (May 5, 2022)

6.3. MAPPING AND UNMAPPING BUFFER DATA

59

Errors If an error occurs, MapBufferRange returns a NULL pointer. An INVALID_VALUE error is generated if offset or length is negative, if offset + length is greater than the value of BUFFER_SIZE, or if access has any bits set other than those defined above. An INVALID_OPERATION error is generated for any of the following conditions: • length is zero. • The buffer is already in a mapped state. • Neither MAP_READ_BIT nor MAP_WRITE_BIT is set. • MAP_READ_BIT is set and any of MAP_INVALIDATE_RANGE_BIT, MAP_INVALIDATE_BUFFER_BIT, or MAP_UNSYNCHRONIZED_BIT is set. • MAP_FLUSH_EXPLICIT_BIT is set and MAP_WRITE_BIT is not set. No error is generated if memory outside the mapped range is modified or queried, but the result is undefined and system errors (possibly including program termination) may occur. If a buffer is mapped with the MAP_FLUSH_EXPLICIT_BIT flag, modifications to the mapped range may be indicated by calling void FlushMappedBufferRange( enum target, intptr offset, sizeiptr length ); with target set to one of the targets listed in table 6.1. offset and length indicate a modified subrange of the mapping, in basic machine units. The specified subrange to flush is relative to the start of the currently mapped range of buffer. FlushMappedBufferRange may be called multiple times to indicate distinct subranges of the mapping which require flushing. Errors An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated if zero is bound to target.

OpenGL ES 3.2 (May 5, 2022)

6.3. MAPPING AND UNMAPPING BUFFER DATA

60

An INVALID_OPERATION error is generated if the buffer bound to target is not mapped, or is mapped without the MAP_FLUSH_EXPLICIT_BIT flag. An INVALID_VALUE error is generated if offset or length is negative, or if offset + length exceeds the size of the mapping.

6.3.1

Unmapping Buffers

After the client has specified the contents of a mapped buffer range, and before the data in that range are dereferenced by any GL commands, the mapping must be relinquished by calling boolean UnmapBuffer( enum target ); with target set to one of the targets listed in table 6.1. Unmapping a mapped buffer object invalidates the pointer to its data store and sets the object’s BUFFER_MAPPED, BUFFER_MAP_POINTER, BUFFER_ACCESS_FLAGS, BUFFER_MAP_OFFSET, and BUFFER_MAP_LENGTH state variables to the initial values shown in table 6.3. UnmapBuffer returns TRUE unless data values in the buffer’s data store have become corrupted during the period that the buffer was mapped. Such corruption can be the result of a screen resolution change or other window system-dependent event that causes system heaps such as those for high-performance graphics memory to be discarded. GL implementations must guarantee that such corruption can occur only during the periods that a buffer’s data store is mapped. If such corruption has occurred, UnmapBuffer returns FALSE, and the contents of the buffer’s data store become undefined. Unmapping that occurs as a side effect of buffer deletion (see section 5.1.2) or reinitialization by BufferData is not an error. Buffer mappings are buffer object state, and are not affected by whether or not a context owing a buffer object is current. Errors An INVALID_OPERATION error is generated if the buffer data store is already in the unmapped state, and FALSE is returned.

6.3.2

Effects of Mapping Buffers on Other GL Commands

Any GL command which attempts to read from, write to, or change the state of a buffer object may generate an INVALID_OPERATION error if all or part of the buffer object is mapped. However, only commands which explicitly describe this OpenGL ES 3.2 (May 5, 2022)

6.4. EFFECTS OF ACCESSING OUTSIDE BUFFER BOUNDS

61

error are required to do so. If an error is not generated, using such commands to perform invalid reads, writes, or state changes will have undefined results and may result in GL interruption or termination.

6.4

Effects of Accessing Outside Buffer Bounds

Most, but not all GL commands operating on buffer objects will detect attempts to read from or write to a location in a bound buffer object at an offset less than zero, or greater than or equal to the buffer’s size. When such an attempt is detected, a GL error is generated. Any command which does not detect these attempts, and performs such an invalid read or write has undefined results, and may result in GL interruption or termination.

6.5

Copying Between Buffers

All or part of the data store of a buffer object may be copied to the data store of another buffer object by calling void CopyBufferSubData( enum readtarget, enum writetarget, intptr readoffset, intptr writeoffset, sizeiptr size ); with readtarget and writetarget each set to one of the targets listed in table 6.1. While any of these targets may be used, the COPY_READ_BUFFER and COPY_WRITE_BUFFER targets are provided specifically for copies, so that they can be done without affecting other buffer binding targets that may be in use. writeoffset and size specify the range of data in the buffer object bound to writetarget that is to be replaced, in terms of basic machine units. readoffset and size specify the range of data in the buffer object bound to readtarget that is to be copied to the corresponding region of writetarget. Errors An INVALID_VALUE error is generated if any of readoffset, writeoffset, or size are negative, if readoffset + size exceeds the size of the buffer object bound to readtarget, or if writeoffset + size exceeds the size of the buffer object bound to writetarget. An INVALID_VALUE error is generated if the same buffer object is bound to both readtarget and writetarget, and the ranges [readoffset, readoffset + size) and [writeoffset, writeoffset + size) overlap.

OpenGL ES 3.2 (May 5, 2022)

6.6. BUFFER OBJECT QUERIES

62

An INVALID_OPERATION error is generated if zero is bound to readtarget or writetarget. An INVALID_OPERATION error is generated if the buffer objects bound to either readtarget or writetarget are mapped

6.6

Buffer Object Queries

The commands void GetBufferParameteriv( enum target, enum pname, int *data ); void GetBufferParameteri64v( enum target, enum pname, int64 *data ); return information about a bound buffer object. target must be one of the targets listed in table 6.1, and pname must be one of the buffer object parameters in table 6.2, other than BUFFER_MAP_POINTER. The value of the specified parameter of the buffer object bound to target is returned in data. Errors An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_ENUM error is generated if pname is not one of the buffer object parameters other than BUFFER_MAP_POINTER. While part or all of the data store of a buffer object is mapped, the pointer to the mapped range of the data store can be queried by calling void GetBufferPointerv( enum target, enum pname, void **params ); with target set to one of the targets listed in table 6.1 and pname set to BUFFER_MAP_POINTER. The single buffer map pointer is returned in params. GetBufferPointerv returns the NULL pointer value if the buffer’s data store is not currently mapped, or if the requesting client did not map the buffer object’s data store, and the implementation is unable to support mappings on multiple clients.

OpenGL ES 3.2 (May 5, 2022)

6.6. BUFFER OBJECT QUERIES

63

Errors An INVALID_ENUM error is generated if target is not one of the targets listed in table 6.1. An INVALID_ENUM error is generated if pname is not BUFFER_MAP_POINTER. An INVALID_OPERATION error is generated if zero is bound to target.

6.6.1

Indexed Buffer Object Limits and Binding Queries

Several types of buffer bindings support an indexed array of binding points for specific use by the GL, in addition to a single generic binding point for general management of buffers of that type. Each type of binding is described in table 6.5 together with the token names used to refer to each buffer in the array of binding points, the starting offset of the binding for each buffer in the array, any constraints on the corresponding offset value passed to BindBufferRange (see section 6.1.1), the size of the binding for each buffer in the array, any constraints on the corresponding size value passed to BindBufferRange, and the size of the array (the number of bind points supported). To query which buffer objects are bound to an indexed array, call GetIntegeri v with target set to the name of the array binding points. index must be in the range zero to the number of bind points supported minus one. The name of the buffer object bound to index is returned in values. If no buffer object is bound for index, zero is returned in values. To query the starting offset or size of the range of a buffer object binding in an indexed array, call GetInteger64i v with target set to respectively the starting offset or binding size name from table 6.5 for that array. index must be in the range zero to the number of bind points supported minus one. If the starting offset or size was not specified when the buffer object was bound (e.g. if it was bound with BindBufferBase), or if no buffer object is bound to the target array at index, zero is returned1 . Errors An INVALID_VALUE error is generated by GetIntegeri v and GetInteger64i v if target is one of the array binding point names, starting offset names, or binding size names from table 6.5 and index is greater than or equal 1

A zero size is a sentinel value indicating that the actual binding range size is determined by the size of the bound buffer at the time the binding is used.

OpenGL ES 3.2 (May 5, 2022)

6.6. BUFFER OBJECT QUERIES

64

Atomic counter array bindings (see sec. 7.7.2) binding points ATOMIC_COUNTER_BUFFER_BINDING starting offset ATOMIC_COUNTER_BUFFER_START offset restriction multiple of 4 binding size ATOMIC_COUNTER_BUFFER_SIZE size restriction none no. of bind points value of MAX_ATOMIC_COUNTER_BUFFER_BINDINGS

Shader storage array bindings (see sec. 7.8) binding points SHADER_STORAGE_BUFFER_BINDING starting offset SHADER_STORAGE_BUFFER_START offset restriction multiple of value of SHADER_STORAGE_binding size size restriction no. of bind points

BUFFER_OFFSET_ALIGNMENT SHADER_STORAGE_BUFFER_SIZE

none value

of

MAX_SHADER_STORAGE_BUFFER_-

BINDINGS

Transform feedback array bindings (see sec. 12.2.2) binding points TRANSFORM_FEEDBACK_BUFFER_BINDING starting offset TRANSFORM_FEEDBACK_BUFFER_START offset restriction multiple of 4 binding size TRANSFORM_FEEDBACK_BUFFER_SIZE size restriction multiple of 4 no. of bind points value of MAX_TRANSFORM_FEEDBACK_SEPARATE_ATTRIBS

Uniform buffer array bindings (see sec. 7.6.3) binding points UNIFORM_BUFFER_BINDING starting offset UNIFORM_BUFFER_START offset restriction multiple of value of UNIFORM_BUFFER_binding size size restriction no. of bind points

OFFSET_ALIGNMENT UNIFORM_BUFFER_SIZE

none value of MAX_UNIFORM_BUFFER_BINDINGS

Table 6.5: Indexed buffer object limits and binding queries

OpenGL ES 3.2 (May 5, 2022)

6.7. BUFFER OBJECT STATE

65

to the number of binding points for target as described in the same table.

6.7

Buffer Object State

The state required to support buffer objects consists of binding names for each of the buffer targets in table 6.1, and for each of the indexed buffer targets in section 6.1.1. The state required for index buffer targets for atomic counters, shader storage, transform feedback, and uniform buffer array bindings is summarized in tables 21.32, 21.34, 21.35 and 21.36, respectively. Additionally, each vertex array has an associated binding so there is a buffer object binding for each of the vertex attribute arrays. The initial values for all buffer object bindings is zero. The state of each buffer object consists of a buffer size in basic machine units, a usage parameter, an access parameter, a mapped boolean, two integers for the offset and size of the mapped region, a pointer to the mapped buffer (NULL if unmapped), and the sized array of basic machine units for the buffer data.

OpenGL ES 3.2 (May 5, 2022)

Chapter 7

Programs and Shaders This chapter specifies commands to create, manage, and destroy program and shader objects. Commands and functionality applicable only to specific shader stages (for example, vertex attributes used as inputs by vertex shaders) are described together with those stages in chapters 10 and 14. A shader specifies operations that are meant to occur on data as it moves through different programmable stages of the OpenGL ES processing pipeline, starting with vertices specified by the application and ending with fragments prior to being written to the framebuffer. The programming language used for shaders is described in the OpenGL ES Shading Language Specification. To use a shader, shader source code is first loaded into a shader object and then compiled. A shader object corresponds to a stage in the rendering pipeline referred to as its shader stage or shader type. Alternatively, pre-compiled shader binary code may be directly loaded into a shader object. An implementation must support shader compilation (the boolean value SHADER_COMPILER must be TRUE). If the integer value of NUM_SHADER_BINARY_FORMATS is greater than zero, then shader binary loading is supported. One or more shader objects are attached to a program object. The program object is then linked, which generates executable code from all the compiled shader objects attached to the program. Alternatively, pre-compiled program binary code may be directly loaded into a program object (see section 7.5). When program objects are bound to a shader stage, they become the current program object for that stage. When the current program object for a shader stage includes a shader of that type, it is considered the active program object for that stage. The current program object for all stages may be set at once using a single unified program object, or the current program object may be set for each stage

66

7.1. SHADER OBJECTS

67

individually using a separable program object where different separable program objects may be current for other stages. The set of separable program objects current for all stages are collected in a program pipeline object that must be bound for use. When a linked program object is made active for one of the stages, the corresponding executable code is used to perform processing for that stage. Shader stages including vertex shaders, tessellation control shaders, tessellation evaluation shaders, geometry shaders, fragment shaders, and compute shaders can be created, compiled, and linked into program objects. Vertex shaders describe the operations that occur on vertex attributes. Tessellation control and evaluation shaders are used to control the operation of the tessellator (see section 11.2). Geometry shaders affect the processing of primitives assembled from vertices (see section 11.3). Fragment shaders affect the processing of fragments during rasterization (see section 14). A single program object can contain all of these shaders, or any subset thereof. Compute shaders perform general-purpose computation for dispatched arrays of shader invocations (see section 17), but do not operate on primitives processed by the other shader types. Shaders can reference several types of variables as they execute. Uniforms are per-program variables that are constant during program execution (see section 7.6). Buffer variables (see section 7.8) are similar to uniforms, but are stored in buffer object memory which may be written to, and is persistent across multiple shader invocations. Samplers (see section 7.9) are a special form of uniform used for texturing (see chapter 8). Images (see section 7.10) are a special form of uniform identifying a level of a texture to be accessed using built-in shader functions as described in section 8.23. Output variables hold the results of shader execution that are used later in the pipeline. Each of these variable types is described in more detail below.

7.1

Shader Objects

The name space for shader objects is the unsigned integers, with zero reserved for the GL. This name space is shared with program objects. The following sections define commands that operate on shader and program objects. To create a shader object, use the command uint CreateShader( enum type ); The shader object is empty when it is created. The type argument specifies the type of shader object to be created and must be one of the values in table 7.1 indicating

OpenGL ES 3.2 (May 5, 2022)

7.1. SHADER OBJECTS type VERTEX_SHADER TESS_CONTROL_SHADER TESS_EVALUATION_SHADER GEOMETRY_SHADER FRAGMENT_SHADER COMPUTE_SHADER

68 Shader Stage Vertex shader Tessellation control shader Tessellation evaluation shader Geometry shader Fragment shader Compute shader

Table 7.1: CreateShader type values and the corresponding shader stages.

the corresponding shader stage. A non-zero name that can be used to reference the shader object is returned. Errors An INVALID_ENUM error is generated and zero is returned if type is not one of the values in table 7.1, The command void ShaderSource( uint shader, sizei count, const char * const *string, const int *length ); loads source code into the shader object named shader. string is an array of count pointers to optionally null-terminated character strings that make up the source code. The length argument is an array with the number of chars in each string (the string length). If an element in length is negative, its accompanying string is nullterminated. If length is NULL, all strings in the string argument are considered nullterminated. The ShaderSource command sets the source code for the shader to the text strings in the string array. If shader previously had source code loaded into it, the existing source code is completely replaced. Any length passed in excludes the null terminator in its count. The strings that are loaded into a shader object are expected to form the source code for a valid shader as defined in the OpenGL ES Shading Language Specification. Errors An INVALID_VALUE error is generated if shader is not the name of either

OpenGL ES 3.2 (May 5, 2022)

7.1. SHADER OBJECTS

69

a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. An INVALID_VALUE error is generated if count is negative. Once the source code for a shader has been loaded, a shader object can be compiled with the command void CompileShader( uint shader ); Each shader object has a boolean status, COMPILE_STATUS, that is modified as a result of compilation. This status can be queried with GetShaderiv (see section 7.12). This status will be set to TRUE if shader was compiled without errors and is ready for use, and FALSE otherwise. Compilation can fail for a variety of reasons as listed in the OpenGL ES Shading Language Specification. If CompileShader failed, any information about a previous compile is lost. Thus a failed compile does not restore the old state of shader. Changing the source code of a shader object with ShaderSource does not change its compile status or the compiled shader code. Each shader object has an information log, which is a text string that is overwritten as a result of compilation. This information log can be queried with GetShaderInfoLog to obtain more information about the compilation attempt (see section 7.12). Errors An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. Resources allocated by the shader compiler may be released with the command void ReleaseShaderCompiler( void ); This is a hint from the application, and does not prevent later use of the shader compiler. If shader source is loaded and compiled after ReleaseShaderCompiler has been called, CompileShader must succeed provided there are no errors in the shader source. The range and precision for different numeric formats supported by the shader compiler may be determined with the command GetShaderPrecisionFormat (see section 7.12). Shader objects can be deleted with the command OpenGL ES 3.2 (May 5, 2022)

7.2. SHADER BINARIES

70

void DeleteShader( uint shader ); If shader is not attached to any program object, it is deleted immediately. Otherwise, shader is flagged for deletion and will be deleted when it is no longer attached to any program object. If an object is flagged for deletion, its boolean status bit DELETE_STATUS is set to true. The value of DELETE_STATUS can be queried with GetShaderiv (see section 7.12). DeleteShader will silently ignore the value zero. Errors An INVALID_VALUE error is generated if shader is neither zero nor the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is not zero and is the name of a program object. The command boolean IsShader( uint shader ); returns TRUE if shader is the name of a shader object. If shader is zero, or a nonzero value that is not the name of a shader object, IsShader returns FALSE. No error is generated if shader is not a valid shader object name.

7.2

Shader Binaries

Precompiled shader binaries may be loaded with the command void ShaderBinary( sizei count, const uint *shaders, enum binaryformat, const void *binary, sizei length ); shaders contains a list of count shader object handles. Each handle refers to a unique shader type, and may correspond to any of the shader stages in table 7.1. binary points to length bytes of pre-compiled binary shader code in client memory, and binaryformat denotes the format of the pre-compiled code. The binary image will be decoded according to the extension specification defining the specified binaryformat. OpenGL ES defines no specific binary formats, but does provide a mechanism to obtain token values for such formats provided by extensions. The number of shader binary formats supported can be obtained by querying the value of NUM_SHADER_BINARY_FORMATS. The list of specific binary formats supported can be obtained by querying the value of SHADER_BINARY_FORMATS. OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

71

Depending on the types of the shader objects in shaders, ShaderBinary will individually load binary shaders, or load an executable binary that contains an optimized set of shaders stored in the same binary. Errors An INVALID_VALUE error is generated if count or length is negative. An INVALID_ENUM error is generated if binaryformat is not a supported format returned in SHADER_BINARY_FORMATS. An INVALID_VALUE error is generated if the data pointed to by binary does not match the specified binaryformat. An INVALID_VALUE error is generated if any of the handles in shaders is not the name of either a program or shader object. An INVALID_OPERATION error is generated if any of the handles in shaders is the name of a program object. An INVALID_OPERATION error is generated if more than one of the handles in shaders refers to the same type of shader object. Additional errors corresponding to specific binary formats may be generated as specified by the extensions defining those formats. If ShaderBinary succeeds, the COMPILE_STATUS of the shader is set to TRUE. If ShaderBinary fails, the old state of shader objects for which the binary was being loaded will not be restored. Note that if shader binary interfaces are supported, then a GL implementation may require that an optimized set of shader binaries that were compiled together be specified to LinkProgram. Not specifying an optimized set may cause LinkProgram to fail.

7.3

Program Objects

A program object is created with the command uint CreateProgram( void ); Program objects are empty when they are created. A non-zero name that can be used to reference the program object is returned. If an error occurs, zero will be returned. To attach a shader object to a program object, use the command void AttachShader( uint program, uint shader ); OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

72

Shader objects may be attached to program objects before source code has been loaded into the shader object, or before the shader object has been compiled. Multiple shader objects of the same type may not be attached to a single program object. However, a single shader object may be attached to more than one program object. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. An INVALID_OPERATION error is generated if shader is already attached to program, or if another shader object of the same type as shader is already attached to program. To detach a shader object from a program object, use the command void DetachShader( uint program, uint shader ); If shader has been flagged for deletion and is not attached to any other program object, it is deleted. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. An INVALID_OPERATION error is generated if shader is not attached to program. In order to use the shader objects contained in a program object, the program object must be linked. The command OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

73

void LinkProgram( uint program ); will link the program object named program. Each program object has a boolean status, LINK_STATUS, that is modified as a result of linking. This status can be queried with GetProgramiv (see section 7.12). This status will be set to TRUE if a valid executable is created, and FALSE otherwise. Linking can fail for a variety of reasons as specified in the OpenGL ES Shading Language Specification, as well as any of the following reasons: • No shader objects are attached to program. • One or more of the shader objects attached to program are not compiled successfully. • More active uniform or active sampler variables are used in program than allowed (see sections 7.6, 7.9, and 11.3.3). • program contains objects to form either a vertex shader or fragment shader, and – program is not separable, and does not contain objects to form both a vertex shader and fragment shader. • program contains an object to form a tessellation control shader (see section 11.2.1), and – the program is not separable and contains no object to form a vertex shader; or – the program is not separable and contains no object to form a tessellation evaluation shader; or – the output patch vertex count is not specified in the compiled tessellation control shader object. • program contains an object to form a tessellation evaluation shader (see section 11.2.3), and – the program is not separable and contains no object to form a vertex shader; or – the program is not separable and contains no object to form a tessellation control shader; or

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

74

– the tessellation primitive mode is not specified in the compiled tessellation evaluation shader object. • program contains objects to form a geometry shader (see section 11.3), and – program is not separable and contains no objects to form a vertex shader; or – the input primitive type, output primitive type, or maximum output vertex count is not specified in the compiled geometry shader object. • program contains objects to form a compute shader (see section 17) and – program also contains objects to form any other type of shader. • The shaders do not use the same shader language version. If LinkProgram failed, any information about a previous link of that program object is lost. Thus, a failed link does not restore the old state of program. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. When program objects which have been linked successfully are used for rendering operations, they may access GL state and interface with other stages of the GL pipeline through active variables and active interface blocks. The GL provides various commands allowing applications to enumerate and query properties of active variables and interface blocks for a specified program. If one of these commands is called with a program for which LinkProgram succeeded, the information recorded when the program was linked is returned. If one of these commands is called with a program for which LinkProgram failed, no error is generated unless otherwise noted. Implementations may return information on variables and interface blocks that would have been active had the program been linked successfully. In cases where the link failed because the program required too many resources, these commands may help applications determine why limits were exceeded. However, the information returned in this case is implementation-dependent and may be incomplete. If one of these commands is called with a program for which LinkProgram had never been called, no error is generated unless otherwise noted, and the program object is considered to have no active variables or interface blocks. OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

75

Each program object has an information log that is overwritten as a result of a link operation. This information log can be queried with GetProgramInfoLog to obtain more information about the link operation or the validation information (see section 7.12). If a program has been linked successfully by LinkProgram or loaded by ProgramBinary (see section 7.5), it can be made part of the current rendering state for all shader stages with the command void UseProgram( uint program ); If program is non-zero, this command will make program the current program object. This will install executable code as part of the current rendering state for each shader stage present when the program was last linked successfully. If UseProgram is called with program set to zero, then there is no current program object. The executable code for an individual shader stage is taken from the current program for that stage. If there is a current program object established by UseProgram, that program is considered current for all stages. Otherwise, if there is a bound program pipeline object (see section 7.4), the program bound to the appropriate stage of the pipeline object is considered current. If there is no current program object or bound program pipeline object, no program is current for any stage. The current program for a stage is considered active if it contains executable code for that stage; otherwise, no program is considered active for that stage. If there is no active program for the vertex or fragment shader stages, the results of vertex and fragment shader execution will respectively be undefined. However, this is not an error. If there is no active program for the tessellation control, tessellation evaluation, or geometry shader stages, those stages are ignored. If there is no active program for the compute shader stage, compute dispatches will generate an error. The active program for the compute shader stage has no effect on the processing of vertices, geometric primitives, and fragments, and the active program for all other shader stages has no effect on compute dispatches1 . Errors An INVALID_VALUE error is generated if program is neither zero nor the name of either a program or shader object. An INVALID_OPERATION error is generated if program is not zero and is the name of a shader object. An INVALID_OPERATION error is generated if program has not been 1

It is possible for a single program pipeline object to contain active programs for all shader stages, even though not all of them will be used while executing drawing commands or compute dispatch.

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

76

linked successfully. The current rendering state is not modified. While a program object is in use, applications are free to modify attached shader objects, compile attached shader objects, attach additional shader objects, and detach shader objects. These operations do not affect the link status or executable code of the program object. If LinkProgram or ProgramBinary successfully re-links a program object that is active for any shader stage, then the newly generated executable code will be installed as part of the current rendering state for all shader stages where the program is active. Additionally, the newly generated executable code is made part of the state of any program pipeline for all stages where the program is attached. If a program object that is active for any shader stage is re-linked unsuccessfully, the link status will be set to FALSE, but any existing executables and associated state will remain part of the current rendering state until a subsequent call to UseProgram, UseProgramStages, or BindProgramPipeline removes them from use. If such a program is attached to any program pipeline object, the existing executables and associated state will remain part of the program pipeline object until a subsequent call to UseProgramStages removes them from use. A program which has not been linked successfully may not be made part of the current rendering state by UseProgram or added to program pipeline objects by UseProgramStages until it is re-linked successfully. If such a program was attached to a program pipeline at the time of a failed link, its existing executable may still be made part of the current rendering state indirectly by BindProgramPipeline. To set a program object parameter, call void ProgramParameteri( uint program, enum pname, int value ); pname identifies which parameter to set for program. value holds the value being set. If pname is PROGRAM_SEPARABLE, value must be TRUE or FALSE, and indicates whether program can be bound for individual pipeline stages using UseProgramStages after it is next linked. If pname is PROGRAM_BINARY_RETRIEVABLE_HINT, value must be TRUE or FALSE, and indicates whether a program binary is likely to be retrieved later, as described for ProgramBinary in section 7.5. State set with this command does not take effect until after the next time LinkProgram or ProgramBinary is called successfully.

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

77

Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if pname is not PROGRAM_SEPARABLE or PROGRAM_BINARY_RETRIEVABLE_HINT. An INVALID_VALUE error is generated if value is not TRUE or FALSE. Program objects can be deleted with the command void DeleteProgram( uint program ); If program is not current for any GL context, is not the active program for any program pipeline object, and is not the current program for any stage of any program pipeline object, it is deleted immediately. Otherwise, program is flagged for deletion and will be deleted after all of these conditions become true. When a program object is deleted, all shader objects attached to it are detached. DeleteProgram will silently ignore the value zero. Errors An INVALID_VALUE error is generated if program is neither zero nor the name of either a program or shader object. An INVALID_OPERATION error is generated if program is not zero and is the name of a shader object. The command boolean IsProgram( uint program ); returns TRUE if program is the name of a program object. If program is zero, or a non-zero value that is not the name of a program object, IsProgram returns FALSE. No error is generated if program is not a valid program object name. The command uint CreateShaderProgramv( enum type, sizei count, const char * const *strings ); creates a stand-alone program from an array of null-terminated source code strings for a single shader type. CreateShaderProgramv is equivalent (assuming no errors are generated) to: OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

78

const uint shader = CreateShader(type); if (shader) { ShaderSource(shader, count, strings, NULL); CompileShader(shader); const uint program = CreateProgram(); if (program) { int compiled = FALSE; GetShaderiv(shader, COMPILE_STATUS, &compiled); ProgramParameteri(program, PROGRAM_SEPARABLE, TRUE); if (compiled) { AttachShader(program, shader); LinkProgram(program); DetachShader(program, shader); } append-shader-info-log-to-program-info-log } DeleteShader(shader); return program; } else { return 0; } Because no shader is returned by CreateShaderProgramv and the shader that is created is deleted in the course of the command sequence, the info log of the shader object is copied to the program so the shader’s failed info log for the failed compilation is accessible to the application. If an error is generated, zero is returned. Errors An INVALID_ENUM error is generated if type is not one of the values in table 7.1. An INVALID_VALUE error is generated if count is negative. Other errors are generated if the supplied shader code fails to compile and link, as described for the commands in the pseudocode sequence above, but all such errors are generated without any side effects of executing those commands.

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

7.3.1

79

Program Interfaces

When a program object is made part of the current rendering state, its executable code may communicate with other GL pipeline stages or application code through a variety of interfaces. When a program is linked, the GL builds a list of active resources for each interface. Examples of active resources include variables and interface blocks used by shader code. Resources referenced in shader code are considered active unless the compiler and linker can conclusively determine that they have no observable effect on the results produced by the executable code of the program. For example, variables might be considered inactive if they are declared but not used in executable code, used only in a clause of an if statement that would never be executed, used only in functions that are never called, or used only in computations of temporary variables having no effect on any shader output. In cases where the compiler or linker cannot make a conclusive determination, any resource referenced by shader code will be considered active. The set of active resources for any interface is implementation-dependent because it depends on various analysis and optimizations performed by the compiler and linker. If a program is linked successfully, the GL will generate lists of active resources based on the executable code produced by the link. If a program is not linked successfully, the link may have failed for a number of reasons, including cases where the program required more resources than supported by the implementation. Implementations are permitted, but not required, to record lists of resources that would have been considered active had the program linked successfully. If an implementation does not record information for any given interface, the corresponding list of active resources is considered empty. If a program has never been linked, all lists of active resources are considered empty. The GL provides a number of commands to query properties of the interfaces of a program object. Each such command accepts a programInterface token, identifying a specific interface. The supported values for programInterface are as follows: • UNIFORM corresponds to the set of active uniform variables (see section 7.6) used by program. • UNIFORM_BLOCK corresponds to the set of active uniform blocks (see section 7.6) used by program. • ATOMIC_COUNTER_BUFFER corresponds to the set of active atomic counter buffer binding points (see section 7.6) used by program. • PROGRAM_INPUT corresponds to the set of active input variables used by the first shader stage of program. If program includes multiple shader stages,

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

80

input variables from any shader stage other than the first will not be enumerated. • PROGRAM_OUTPUT corresponds to the set of active output variables (see section 11.1.2.1) used by the last shader stage of program. If program includes multiple shader stages, output variables from any shader stage other than the last will not be enumerated. • TRANSFORM_FEEDBACK_VARYING corresponds to the set of output variables in the last non-fragment stage of program that would be captured when transform feedback is active (see section 11.1.2.1). The resources enumerated by this query are listed as specified by the most recent call to TransformFeedbackVaryings before the last call to LinkProgram. When the resource names an output array variable either a single element of the array or the whole array is captured. If the variable name is specified with an array index syntax "name[x]", name is the name of the array resource and x is the constant-integer index of the element captured. If the resource name is an array and has no array index and square bracket, then the whole array is captured. • BUFFER_VARIABLE corresponds to the set of active buffer variables used by program (see section 7.8). • SHADER_STORAGE_BLOCK corresponds to the set of active shader storage blocks used by program (see section 7.8) 7.3.1.1

Naming Active Resources

When building a list of active variable or interface blocks, resources with aggregate types (such as arrays or structures) may produce multiple entries in the active resource list for the corresponding interface. Additionally, each active variable or interface block in the list is assigned an associated name string that can be used by applications to refer to the resource. For interfaces enumerating active variables, resource list entries for variables declared outside interface blocks are generated as follows: • For an active variable declared as a single instance of a basic type, a single entry will be generated, using the variable name from the shader source. • For an active variable declared as an array of basic types (e.g. not an array of structures or an array of arrays), a single entry will be generated, with its name string formed by concatenating the name of the array and the string "[0]". OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

81

• For an active variable declared as a structure, a separate entry will be generated for each active structure member. The name of each entry is formed by concatenating the name of the structure, the "." character, and the name of the structure member. If a structure member to enumerate is itself a structure or array, these enumeration rules are applied recursively. • For an active variable declared as an array of an aggregate data type (structures or arrays), a separate entry will be generated for each active array element, unless noted immediately below. The name of each entry is formed by concatenating the name of the array, the "[" character, an integer identifying the element number, and the "]" character. These enumeration rules are applied recursively, treating each enumerated array element as a separate active variable. For interfaces enumerating active variables, resource list entries for active members of interface blocks are generated as follows: • For active members of an interface block with no instance name, resource list entries will be generated by treating the block member as though it were declared as a variable outside an interface block. The name strings for these entries will not include the block name. • For active members of an interface block with an instance name, resource list entries will be generated by applying the rules for variables declared outside an interface block. The name string for each such entry is formed by concatenating the name of the interface block (not the instance name), the "." character, and the name string that would be generated for an equivalent variable declared outside an interface block. • For active members of an array of interface blocks, entries will be generated using the same rules as for members of a single interface block with an instance name. There will not be separate entries for each instance of the interface block, and the name strings will not contain any text like "[0]" to indicate that the member belongs to an array of interface blocks. • For active shader storage block members that are declared as arrays of an aggregate type, entries will not be generated for each array element. Such block members are referred to as top-level arrays and will generate entries only for the first array element. For interfaces enumerating active interface blocks, the entries of active resource lists are generated as follows: OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

82

• For an active interface block not declared as an array of block instances, a single entry will be generated, using the block name from the shader source. • For an active interface block declared as an array of arrays, a separate entry will be generated for each active instance. The name of each instance is formed by concatenating the block name, the "[" character, an integer identifying the instance number, and the "]" character. These enumeration rules are applied recursively, treating each enumerated array element as a separate active interface block. When an integer array element or block instance number is part of the name string, it will be specified in decimal form without a "+" or "-" sign or any extra leading zeroes. Additionally, the name string will not include white space anywhere in the string. The order of the active resource list is implementation-dependent for all interfaces except for TRANSFORM_FEEDBACK_VARYING. For TRANSFORM_FEEDBACK_VARYING, the active resource list will use the variable order specified in the most recent call to TransformFeedbackVaryings before the last call to LinkProgram. For the ATOMIC_COUNTER_BUFFER interface, the list of active buffer binding points is built by identifying each unique binding point associated with one or more active atomic counter uniform variables. Active atomic counter buffers do not have an associated name string. For the UNIFORM, PROGRAM_INPUT, PROGRAM_OUTPUT, and TRANSFORM_FEEDBACK_VARYING interfaces, the active resource list will include all active variables for the interface, including any active built-in variables. When a program is linked successfully, active variables in the UNIFORM, PROGRAM_INPUT, or PROGRAM_OUTPUT interfaces are assigned one or more signed integer locations. These locations can be used by commands to assign values to uniforms, to identify generic vertex attributes associated with vertex shader inputs, or to identify fragment color output numbers associated with fragment shader outputs. For such variables declared as arrays, separate locations will be assigned to each active array element and are not required to be sequential. The location for "a[1]" may or may not be equal to the location for "a[0]" +1. Furthermore, since unused elements at the end of uniform arrays may be trimmed, the location of the i + 1’th array element may not be valid even if the location of the i’th element is valid. As a direct consequence, the value of the location of "a[0]" +1 may refer to a different uniform entirely. Applications that wish to set individual array elements should query the locations of each element separately. Not all active variables are assigned valid locations; the following variables will have an effective location of -1: OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

83

• uniforms declared as atomic counters • members of a uniform block • built-in inputs, outputs, and uniforms (starting with gl_) • inputs (except for vertex shader inputs) not declared with a location layout qualifier • outputs (except for fragment shader outputs) not declared with a location layout qualifier If a program has not been linked successfully, no locations will be assigned. The command void GetProgramInterfaceiv( uint program, enum programInterface, enum pname, int *params ); queries a property of the interface programInterface in program program, returning its value in params. The property to return is specified by pname. If pname is ACTIVE_RESOURCES, the value returned is the number of resources in the active resource list for programInterface. If the list of active resources for programInterface is empty, zero is returned. If pname is MAX_NAME_LENGTH, the value returned is the length of the longest active name string for an active resource in programInterface. This length includes an extra character for the null terminator. If the list of active resources for programInterface is empty, zero is returned. If pname is MAX_NUM_ACTIVE_VARIABLES, the value returned is the number of active variables belonging to the interface block or atomic counter buffer resource in programInterface with the most active variables. If the list of active resources for programInterface is empty, zero is returned. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if programInterface is not one of the interfaces described in the introduction to section 7.3.1. An INVALID_ENUM error is generated if pname is not ACTIVE_RESOURCES, MAX_NAME_LENGTH, or MAX_NUM_ACTIVE_VARIABLES.

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

84

An INVALID_OPERATION error is generated if pname is MAX_NAME_LENGTH and programInterface is ATOMIC_COUNTER_BUFFER, since active atomic counter resources are not assigned name strings. An INVALID_OPERATION error is generated if pname is MAX_NUM_ACTIVE_VARIABLES and programInterface is not ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BLOCK, or UNIFORM_BLOCK. Each entry in the active resource list for an interface is assigned a unique unsigned integer index in the range zero to N − 1, where N is the number of entries in the active resource list. The command uint GetProgramResourceIndex( uint program, enum programInterface, const char *name ); returns the unsigned integer index assigned to a resource named name in the interface type programInterface of program object program. If name exactly matches the name string of one of the active resources for programInterface, the index of the matched resource is returned. • For TRANSFORM_FEEDBACK_VARYING resources, name must match one of the variables to be captured as specified by a previous call to TransformFeedbackVaryings. Otherwise, • For all other resource types, if name would exactly match the name string of an active resource if "[0]" were appended to name, the index of the matched resource is returned. Otherwise, name is considered not to be the name of an active resource, and INVALID_INDEX is returned. Note that if an interface enumerates a single active resource list entry for an array variable (e.g., "a[0]"), a name identifying any array element other than the first (e.g., "a[1]") is not considered to match. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. If name does not match a resource as described above, the value INVALID_INDEX is returned, but no GL error is generated. An INVALID_ENUM error is generated if programInterface is not one of the interfaces described in the introduction to section 7.3.1.

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

85

An INVALID_ENUM error is generated if programInterface is ATOMIC_COUNTER_BUFFER, since active atomic counter resources are not assigned name strings. The command void GetProgramResourceName( uint program, enum programInterface, uint index, sizei bufSize, sizei *length, char *name ); returns the name string assigned to the single active resource with an index of index in the interface programInterface of program object program. The name string assigned to the active resource identified by index is returned as a null-terminated string in name. The actual number of characters written into name, excluding the null terminator, is returned in length. If length is NULL, no length is returned. The maximum number of characters that may be written into name, including the null terminator, is specified by bufSize. If the length of the name string (including the null terminator) is greater than bufSize, the first bufSize − 1 characters of the name string will be written to name, followed by a null terminator. If bufSize is zero, no error is generated but no characters will be written to name. The length of the longest name string for programInterface, including a null terminator, can be queried by calling GetProgramInterfaceiv with a pname of MAX_NAME_LENGTH. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if programInterface is not one of the interfaces described in the introduction to section 7.3.1. An INVALID_ENUM error is generated if programInterface is ATOMIC_COUNTER_BUFFER, since active atomic counter resources are not assigned name strings. An INVALID_VALUE error is generated if index is greater than or equal to the number of entries in the active resource list for programInterface. An INVALID_VALUE error is generated if bufSize is negative. The command

OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

86

void GetProgramResourceiv( uint program, enum programInterface, uint index, sizei propCount, const enum *props, sizei count, sizei *length, int *params ); returns values for multiple properties of a single active resource with an index of index in the interface programInterface of program object program. Values for propCount properties specified by the array props are returned. The values associated with the properties of the active resource are written to consecutive entries in params, in increasing order according to position in props. If no error is generated, only the first count integer values will be written to params; any extra values will not be written. If length is not NULL, the actual number of values written to params will be written to length. Supported Interfaces

Property ACTIVE_VARIABLES, BUFFER_BINDING, NUM_ACTIVE_VARIABLES ARRAY_SIZE

ARRAY_STRIDE, BLOCK_INDEX, IS_ROW_MAJOR, MATRIX_STRIDE ATOMIC_COUNTER_BUFFER_INDEX BUFFER_DATA_SIZE IS_PER_PATCH LOCATION NAME_LENGTH OFFSET REFERENCED_BY_VERTEX_SHADER, REFERENCED_BY_TESS_CONTROL_SHADER, REFERENCED_BY_TESS_EVALUATION_SHADER, REFERENCED_BY_GEOMETRY_SHADER, REFERENCED_BY_FRAGMENT_SHADER, REFERENCED_BY_COMPUTE_SHADER

ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BLOCK, UNIFORM_BLOCK BUFFER_VARIABLE, PROGRAM_INPUT, PROGRAM_OUTPUT, TRANSFORM_FEEDBACK_VARYING, UNIFORM BUFFER_VARIABLE, UNIFORM UNIFORM ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BLOCK, UNIFORM_BLOCK PROGRAM_INPUT, PROGRAM_OUTPUT PROGRAM_INPUT, PROGRAM_OUTPUT, UNIFORM all but ATOMIC_COUNTER_BUFFER BUFFER_VARIABLE, UNIFORM ATOMIC_COUNTER_BUFFER, BUFFER_VARIABLE, PROGRAM_INPUT, PROGRAM_OUTPUT, SHADER_STORAGE_BLOCK, UNIFORM, UNIFORM_BLOCK

GetProgramResourceiv properties continued on next page OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

87

GetProgramResourceiv properties continued from previous page Property Supported Interfaces TOP_LEVEL_ARRAY_SIZE, TOP_- BUFFER_VARIABLE LEVEL_ARRAY_STRIDE TYPE

BUFFER_VARIABLE, PROGRAM_INPUT, PROGRAM_OUTPUT, TRANSFORM_FEEDBACK_VARYING, UNIFORM

Table 7.2: GetProgramResourceiv properties and supported interfaces

For the property ACTIVE_VARIABLES, an array of active variable indices associated with an atomic counter buffer, active uniform block, or shader storage block is written to params. The number of values written to params for an active resource is given by the value of the property NUM_ACTIVE_VARIABLES for the resource. For the property ARRAY_SIZE, a single integer identifying the number of active array elements of an active variable is written to params. The array size returned is in units of the type associated with the property TYPE. For active variables not corresponding to an array of basic types, the value one is written to params. If the variable is an array whose size is not declared or determined when the program is linked, the value zero is written to params. For the property ARRAY_STRIDE, a single integer identifying the stride between array elements in an active variable is written to params. For active variables declared as an array of basic types, the value written is the difference, in basic machine units, between the offsets of consecutive elements in an array. For active variables not declared as an array of basic types, zero is written to params. For active variables not backed by a buffer object, -1 is written to params, regardless of the variable type. For the property ATOMIC_COUNTER_BUFFER_INDEX, a single integer identifying the index of the active atomic counter buffer containing an active variable is written to params. If the variable is not an atomic counter uniform, the value -1 is written to params. For the property BLOCK_INDEX, a single integer identifying the index of the active interface block containing an active variable is written to params. The index written for a member of an interface block declared as an array of block instances is the index of the first block of the array. If the variable is not a member of an interface block, the value -1 is written to params. OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

88

For the property BUFFER_BINDING, a single integer identifying the index of the buffer binding point associated with the active uniform block, atomic counter buffer, or shader storage block is written to params. For the property BUFFER_DATA_SIZE, a single integer identifying the implementation-dependent minimum total buffer object size is written to params. This value is the size, in basic machine units, required to hold all active variables associated with an active uniform block, atomic counter buffer, or shader storage block. If the final member of an active shader storage block is an array with no declared size, the minimum buffer size is computed assuming the array was declared as an array with one element. For the property IS_PER_PATCH, a single integer identifying whether the input or output is a per-patch attribute is written to params. If the active variable is a per-patch attribute (declared with the patch qualifier), the value one is written to params; otherwise, the value zero is written to params. For the property IS_ROW_MAJOR, a single integer identifying whether an active variable is a row-major matrix is written to params. For active variables backed by a buffer object, declared as a single matrix or array of matrices, and stored in rowmajor order, one is written to params. For all other active variables, zero is written to params. For the property LOCATION, a single integer identifying the assigned location for an active uniform, input, or output variable is written to params. For input, output, or uniform variables with locations specified by a layout qualifier, the specified location is used. For vertex shader input, fragment shader output, or uniform variables without a layout qualifier, the location assigned when a program is linked is written to params. For all other input and output variables, the value -1 is written to params. For atomic counter uniforms and uniforms in uniform blocks, the value -1 is written to params. For the property MATRIX_STRIDE, a single integer identifying the stride between columns of a column-major matrix or rows of a row-major matrix is written to params. For active variables declared a single matrix or array of matrices, the value written is the difference, in basic machine units, between the offsets of consecutive columns or rows in each matrix. For active variables not declared as a matrix or array of matrices, zero is written to params. For active variables not backed by a buffer object, -1 is written to params, regardless of the variable type. For the property NAME_LENGTH, a single integer identifying the length of the name string associated with an active variable or interface block is written to params. The name length includes a terminating null character. For the property NUM_ACTIVE_VARIABLES, a single integer identifying the number of active variables associated with an active uniform block, atomic counter buffer, or shader storage block is written to params. OpenGL ES 3.2 (May 5, 2022)

7.3. PROGRAM OBJECTS

89

For the property OFFSET, a single integer identifying the offset of an active variable is written to params. For variables in the BUFFER_VARIABLE and UNIFORM interfaces that are backed by a buffer object, the value written is the offset of that variable relative to the base of the buffer range holding its value. For active variables not backed by a buffer object, an offset of -1 is written to params. For the properties REFERENCED_BY_VERTEX_SHADER, REFERENCED_BY_TESS_CONTROL_SHADER, REFERENCED_BY_TESS_EVALUATION_SHADER, REFERENCED_BY_GEOMETRY_SHADER, REFERENCED_BY_FRAGMENT_SHADER, and REFERENCED_BY_COMPUTE_SHADER, a single integer is written to params, identifying whether the active resource is referenced by the vertex, tessellation control, tessellation evaluation, geometry, fragment, or compute shaders, respectively, in the program object. The value one is written to params if an active variable is referenced by the corresponding shader, or if an active uniform block, shader storage block, or atomic counter buffer contains at least one variable referenced by the corresponding shader. Otherwise, the value zero is written to params. For the property TOP_LEVEL_ARRAY_SIZE, a single integer identifying the number of active array elements of the top-level shader storage block member containing to the active variable is written to params. If the top-level block member is not declared as an array of an aggregate type, the value one is written to params. If the top-level block member is an array of an aggregate type whose size is not declared or determined when the program is linked, the value zero is written to params. For the property TOP_LEVEL_ARRAY_STRIDE, a single integer identifying the stride between array elements of the top-level shader storage block member containing the active variable is written to params. For top-level block members declared as arrays of an aggregate type, the value written is the difference, in basic machine units, between the offsets of the active variable for consecutive elements in the top-level array. For top-level block members not declared as an array of an aggregate type, zero is written to params. For the property TYPE, a single integer identifying the type of an active variable is written to params. The integer returned is one of the values found in table 7.3.

Type Name Token FLOAT FLOAT_VEC2 FLOAT_VEC3 FLOAT_VEC4

Keyword

Attrib Xfb

Buffer

float vec2 vec3 vec4

   

   

(Continued on next page) OpenGL ES 3.2 (May 5, 2022)

   

7.3. PROGRAM OBJECTS

90

OpenGL ES Shading Language Type Tokens (continued) Type Name Token Keyword Attrib Xfb INT int   INT_VEC2 ivec2   INT_VEC3 ivec3   INT_VEC4 ivec4   UNSIGNED_INT uint   UNSIGNED_INT_VEC2 uvec2   UNSIGNED_INT_VEC3 uvec3   UNSIGNED_INT_VEC4 uvec4   BOOL BOOL_VEC2 BOOL_VEC3 BOOL_VEC4 FLOAT_MAT2 FLOAT_MAT3 FLOAT_MAT4 FLOAT_MAT2x3 FLOAT_MAT2x4 FLOAT_MAT3x2 FLOAT_MAT3x4 FLOAT_MAT4x2 FLOAT_MAT4x3 SAMPLER_2D SAMPLER_3D SAMPLER_CUBE SAMPLER_2D_SHADOW SAMPLER_2D_ARRAY SAMPLER_CUBE_MAP_ARRAY SAMPLER_2D_ARRAY_SHADOW SAMPLER_2D_MULTISAMPLE SAMPLER_2D_MULTISAMPLE_ARRAY SAMPLER_CUBE_SHADOW SAMPLER_CUBE_MAP_ARRAY_SHADOW SAMPLER_BUFFER

bool bvec2 bvec3 bvec4 mat2 mat3 mat4 mat2x3 mat2x4 mat3x2 mat3x4 mat4x2 mat4x3 sampler2D sampler3D samplerCube sampler2DShadow sampler2DArray samplerCubeArray sampler2DArrayShadow sampler2DMS sampler2DMSArray samplerCubeShadow samplerCubeArrayShadow samplerBuffer

(Continued on next page)

OpenGL ES 3.2 (May 5, 2022)

        

        

Buffer                     

7.3. PROGRAM OBJECTS

91

OpenGL ES Shading Language Type Tokens (continued) Type Name Token Keyword Attrib Xfb INT_SAMPLER_2D INT_SAMPLER_3D INT_SAMPLER_CUBE INT_SAMPLER_2D_ARRAY INT_SAMPLER_CUBE_MAP_ARRAY INT_SAMPLER_2D_MULTISAMPLE INT_SAMPLER_2D_MULTISAMPLE_ARRAY INT_SAMPLER_BUFFER UNSIGNED_INT_SAMPLER_2D UNSIGNED_INT_SAMPLER_3D UNSIGNED_INT_SAMPLER_CUBE UNSIGNED_INT_SAMPLER_2D_ARRAY UNSIGNED_INT_SAMPLER_CUBE_MAP_ARRAY UNSIGNED_INT_SAMPLER_2D_MULTISAMPLE UNSIGNED_INT_SAMPLER_2D_MULTISAMPLE_ARRAY UNSIGNED_INT_SAMPLER_BUFFER IMAGE_2D IMAGE_3D IMAGE_CUBE IMAGE_BUFFER IMAGE_2D_ARRAY IMAGE_CUBE_MAP_ARRAY INT_IMAGE_2D INT_IMAGE_3D INT_IMAGE_CUBE INT_IMAGE_BUFFER

isampler2D isampler3D isamplerCube isampler2DArray isamplerCubeArray isampler2DMS isampler2DMSArray isamplerBuffer usampler2D usampler3D usamplerCube usampler2DArray usamplerCubeArray usampler2DMS usampler2DMSArray usamplerBuffer image2D image3D imageCube imageBuffer image2DArray imageCubeArray iimage2D iimage3D iimageCube iimageBuffer

(Continued on next page)

OpenGL ES 3.2 (May 5, 2022)

Buffer

7.3. PROGRAM OBJECTS

92

OpenGL ES Shading Language Type Tokens (continued) Type Name Token Keyword Attrib Xfb INT_IMAGE_2D_ARRAY INT_IMAGE_CUBE_MAP_ARRAY UNSIGNED_INT_IMAGE_2D UNSIGNED_INT_IMAGE_3D UNSIGNED_INT_IMAGE_CUBE UNSIGNED_INT_IMAGE_BUFFER UNSIGNED_INT_IMAGE_2D_ARRAY UNSIGNED_INT_IMAGE_CUBE_MAP_ARRAY UNSIGNED_INT_ATOMIC_COUNTER

iimage2DArray iimageCubeArray uimage2D uimage3D uimageCube uimageBuffer uimage2DArray uimageCubeArray atomic_uint

Table 7.3: OpenGL ES Shading Language type tokens, and corresponding shading language keywords declaring each such type. Types whose “Attrib” column is marked may be declared as vertex attributes (see section 11.1.1). Types whose “Xfb” column is marked may be the types of variables returned by transform feedback (see section 11.1.2.1). Types whose “Buffer” column is marked may be declared as buffer variables (see section 7.8).

Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if programInterface is not one of the interfaces described in the introduction to section 7.3.1. An INVALID_VALUE error is generated if propCount is less than or equal to zero, or if count is negative. An INVALID_ENUM error is generated if any value in props is not one of the properties described above.

OpenGL ES 3.2 (May 5, 2022)

Buffer

7.3. PROGRAM OBJECTS

93

An INVALID_OPERATION error is generated if any value in props is not allowed for programInterface. The set of allowed programInterface values for each property can be found in table 7.2. The command int GetProgramResourceLocation( uint program, enum programInterface, const char *name ); returns the location assigned to the variable named name in interface programInterface of program object program. programInterface must be one of UNIFORM, PROGRAM_INPUT, or PROGRAM_OUTPUT. The value -1 will be returned if an error occurs, if name does not identify an active variable on programInterface, or if name identifies an active variable that does not have a valid location assigned, as described above. The locations returned by these commands are the same locations returned when querying the LOCATION resource properties. A string provided to GetProgramResourceLocation is considered to match an active variable if • the string exactly matches the name of the active variable; • if the string identifies the base name of an active array, where the string would exactly match the name of the variable if the suffix "[0]" were appended to the string; or • if the string identifies an active element of the array, where the string ends with the concatenation of the "[" character, an integer (with no "+" sign, extra leading zeroes, or whitespace) identifying an array element, and the "]" character, the integer is less than the number of active elements of the array variable, and where the string would exactly match the enumerated name of the array if the decimal integer were replaced with zero. Any other string is considered not to identify an active variable. If the string specifies an element of an array variable, GetProgramResourceLocation returns the location assigned to that element. If it specifies the base name of an array, it identifies the resources associated with the first element of the array. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a

OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS

94

shader object. An INVALID_OPERATION error is generated if program has not been linked successfully. An INVALID_ENUM error is generated if programInterface is not one of the interfaces named above.

7.4

Program Pipeline Objects

Instead of packaging all shader stages into a single program object, shader types might be contained in multiple program objects each consisting of part of the complete pipeline. A program object may even contain only a single shader stage. This facilitates greater flexibility when combining different shaders in various ways without requiring a program object for each combination. A program pipeline object contains bindings for each shader type associating that shader type with a program object. The command void GenProgramPipelines( sizei n, uint *pipelines ); returns n previously unused program pipeline object names in pipelines. These names are marked as used, for the purposes of GenProgramPipelines only, but they acquire state only when they are first bound. Errors An INVALID_VALUE error is generated if n is negative. Program pipeline objects are deleted by calling void DeleteProgramPipelines( sizei n, const uint *pipelines ); pipelines contains n names of program pipeline objects to be deleted. Once a program pipeline object is deleted, it has no contents and its name becomes unused. If an object that is currently bound is deleted, the binding for that object reverts to zero and no program pipeline object becomes current. Unused names in pipelines that have been marked as used for the purposes of GenProgramPipelines are marked as unused again. Unused names in pipelines are silently ignored, as is the value zero.

OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS

95

Errors An INVALID_VALUE error is generated if n is negative. The command boolean IsProgramPipeline( uint pipeline ); returns TRUE if pipeline is the name of a program pipeline object. If pipeline is zero, or a non-zero value that is not the name of a program pipeline object, IsProgramPipeline returns FALSE. No error is generated if pipeline is not a valid program pipeline object name. A program pipeline object is created by binding a name returned by GenProgramPipelines with the command void BindProgramPipeline( uint pipeline ); pipeline is the program pipeline object name. The resulting program pipeline object is a new state vector, comprising all the state and with the same initial values listed in table 21.20. BindProgramPipeline may also be used to bind an existing program pipeline object. If the bind is successful, no change is made to the state of the bound program pipeline object, and any previous binding is broken. If BindProgramPipeline is called with pipeline set to zero, then there is no current program pipeline object. If no current program object has been established by UseProgram, the program objects used for each shader stage and for uniform updates are taken from the bound program pipeline object, if any. If there is a current program object established by UseProgram, the bound program pipeline object has no effect on rendering or uniform updates. When a bound program pipeline object is used for rendering, individual shader executables are taken from its program objects as described in the discussion of UseProgram in section 7.3). Errors An INVALID_OPERATION error is generated if pipeline is not zero or a name returned from a previous call to GenProgramPipelines, or if such a name has since been deleted with DeleteProgramPipelines. The executables in a program object associated with one or more shader stages can be made part of the program pipeline state for those shader stages with the command OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS

96

void UseProgramStages( uint pipeline, bitfield stages, uint program ); where pipeline is the program pipeline object to be updated, stages is the bitwise OR of accepted constants representing shader stages, and program is the program object from which the executables are taken. The bits set in stages indicate the program stages for which the program object named by program becomes current. These stages may include compute, vertex, tessellation control, tessellation evaluation, geometry, or fragment, indicated respectively by COMPUTE_SHADER_BIT, VERTEX_SHADER_BIT, TESS_CONTROL_SHADER_BIT, TESS_EVALUATION_SHADER_BIT, GEOMETRY_SHADER_BIT, or FRAGMENT_SHADER_BIT. The constant ALL_SHADER_BITS indicates program is to be made current for all shader stages. If program refers to a program object with a valid shader attached for an indicated shader stage, this call installs the executable code for that stage in the indicated program pipeline object state. If UseProgramStages is called with program set to zero or with a program object that contains no executable code for any stage in stages, it is as if the pipeline object has no programmable stage configured for that stage. If pipeline is a name that has been generated (without subsequent deletion) by GenProgramPipelines, but refers to a program pipeline object that has not been previously bound, the GL first creates a new state vector in the same manner as when BindProgramPipeline creates a new program pipeline object. Errors An INVALID_VALUE error is generated if stages is not the special value ALL_SHADER_BITS, and has any bits set other than COMPUTE_SHADER_BIT, VERTEX_SHADER_BIT, TESS_CONTROL_SHADER_BIT, TESS_EVALUATION_SHADER_BIT, GEOMETRY_SHADER_BIT, and FRAGMENT_SHADER_BIT. An INVALID_VALUE error is generated if program is not zero and is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated if program is not zero and was linked without the PROGRAM_SEPARABLE parameter set, or has not been linked successfully. The corresponding shader stages in pipeline are not modified.

OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS

97

An INVALID_OPERATION error is generated if pipeline is not a name returned from a previous call to GenProgramPipelines or if such a name has since been deleted by DeleteProgramPipelines. The command void ActiveShaderProgram( uint pipeline, uint program ); sets the linked program named by program to be the active program (see section 7.6.1) used for uniform updates for the program pipeline object pipeline. If program is zero, then it is as if there is no active program for pipeline. If pipeline is a name that has been generated (without subsequent deletion) by GenProgramPipelines, but refers to a program pipeline object that has not been previously bound, the GL first creates a new state vector in the same manner as when BindProgramPipeline creates a new program pipeline object. Errors An INVALID_OPERATION error is generated if pipeline is not a name returned from a previous call to GenProgramPipelines or if such a name has since been deleted by DeleteProgramPipelines. An INVALID_VALUE error is generated if program is not zero and is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated if program is not zero and has not been linked successfully. The active program is not modified.

7.4.1

Shader Interface Matching

When multiple shader stages are active, the outputs of one stage form an interface with the inputs of the next stage. At each such interface, shader inputs are matched up against outputs from the previous stage: • An output block is considered to match an input block in the subsequent shader if the two blocks have the same block name, and the members of the block match exactly in name, type, qualification, and declaration order. – For the purposes of shader interface matching, the gl_PointSize member of the intrinsically declared gl_PerVertex shader interface block is ignored. OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS

98

– Output blocks that do not match in name, but have a location and match in every other way listed above may be considered to match by some implementations, but not all - so this behaviour should not be relied upon. • An output variable is considered to match an input variable in the subsequent shader if: – the two variables match in name, type, and qualification, and neither has a location qualifier; or – the two variables are declared with the same location qualifier and match in type and qualification. Variables or block members declared as structures are considered to match in type if and only if structure members match in name, type, qualification, and declaration order. Variables or block members declared as arrays are considered to match in type only if both declarations specify the same element type and array size. The rules for determining if variables or block members match in qualification are found in the OpenGL ES Shading Language Specification. Tessellation control shader per-vertex output variables and blocks and tessellation control, tessellation evaluation, and geometry shader per-vertex input variables and blocks are required to be declared as arrays, with each element representing input or output values for a single vertex of a multi-vertex primitive. For the purposes of interface matching, such variables and blocks are treated as though they were not declared as arrays. For program objects containing multiple shaders, LinkProgram will check for mismatches on interfaces between shader stages in the program being linked and generate a link error if a mismatch is detected. A link error is generated if any statically referenced input variable or block does not have a matching output. With separable program objects, interfaces between shader stages may involve the outputs from one program object and the inputs from a second program object. For such interfaces, it is not possible to detect mismatches at link time, because the programs are linked separately. When each such program is linked, all inputs or outputs interfacing with another program stage are treated as active. The linker will generate an executable that assumes the presence of a compatible program on the other side of the interface. If a mismatch between programs occurs, using the programs together in a program pipeline will result in a validation failure (see section 11.1.3.11). At an interface between program objects, the set of inputs and outputs are considered to match exactly if and only if: OpenGL ES 3.2 (May 5, 2022)

7.4. PROGRAM PIPELINE OBJECTS • Every declared input block or variable has a matching output, as described above. However, the intrinsically declared gl_PerVertex shader interface block must be redeclared, and all members of the redeclared gl_PerVertex shader interface block, including the gl_PointSize member if present in the redeclaration, must match exactly in name, type, qualification and declaration order. • There are no output blocks or user-defined output variables declared without a matching input block or variable declaration. • All matched input and output variables (in a block or otherwise) have identical precision qualification. When the set of inputs and outputs on an interface between programs matches exactly, all inputs are well-defined except when the corresponding outputs were not written in the previous shader. However, any mismatch between inputs and outputs will result in a validation failure. As described above, an exact interface match requires matching built-in input and output blocks. At an interface between two non-fragment shader stages, the gl_PerVertex input and output blocks are considered to match if and only if the block members match exactly in name, type, qualification, and declaration order. At an interface involving the fragment shader stage, the presence or absence of any built-in output does not affect interface matching.

7.4.2

Program Pipeline Object State

The state required to support program pipeline objects consists of a single binding name of the current program pipeline object. This binding is initially zero indicating no program pipeline object is bound. The state of each program pipeline object consists of: • Unsigned integers holding the names of the active program and each of the current vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute stage programs. Each integer is initially zero. • A boolean holding the status of the last validation attempt, initially false. • An array of type char containing the information log (see section 7.12), initially empty. • An integer holding the length of the information log.

OpenGL ES 3.2 (May 5, 2022)

99

7.5. PROGRAM BINARIES

7.5

100

Program Binaries

The command void GetProgramBinary( uint program, sizei bufSize, sizei *length, enum *binaryFormat, void *binary ); returns a binary representation of the program object’s compiled and linked executable source, henceforth referred to as its program binary. The maximum number of bytes that may be written into binary is specified by bufSize. The actual number of bytes written into binary is returned in length and its format is returned in binaryFormat. If length is NULL, then no length is returned. The number of bytes in the program binary can be queried by calling GetProgramiv with pname PROGRAM_BINARY_LENGTH. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated if the value of NUM_PROGRAM_BINARY_FORMATS is zero. An INVALID_OPERATION error is generated if program has not been linked successfully. In this case its program binary length is zero. An INVALID_VALUE error is generated if bufSize is negative. An INVALID_OPERATION error is generated if bufSize is less than the number of bytes in the program binary. The command void ProgramBinary( uint program, enum binaryFormat, const void *binary, sizei length ); loads a program object with a program binary previously returned from GetProgramBinary. This is useful to avoid online compilation, while still using OpenGL ES Shading Language source shaders as a portable initial format. binaryFormat and binary must be those returned by a previous call to GetProgramBinary, and length must be the length of the program binary as returned by GetProgramBinary or GetProgramiv with pname PROGRAM_BINARY_LENGTH. Loading the program binary will fail, setting the LINK_STATUS of program to FALSE, if these conditions are not met. OpenGL ES 3.2 (May 5, 2022)

7.5. PROGRAM BINARIES

101

Loading a program binary may also fail if the implementation determines that there has been a change in hardware or software configuration from when the program binary was produced such as having been compiled with an incompatible or outdated version of the compiler. In this case the application should fall back to providing the original OpenGL ES Shading Language source shaders, and perhaps again retrieve the program binary for future use. A program object’s program binary is replaced by calls to LinkProgram or ProgramBinary. Where linking success or failure is concerned, ProgramBinary can be considered to perform an implicit linking operation. LinkProgram and ProgramBinary both set the program object’s LINK_STATUS to TRUE or FALSE, as queried with GetProgramiv, to reflect success or failure and update the information log, queried with GetProgramInfoLog, to provide details about warnings or errors. A successful call to ProgramBinary will reset all uniform variables in the default uniform block, all uniform block buffer bindings, and all shader storage block buffer bindings to their initial values. The initial value is either the value of the variable’s initializer as specified in the original shader source, or zero if no initializer was present. Additionally, values of the following state that were in effect when the program was linked before saving are restored when ProgramBinary is called successfully: • Program parameter PROGRAM_SEPARABLE, • all vertex shader input and fragment shader output assignments, and • atomic counter binding, offset and stride assignments. If ProgramBinary fails to load a binary, no error is generated, but any information about a previous link or load of that program object is lost. Thus, a failed load does not restore the old state of program. The failure does not alter other program state not affected by linking such as the attached shaders, and the vertex attribute bindings as set by BindAttribLocation. OpenGL ES defines no specific binary formats. Queries of values NUM_PROGRAM_BINARY_FORMATS and PROGRAM_BINARY_FORMATS return the number of program binary formats and the list of program binary format values supported by an implementation. The binaryFormat returned by GetProgramBinary must be present in this list. Any program binary retrieved using GetProgramBinary and submitted using ProgramBinary under the same configuration must be successful. Any programs loaded successfully by ProgramBinary must be run properly with any legal GL state vector. OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

102

If an implementation needs to recompile or otherwise modify program executables based on GL state outside the program, GetProgramBinary is required to save enough information to allow such recompilation. To indicate that a program binary is likely to be retrieved, ProgramParameteri should be called with pname set to PROGRAM_BINARY_RETRIEVABLE_HINT and value set to TRUE. This setting will not be in effect until the next time LinkProgram or ProgramBinary has been called successfully. Additionally, the application may defer GetProgramBinary calls until after using the program with all non-program state vectors that it is likely to encounter. Such deferral may allow implementations to save additional information in the program binary that would minimize recompilation in future uses of the program binary. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if binaryFormat is not a binary format present in the list of specific binary formats supported. An INVALID_VALUE error is generated if length is negative.

7.6

Uniform Variables

Shaders can declare named uniform variables, as described in the OpenGL ES Shading Language Specification. A uniform is considered an active uniform if the compiler and linker determine that the uniform will actually be accessed when the executable code is executed. In cases where the compiler and linker cannot make a conclusive determination, the uniform will be considered active. Sets of uniforms, except for atomic counters, images, and samplers, can be grouped into uniform blocks. Named uniform blocks, as described in the OpenGL ES Shading Language Specification, store uniform values in the data store of a buffer object corresponding to the uniform block. Such blocks are assigned a uniform block index. Uniforms that are declared outside of a named uniform block are part of the default uniform block. The default uniform block has no name or uniform block index. Uniforms in the default uniform block are program object-specific state. They retain their values once loaded, and their values are restored whenever a program object is used, as long as the program object has not been re-linked.

OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES Shader Stage Vertex (see section 11.1.2) Tessellation control (see section 11.2.1.1) Tessellation evaluation (see section 11.2.3.1) Geometry (see section 11.3.3) Fragment (see section 14.1) Compute (see section 17.1)

103 pname for querying default uniform block storage, in components MAX_VERTEX_UNIFORM_COMPONENTS MAX_TESS_CONTROL_UNIFORM_COMPONENTS MAX_TESS_EVALUATION_UNIFORM_COMPONENTS MAX_GEOMETRY_UNIFORM_COMPONENTS MAX_FRAGMENT_UNIFORM_COMPONENTS MAX_COMPUTE_UNIFORM_COMPONENTS

Table 7.4: Query targets for default uniform block storage, in components.

Like uniforms, uniform blocks can be active or inactive. Active uniform blocks are those that contain active uniforms after a program has been compiled and linked. Uniform blocks declared in an array are considered active if any member of the array would otherwise be considered active. All members of a named uniform block declared with a shared or std140 layout qualifier are considered active, even if they are not referenced in any shader in the program. Note that this means that the uniform block itself is also considered active, even if no member of the block is referenced. The implementation-dependent amount of storage available for uniform variables, except for atomic counters, in the default uniform block accessed by a shader for a particular shader stage can be queried by calling GetIntegerv with pname as specified in table 7.4 for that stage. The implementation-dependent constants MAX_VERTEX_UNIFORM_VECTORS and MAX_FRAGMENT_UNIFORM_VECTORS have values respectively equal to the values of MAX_VERTEX_UNIFORM_COMPONENTS and MAX_FRAGMENT_UNIFORM_COMPONENTS divided by four. The total amount of combined storage available for uniform variables in all uniform blocks accessed by a shader for a particular shader stage can be queried by calling GetIntegerv with pname as specified in table 7.5 for that stage. These values represent the numbers of individual floating-point, integer, or boolean values that can be held in uniform variable storage for a shader. For uniforms with boolean, integer, or floating-point components, • A scalar uniform will consume no more than 1 component • A vector uniform will consume no more than n components, where n is the vector component count

OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES Shader Stage Vertex Tessellation control Tessellation evaluation Geometry Fragment Compute

104

pname for querying combined uniform block storage, in components MAX_COMBINED_VERTEX_UNIFORM_COMPONENTS MAX_COMBINED_TESS_CONTROL_UNIFORM_COMPONENTS MAX_COMBINED_TESS_EVALUATION_UNIFORM_COMPONENTS MAX_COMBINED_GEOMETRY_UNIFORM_COMPONENTS MAX_COMBINED_FRAGMENT_UNIFORM_COMPONENTS MAX_COMBINED_COMPUTE_UNIFORM_COMPONENTS

Table 7.5: Query targets for combined uniform block storage, in components.

• A matrix uniform will consume no more than 4 × min(r, c) components, where r and c are the number of rows and columns in the matrix. Scalar, vector, and matrix uniforms with double-precision components will consume no more than twice the number of components of equivalent uniforms with floating-point components. Errors A link error is generated if an attempt is made to utilize more than the space available for uniform variables in a shader stage. When a program is linked successfully, all active uniforms, except for atomic counters, belonging to the program object’s default uniform block are initialized as defined by the version of the OpenGL ES Shading Language used to compile the program. A successful link will also generate a location for each active uniform in the default uniform block which doesn’t already have an explicit location defined in the shader. The generated locations will never take the location of a uniform with an explicit location defined in the shader, even if that uniform is determined to be inactive. The values of active uniforms in the default uniform block can be changed using this location and the appropriate Uniform* or ProgramUniform* command (see section 7.6.1). These generated locations are invalidated and new ones assigned after each successful re-link. The explicitly defined locations and the generated locations must be in the range of 0 to the value of MAX_UNIFORM_LOCATIONS minus one. Similarly, when a program is linked successfully, all active atomic counters are assigned bindings, offsets (and strides for arrays of atomic counters) according to layout rules described in section 7.6.2.2. Atomic counter uniform buffer objects OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

105

provide the storage for atomic counters, so the values of atomic counters may be changed by modifying the contents of the buffer object using the commands in sections 6.2, 6.3, and 6.5. Atomic counters are not assigned a location and may not be modified using the Uniform* commands. The bindings, offsets, and strides belonging to atomic counters of a program object are invalidated and new ones assigned after each successful re-link. Similarly, when a program is linked successfully, all active uniforms belonging to the program’s named uniform blocks are assigned offsets (and strides for array and matrix type uniforms) within the uniform block according to layout rules described below. Uniform buffer objects provide the storage for named uniform blocks, so the values of active uniforms in named uniform blocks may be changed by modifying the contents of the buffer object. Uniforms in a named uniform block are not assigned a location and may not be modified using the Uniform* commands. The offsets and strides of all active uniforms belonging to named uniform blocks of a program object are invalidated and new ones assigned after each successful re-link. To determine the set of active uniform variables used by a program, applications can query the properties and active resources of the UNIFORM interface of a program. Additionally, several dedicated commands are provided to query properties of active uniforms. The command int GetUniformLocation( uint program, const char *name ); is equivalent to GetProgramResourceLocation(program, UNIFORM, name); The command void GetUniformIndices( uint program, sizei uniformCount, const char * const *uniformNames, uint *uniformIndices ); is equivalent to for (int i = 0; i < uniformCount; i++) { uniformIndices[i] = GetProgramResourceIndex(program, UNIFORM, uniformNames[i]); } OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

106

The command void GetActiveUniform( uint program, uint index, sizei bufSize, sizei *length, int *size, enum *type, char *name ); is equivalent to const enum props[] = { ARRAY_SIZE, TYPE }; GetProgramResourceName(program, UNIFORM, index, bufSize, length, name); GetProgramResourceiv(program, UNIFORM, index, 1, &props[0], 1, NULL, size); GetProgramResourceiv(program, UNIFORM, index, 1, &props[1], 1, NULL, (int *)type); The command void GetActiveUniformsiv( uint program, sizei uniformCount, const uint *uniformIndices, enum pname, int *params ); is equivalent to GLenum prop; for (int i = 0; i < uniformCount; i++) { GetProgramResourceiv(program, UNIFORM, uniformIndices[i], 1, &prop, 1, NULL, &params[i]); } where the value of prop is taken from table 7.6, based on the value of pname. To determine the set of active uniform blocks used by a program, applications can query the properties and active resources of the UNIFORM_BLOCK interface. Additionally, several commands are provided to query properties of active uniform blocks. The command uint GetUniformBlockIndex( uint program, const char *uniformBlockName ); is equivalent to GetProgramResourceIndex(program, UNIFORM_BLOCK, uniformBlockName); OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

107

pname

prop

UNIFORM_TYPE UNIFORM_SIZE UNIFORM_NAME_LENGTH UNIFORM_BLOCK_INDEX UNIFORM_OFFSET UNIFORM_ARRAY_STRIDE UNIFORM_MATRIX_STRIDE UNIFORM_IS_ROW_MAJOR

TYPE ARRAY_SIZE NAME_LENGTH BLOCK_INDEX OFFSET ARRAY_STRIDE MATRIX_STRIDE IS_ROW_MAJOR

Table 7.6: GetProgramResourceiv properties used by GetActiveUniformsiv.

The command void GetActiveUniformBlockName( uint program, uint uniformBlockIndex, sizei bufSize, sizei length, char *uniformBlockName ); is equivalent to GetProgramResourceName(program, UNIFORM_BLOCK, uniformBlockIndex , bufSize, length, uniformBlockName); The command void GetActiveUniformBlockiv( uint program, uint uniformBlockIndex, enum pname, int *params ); is equivalent to GLenum prop; GetProgramResourceiv(program, UNIFORM_BLOCK, uniformBlockIndex , 1, &prop, maxSize, NULL, params); where the value of prop is taken from table 7.7, based on the value of pname, and maxSize is taken to specify a sufficiently large buffer to receive all values that would be written to params. To determine the set of active atomic counter buffer binding points used by a program, applications can query the properties and active resources of the ATOMIC_COUNTER_BUFFER interface of a program. OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

108

pname

prop

UNIFORM_BLOCK_BINDING UNIFORM_BLOCK_DATA_SIZE UNIFORM_BLOCK_NAME_LENGTH UNIFORM_BLOCK_ACTIVE_UNIFORMS UNIFORM_BLOCK_ACTIVE_UNIFORM_INDICES UNIFORM_BLOCK_REFERENCED_BY_VERTEX_SHADER

BUFFER_BINDING BUFFER_DATA_SIZE NAME_LENGTH NUM_ACTIVE_VARIABLES ACTIVE_VARIABLES

UNIFORM_BLOCK_REFERENCED_BY_FRAGMENT_SHADER

REFERENCED_BY_FRAGMENT_SHADER

Table 7.7: Blockiv.

7.6.1

REFERENCED_BY_VERTEX_SHADER

GetProgramResourceiv properties used by GetActiveUniform-

Loading Uniform Variables In The Default Uniform Block

To load values into the uniform variables, except for atomic counters, of the default uniform block of the active program object, use the commands void Uniform{1234}{if ui}( int location, T value ); void Uniform{1234}{if ui}v( int location, sizei count, const T *value ); void UniformMatrix{234}fv( int location, sizei count, boolean transpose, const float *value ); void UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}fv( int location, sizei count, boolean transpose, const float *value ); If a non-zero program object is bound by UseProgram, it is the active program object whose uniforms are updated by these commands. If no program object is bound using UseProgram, the active program object of the current program pipeline object set by ActiveShaderProgram is the active program object. If the current program pipeline object has no active program or there is no current program pipeline object, then there is no active program. The given values are loaded into the default uniform block uniform variable location identified by location and associated with a uniform variable. OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

109

The Uniform*f{v} commands will load count sets of one to four floating-point values into a uniform defined as a float, a floating-point vector, or an array of either of these types. The Uniform*i{v} commands will load count sets of one to four integer values into a uniform defined as a sampler, an integer, an integer vector, or an array of either of these types. Only the Uniform1i{v} commands can be used to load sampler values (see section 7.9). The Uniform*ui{v} commands will load count sets of one to four unsigned integer values into a uniform defined as a unsigned integer, an unsigned integer vector, or an array of either of these types. The UniformMatrix{234}fv commands will load count 2 × 2, 3 × 3, or 4 × 4 matrices (corresponding to 2, 3, or 4 in the command name) of floating-point values into a uniform defined as a matrix or an array of matrices. If transpose is FALSE, the matrix is specified in column major order, otherwise in row major order. The UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}fv commands will load count 2 × 3, 3 × 2, 2 × 4, 4 × 2, 3 × 4, or 4 × 3 matrices (corresponding to the numbers in the command name) of floating-point values into a uniform defined as a matrix or an array of matrices. The first number in the command name is the number of columns; the second is the number of rows. For example, UniformMatrix2x4fv is used to load a matrix consisting of two columns and four rows. If transpose is FALSE, the matrix is specified in column major order, otherwise in row major order. When loading values for a uniform declared as a boolean, a boolean vector, or an array of either of these types, any of the Uniform*i{v}, Uniform*ui{v}, and Uniform*f{v} commands can be used. Type conversion is done by the GL. Boolean values are set to FALSE if the corresponding input value is 0 or 0.0f, and set to TRUE otherwise. The Uniform* command used must match the size of the uniform, as declared in the shader. For example, to load a uniform declared as a bvec2, any of the Uniform2{if ui}* commands may be used. For all other uniform types loadable with Uniform* commands, the command used must match the size and type of the uniform, as declared in the shader, and no type conversions are done. For example, to load a uniform declared as a vec4, Uniform4f{v} must be used, and to load a uniform declared as a mat3, UniformMatrix3fv must be used. When loading N elements starting at an arbitrary position k in a uniform declared as an array, elements k through k + N − 1 in the array will be replaced with the new values. Values for any array element that exceeds the highest array element index used, as reported by GetActiveUniform, will be ignored by the GL. If the value of location is -1, the Uniform* commands will silently ignore the data passed in, and the current uniform values will not be changed. OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

Errors An INVALID_VALUE error is generated if count is negative. An INVALID_VALUE error is generated if Uniform1i{v} is used to set a sampler uniform to a value less than zero or greater than or equal to the value of MAX_COMBINED_TEXTURE_IMAGE_UNITS. An INVALID_OPERATION error is generated if any of the following conditions occur: • the size indicated in the name of the Uniform* command used does not match the size of the uniform declared in the shader, • the component type and count indicated in the name of the Uniform* command used does not match the type of the uniform declared in the shader, where a boolean uniform component type is considered to match any of the Uniform*i{v}, Uniform*ui{v}, or Uniform*f{v} commands. • count is greater than one, and the uniform declared in the shader is not an array variable, • no variable with a location of location exists in the program object currently in use and location is not -1, or • a sampler uniform is loaded with any of the Uniform* commands other than Uniform1i{v}. • an image uniform is loaded with any of the Uniform* commands. • there is no active program object in use. To load values into the uniform variables of the default uniform block of a program which may not necessarily be bound, use the commands void ProgramUniform{1234}{if}( uint program, int location, T value ); void ProgramUniform{1234}{if}v( uint program, int location, sizei count, const T *value ); void ProgramUniform{1234}ui( uint program, int location, T value ); void ProgramUniform{1234}uiv( uint program, int location, sizei count, const T *value ); OpenGL ES 3.2 (May 5, 2022)

110

7.6. UNIFORM VARIABLES

111

void ProgramUniformMatrix{234}{f}v( uint program, int location, sizei count, boolean transpose, const T *value ); void ProgramUniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}{f}v( uint program, int location, sizei count, boolean transpose, const T *value ); These commands operate identically to the corresponding commands above without Program in the command name except, rather than updating the currently active program object, these Program commands update the program object named by the initial program parameter.The remaining parameters following the initial program parameter match the parameters for the corresponding nonProgram uniform command. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated if program has not been linked successfully. In addition, all errors described for the corresponding Uniform* commands apply.

7.6.2

Uniform Blocks

The values of uniforms arranged in named uniform blocks are extracted from buffer object storage. The mechanisms for placing individual uniforms in a buffer object and connecting a uniform block to an individual buffer object are described below. If the number of active uniform blocks referenced by the shaders in a program exceeds implementation-dependent limits, the program will fail to link. The limits for vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute shaders can be obtained by calling GetIntegerv with pname values of MAX_VERTEX_UNIFORM_BLOCKS, MAX_TESS_CONTROL_UNIFORM_BLOCKS, MAX_TESS_EVALUATION_UNIFORM_BLOCKS, MAX_GEOMETRY_UNIFORM_BLOCKS, MAX_FRAGMENT_UNIFORM_BLOCKS, and MAX_COMPUTE_UNIFORM_BLOCKS, respectively. Additionally, a program will fail to link if the sum of the number of active uniform blocks referenced by each shader stage in a program exceeds the value of OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

112

the implementation-dependent limit MAX_COMBINED_UNIFORM_BLOCKS. If a uniform block is referenced by multiple shaders, each such reference counts separately against this combined limit. Finally, the total amount of buffer object storage available for any given uniform block is subject to an implementation-dependent limit. The maximum amount of available space, in basic machine units, can be queried by calling GetIntegerv with a pname of MAX_UNIFORM_BLOCK_SIZE. If the amount of storage required for a uniform block exceeds this limit, a program will fail to link. When a named uniform block is declared by multiple shaders in a program, it must be declared identically in each shader. The uniforms within the block must be declared with the same names, types, and layout qualifiers, in the same order. If a program contains multiple shaders with different declarations for the same named uniform block, the program will fail to link. 7.6.2.1

Uniform Buffer Object Storage

When stored in buffer objects associated with uniform blocks, uniforms are represented in memory as follows: • Members of type bool, int, uint, and float are respectively extracted from a buffer object by reading a single uint, int, uint, or float value at the specified offset. • Vectors with N elements with basic data types of bool, int, uint, or float are extracted as N values in consecutive memory locations beginning at the specified offset, with components stored in order with the first (X) component at the lowest offset. The GL data type used for component extraction is derived according to the rules for scalar members above. • Column-major matrices with C columns and R rows (using the type matCxR or simply matC if C = R) are treated as an array of C column vectors, each consisting of R floating-point components. The column vectors will be stored in order, with column zero at the lowest offset. The difference in offsets between consecutive columns of the matrix will be referred to as the column stride, and is constant across the matrix. The column stride is an implementation-dependent function of the matrix type, and may be determined after a program is linked by querying the MATRIX_STRIDE interface using GetProgramResourceiv (see section 7.3.1). • Row-major matrices with C columns and R rows (using the type matCxR, or simply matC if C = R) are treated as an array of R row vectors, each OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

113

consisting of C floating-point components. The row vectors will be stored in order, with row zero at the lowest offset. The difference in offsets between consecutive rows of the matrix will be referred to as the row stride, and is constant across the matrix. The row stride is an implementation-dependent function of the matrix type, and may be determined after a program is linked by querying the MATRIX_STRIDE interface using GetProgramResourceiv (see section 7.3.1). • Arrays of scalars, vectors, and matrices are stored in memory by element order, with array member zero at the lowest offset. The difference in offsets between each pair of elements in the array in basic machine units is referred to as the array stride, and is constant across the entire array. The array stride is an implementation-dependent function of the array type, and may be determined after a program is linked by querying the ARRAY_STRIDE interface using GetProgramResourceiv (see section 7.3.1). 7.6.2.2

Standard Uniform Block Layout

By default, uniforms contained within a uniform block are extracted from buffer storage in an implementation-dependent manner. Applications may query the offsets assigned to uniforms inside uniform blocks with query functions provided by the GL. The layout qualifier provides shaders with control of the layout of uniforms within a uniform block. When the std140 layout is specified, the offset of each uniform in a uniform block can be derived from the definition of the uniform block by applying the set of rules described below. When using the std140 storage layout, structures will be laid out in buffer storage with its members stored in monotonically increasing order based on their location in the declaration. A structure and each structure member have a base offset and a base alignment, from which an aligned offset is computed by rounding the base offset up to a multiple of the base alignment. The base offset of the first member of a structure is taken from the aligned offset of the structure itself. The base offset of all other structure members is derived by taking the offset of the last basic machine unit consumed by the previous member and adding one. Each structure member is stored in memory at its aligned offset. The members of a toplevel uniform block are laid out in buffer storage by treating the uniform block as a structure with a base offset of zero. 1. If the member is a scalar consuming N basic machine units, the base alignment is N .

OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

114

2. If the member is a two- or four-component vector with components consuming N basic machine units, the base alignment is 2N or 4N , respectively. 3. If the member is a three-component vector with components consuming N basic machine units, the base alignment is 4N . 4. If the member is an array of scalars or vectors, the base alignment and array stride are set to match the base alignment of a single array element, according to rules (1), (2), and (3), and rounded up to the base alignment of a vec4. The array may have padding at the end; the base offset of the member following the array is rounded up to the next multiple of the base alignment. 5. If the member is a column-major matrix with C columns and R rows, the matrix is stored identically to an array of C column vectors with R components each, according to rule (4). 6. If the member is an array of S column-major matrices with C columns and R rows, the matrix is stored identically to a row of S × C column vectors with R components each, according to rule (4). 7. If the member is a row-major matrix with C columns and R rows, the matrix is stored identically to an array of R row vectors with C components each, according to rule (4). 8. If the member is an array of S row-major matrices with C columns and R rows, the matrix is stored identically to a row of S × R row vectors with C components each, according to rule (4). 9. If the member is a structure, the base alignment of the structure is N , where N is the largest base alignment value of any of its members, and rounded up to the base alignment of a vec4. The individual members of this substructure are then assigned offsets by applying this set of rules recursively, where the base offset of the first member of the sub-structure is equal to the aligned offset of the structure. The structure may have padding at the end; the base offset of the member following the sub-structure is rounded up to the next multiple of the base alignment of the structure. If none of the structure members are larger than a vec4, the base alignment of the structure is vec4. 10. If the member is an array of S structures, the S elements of the array are laid out in order, according to rule (9). Shader storage blocks (see section 7.8) also support the std140 layout qualifier, as well as a std430 layout qualifier not supported for uniform blocks. When OpenGL ES 3.2 (May 5, 2022)

7.6. UNIFORM VARIABLES

115

using the std430 storage layout, shader storage blocks will be laid out in buffer storage identically to uniform and shader storage blocks using the std140 layout, except that the base alignment and stride of arrays of scalars and vectors in rule 4 and of structures in rule 9 are not rounded up a multiple of the base alignment of a vec4.

7.6.3

Uniform Buffer Object Bindings

The value of an active uniform inside a named uniform block is extracted from the data store of a buffer object bound to one of an array of uniform buffer binding points. The number of binding points may be queried by calling GetIntegerv with a pname of MAX_COMBINED_UNIFORM_BLOCKS. Regions of buffer objects are bound as storage for uniform blocks by calling BindBuffer* commands (see section 6) with target set to UNIFORM_BUFFER. Each of a program’s active uniform blocks has a corresponding uniform buffer object binding point. The binding is established when a program is linked or relinked, and the initial value of the binding is specified by a layout qualifier (if present), or zero otherwise. The binding point can be assigned by calling: void UniformBlockBinding( uint program, uint uniformBlockIndex, uint uniformBlockBinding ); program is a name of a program object for which the command LinkProgram has been issued in the past. If successful, UniformBlockBinding specifies that program will use the data store of the buffer object bound to the binding point uniformBlockBinding to extract the values of the uniforms in the uniform block identified by uniformBlockIndex. When executing shaders that access uniform blocks, the binding point corresponding to each active uniform block must be populated with a buffer object with a size no smaller than the minimum required size of the uniform block (the value of UNIFORM_BLOCK_DATA_SIZE). For binding points populated by BindBufferRange, the size in question is the value of the size parameter. If any active uniform block is not backed by a sufficiently large buffer object, the results of shader execution may be undefined or modified, as described in section 6.4. Shaders may be executed to process the primitives and vertices specified by any command that transfers vertices to the GL. Errors An INVALID_VALUE error is generated if program is not the name of ei-

OpenGL ES 3.2 (May 5, 2022)

7.7. ATOMIC COUNTER BUFFERS

116

ther a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if uniformBlockIndex is not an active uniform block index of program, or if uniformBlockBinding is greater than or equal to the value of MAX_UNIFORM_BUFFER_BINDINGS.

7.7

Atomic Counter Buffers

The values of atomic counters are backed by buffer object storage. The mechanisms for accessing individual atomic counters in a buffer object and connecting to an atomic counter are described in this section. There is a set of implementation-dependent maximums for the number of active atomic counter buffers referenced by each shader. If the number of atomic counter buffer bindings referenced by any shader in the program exceeds the corresponding limit, the program will fail to link. The limits for vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute shaders can be obtained by calling GetIntegerv with pname values of MAX_VERTEX_ATOMIC_COUNTER_BUFFERS, MAX_TESS_CONTROL_ATOMIC_COUNTER_BUFFERS, MAX_TESS_EVALUATION_ATOMIC_COUNTER_BUFFERS, MAX_GEOMETRY_ATOMIC_COUNTER_BUFFERS, MAX_FRAGMENT_ATOMIC_COUNTER_BUFFERS, and MAX_COMPUTE_ATOMIC_COUNTER_BUFFERS, respectively. Additionally, there is an implementation-dependent limit on the sum of the number of active atomic counter buffers used by each shader stage of a program. If an atomic counter buffer is used by multiple shader stages, each such use counts separately against this combined limit. The combined atomic counter buffer use limit can be obtained by calling GetIntegerv with a pname of MAX_COMBINED_ATOMIC_COUNTER_BUFFERS.

7.7.1

Atomic Counter Buffer Object Storage

Atomic counters stored in buffer objects are represented in memory as follows: • Members of type atomic_uint are extracted from a buffer object by reading a single uint-typed value at the specified offset. • Arrays of type atomic_uint are stored in memory by element order, with array element member zero at the lowest offset. The difference in offsets between each pair of elements in the array in basic machine units is referred to as the array stride, and is constant across the entire array. The array stride OpenGL ES 3.2 (May 5, 2022)

7.8. SHADER BUFFER VARIABLES AND SHADER STORAGE BLOCKS117 (the value of UNIFORM_ARRAY_STRIDE), is always 4, and may be queried after a program is linked.

7.7.2

Atomic Counter Buffer Bindings

The value of an active atomic counter is extracted from or written to the data store of a buffer object bound to one of an array of atomic counter buffer binding points. The number of binding points can be queried by calling GetIntegerv with a pname of MAX_ATOMIC_COUNTER_BUFFER_BINDINGS. Regions of buffer objects are bound as storage for atomic counters by calling one of the BindBuffer* commands (see section 6) with target set to ATOMIC_COUNTER_BUFFER. Each of a program’s active atomic counter buffer bindings has a corresponding atomic counter buffer binding point. This binding point is established with the layout qualifier in the shader text, either explicitly or implicitly, as described in the OpenGL ES Shading Language Specification. When executing shaders that access atomic counters, each active atomic counter buffer must be populated with a buffer object with a size no smaller than the minimum required size for that buffer (the value of BUFFER_DATA_SIZE returned by GetProgramResourceiv). For binding points populated by BindBufferRange, the size in question is the value of the size parameter. If any active atomic counter buffer is not backed by a sufficiently large buffer object, the results of shader execution may be undefined or modified, as described in section 6.4.

7.8

Shader Buffer Variables and Shader Storage Blocks

Shaders can declare named buffer variables, as described in the OpenGL ES Shading Language Specification. Sets of buffer variables are grouped into interface blocks called shader storage blocks. The values of each buffer variable in a shader storage block are read from or written to the data store of a buffer object bound to the binding point associated with the block. Buffer variables are considered active in the same way as uniform variables (see section 7.6). The values of active buffer variables may be changed by executing shaders that assign values to them or perform atomic memory operations on them; by modifying the contents of the bound buffer object’s data store with the commands in sections 6.2, 6.3, and 6.5; by binding a new buffer object to the binding point associated with the block; or by changing the binding point associated with the block. Like buffer variables, shader storage blocks can be active or inactive. Whether a shader storage block is active or inactive is determined in the same way as for

OpenGL ES 3.2 (May 5, 2022)

7.8. SHADER BUFFER VARIABLES AND SHADER STORAGE BLOCKS118 uniform blocks (see section 7.6). Additionally though, all members of a named shader storage block declared with a std430 layout qualifier are considered active, even if they are not referenced in any shader in the program. Buffer variables in shader storage blocks are represented in memory in the same way as uniforms stored in uniform blocks, as described in section 7.6.2.1. When a program is linked successfully, each active buffer variable is assigned an offset relative to the base of the buffer object binding associated with its shader storage block. For buffer variables declared as arrays and matrices, strides between array elements or matrix columns or rows will also be assigned. Offsets and strides of buffer variables will be assigned in an implementation-dependent manner unless the shader storage block is declared using the std140 or std430 storage layout qualifiers. For std140 and std430 shader storage blocks, offsets will be assigned using the method described in section 7.6.2.2. If a program is re-linked, existing buffer variable offsets and strides are invalidated, and a new set of active variables, offsets, and strides will be generated. The total amount of buffer object storage that can be accessed in any shader storage block is subject to an implementation-dependent limit. The maximum amount of available space, in basic machine units, may be queried by calling GetIntegerv with pname MAX_SHADER_STORAGE_BLOCK_SIZE. If the amount of storage required for the fixed size portion of any shader storage block (as reported by BUFFER_DATA_SIZE) exceeds this limit, a program will fail to link. If the number of active shader storage blocks referenced by the shaders in a program exceeds implementation-dependent limits, the program will fail to link. The limits for vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute shaders can be obtained by calling GetIntegerv with pname values of MAX_VERTEX_SHADER_STORAGE_BLOCKS, MAX_TESS_CONTROL_SHADER_STORAGE_BLOCKS, MAX_TESS_EVALUATION_SHADER_STORAGE_BLOCKS, MAX_GEOMETRY_SHADER_STORAGE_BLOCKS, MAX_FRAGMENT_SHADER_STORAGE_BLOCKS, and MAX_COMPUTE_SHADER_STORAGE_BLOCKS, respectively. Additionally, a program will fail to link if the sum of the number of active shader storage blocks referenced by each shader stage in a program exceeds the value of the implementation-dependent limit MAX_COMBINED_SHADER_STORAGE_BLOCKS. If a shader storage block in a program is referenced by multiple shaders, each such reference counts separately against this combined limit. When a named shader storage block is declared by multiple shaders in a program, it must be declared identically in each shader. The buffer variables within the block must be declared with the same names, types, qualification, and declaration order. If a program contains multiple shaders with different declarations for the same named shader storage block, the program will fail to link. OpenGL ES 3.2 (May 5, 2022)

7.9. SAMPLERS

119

Regions of buffer objects are bound as storage for shader storage blocks by calling one of the BindBuffer* commands (see section 6) with target SHADER_STORAGE_BUFFER. Each of a program’s active shader storage blocks has a corresponding shader storage buffer object binding point. When a program object is linked, the shader storage buffer object binding point assigned to each of its active shader storage blocks is reset to the value specified by the corresponding binding layout qualifier, if present, or zero otherwise. It is not possible to change the binding point associated with a shader storage block after a program is linked. When executing shaders that access shader storage blocks, the binding point corresponding to each active shader storage block must be populated with a buffer object with a size no smaller than the minimum required size of the shader storage block (the value of BUFFER_SIZE for the appropriate SHADER_STORAGE_BUFFER resource). For binding points populated by BindBufferRange, the size in question is the value of the size parameter or the size of the buffer minus the value of the offset parameter, whichever is smaller. If any active shader storage block is not backed by a sufficiently large buffer object, the results of shader execution may be undefined or modified, as described in section 6.4.

7.9

Samplers

Samplers are special uniforms used in the OpenGL ES Shading Language to identify the texture object used for each texture lookup. The value of a sampler indicates the texture image unit being accessed. Setting a sampler’s value to i selects texture image unit number i. The values of i ranges from zero to the implementation-dependent maximum supported number of texture image units minus one. The type of the sampler identifies the target on the texture image unit, as shown in table 7.3 for sampler* types. The texture object bound to that texture image unit’s target is then used for the texture lookup. For example, a variable of type sampler2D selects target TEXTURE_2D on its texture image unit. Binding of texture objects to targets is done as usual with BindTexture. Selecting the texture image unit to bind to is done as usual with ActiveTexture. The location of a sampler is queried with GetUniformLocation, just like any uniform variable. Sampler values must be set by calling Uniform1i{v}. Errors

OpenGL ES 3.2 (May 5, 2022)

7.10. IMAGES

120

It is not allowed to have variables of different sampler types pointing to the same texture image unit within a program object. This situation can only be detected at the next rendering command issued which triggers shader invocations, and an INVALID_OPERATION error will then be generated. Active samplers are samplers actually being used in a program object. The LinkProgram command determines if a sampler is active or not. The LinkProgram command will attempt to determine if the active samplers in the shader(s) contained in the program object exceed the maximum allowable limits. If it determines that the count of active samplers exceeds the allowable limits, then the link fails (these limits can be different for different types of shaders). Each active sampler variable counts against the limit, even if multiple samplers refer to the same texture image unit.

7.10

Images

Images are special uniforms used in the OpenGL ES Shading Language to identify a level of a texture to be read or written using built-in image load, store or atomic functions in the manner described in section 8.23. The value of an image uniform is an integer specifying the image unit accessed. Image units are numbered beginning at zero, and there is an implementation-dependent number of available image units (the value of MAX_IMAGE_UNITS). Note that image units used for image variables are independent of the texture image units used for sampler variables; the number of units provided by the implementation may differ. Textures are bound independently and separately to image and texture image units. The type of an image variable must match the texture target of the image currently bound to the image unit, otherwise the result of a load, store or atomic operation is undefined (see section 4.1.7.2 of the OpenGL ES Shading Language Specification for more details). The location of an image variable is queried with GetUniformLocation, just like any uniform variable. There is a limit on the number of active image variables that may be used by a program or by any particular shader.

7.11

Shader Memory Access

As described in the OpenGL ES Shading Language Specification, shaders may perform random-access reads and writes to buffer object memory by reading from, OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

121

assigning to, or performing atomic memory operation on shader buffer variables, or to texture or buffer object memory by using built-in image load, store, or atomic functions operating on shader image variables, or performing atomic operations on atomic counter variables. The ability to perform such random-access reads and writes in systems that may be highly pipelined results in ordering and synchronization issues discussed in the sections below.

7.11.1

Shader Memory Access Ordering

The order in which texture or buffer object memory is read or written by shaders is largely undefined. For some shader types (vertex, and in some cases, fragment), even the number of shader invocations that might perform loads and stores is undefined. In particular, the following rules apply: • While a vertex or tessellation evaluation shader will be executed at least once for each unique vertex specified by the application, (vertex shaders) or generated by the tessellation primitive generator (tessellation evaluation shaders), it may be executed more than once for implementation-dependent reasons. Additionally, if the same vertex is specified multiple times in a collection of primitives (e.g., repeating an index in DrawElements), the vertex shader might be run only once. • For each fragment generated by the GL, the number of fragment shader invocations depends on a number of factors. If the fragment fails the pixel ownership test (see section 13.8.1), scissor test (see section 13.8.2), or is discarded by any of the multisample fragment operations (see section 13.8.3), the fragment shader will not be executed In addition, if early per-fragment tests are enabled (see section 13.8), the fragment shader will not be executed if the fragment is discarded during the early per-fragment tests, and a fragment may not be executed if the fragment will never contribute to the framebuffer. For example, if a fragment A written to a pixel or sample from primitive A will be replaced by a fragment B written to a pixel or sample from primitive B, then fragment A may not be executed even if primitive A is specified prior to primitive B. When fragment shaders are executed, the number of invocations per fragment is exactly one when the framebuffer has no multisample buffer (the value of SAMPLE_BUFFERS is zero). Otherwise, the number of invocations OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

122

is in the range [1, N ] where N is the number of samples covered by the fragment; if the fragment shader specifies per-sample shading, it will be invoked exactly N times. • If a fragment shader is invoked to process fragments or samples not covered by a primitive being rasterized to facilitate the approximation of derivatives for texture lookups, stores have no effect. • The relative order of invocations of the same shader type are undefined. A store issued by a shader when working on primitive B might complete prior to a store for primitive A, even if primitive A is specified prior to primitive B. This applies even to fragment shaders; while fragment shader outputs are written to the framebuffer in primitive order, stores executed by fragment shader invocations are not. • The relative order of invocations of different shader types is undefined. The above limitations on shader invocation order also make some forms of synchronization between shader invocations within a single set of primitives unimplementable. For example, having one invocation poll memory written by another invocation assumes that the other invocation has been launched and can complete its writes. Stores issued to different memory locations within a single shader invocation may not be visible to other invocations in the order they were performed. The builtin function memoryBarrier may be used to provide stronger ordering of reads and writes performed by a single invocation. Calling memoryBarrier guarantees that any memory transactions issued by the shader invocation prior to the call complete prior to the memory transactions issued after the call. Memory barriers may be needed for algorithms that require multiple invocations to access the same memory and require the operations to be performed in a partially-defined relative order. For example, if one shader invocation does a series of writes, followed by a memoryBarrier call, followed by another write, and another invocation sees the result of the final write and calls memoryBarrier, then the second invocation will also see the previous writes. Without either memory barrier, the final write may be visible before the previous writes. The built-in atomic memory transaction functions may be used to read and write a given memory address atomically. While built-in atomic functions issued by multiple shader invocations are executed in undefined order relative to each other, these functions perform both a read and a write of a memory address and guarantee that no other memory transaction will write to the underlying memory between the read and write. Atomics allow shaders to use shared global addresses for mutual exclusion or as counters, among other uses. OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

7.11.2

123

Shader Memory Access Synchronization

Data written to textures or buffer objects by a shader invocation may eventually be read by other shader invocations, sourced by other fixed pipeline stages, or read back by the application. When data is written using API commands such as TexSubImage* or BufferSubData, the GL implementation knows when and where writes occur and can perform implicit synchronization to ensure that operations requested before the update see the original data and that subsequent operations see the modified data. Without logic to track the target address of each shader instruction performing a store, automatic synchronization of stores performed by a shader invocation would require the GL implementation to make worst-case assumptions at significant performance cost. To permit cases where textures or buffers may be read or written in different pipeline stages without the overhead of automatic synchronization, buffer object and texture stores performed by shaders are not automatically synchronized with other GL operations using the same memory. Explicit synchronization is required to ensure that the effects of buffer and texture data stores performed by shaders will be visible to subsequent operations using the same objects and will not overwrite data still to be read by previously requested operations. Without manual synchronization, shader stores for a “new” primitive may complete before processing of an “old” primitive completes. Additionally, stores for an “old” primitive might not be completed before processing of a “new” primitive starts. The command void MemoryBarrier( bitfield barriers ); defines a barrier ordering the memory transactions issued prior to the command relative to those issued after the barrier. For the purposes of this ordering, memory transactions performed by shaders are considered to be issued by the rendering command that triggered the execution of the shader. barriers is a bitfield indicating the set of operations that are synchronized with shader stores; the bits used in barriers are as follows: • VERTEX_ATTRIB_ARRAY_BARRIER_BIT: If set, vertex data sourced from buffer objects after the barrier will reflect data written by shaders prior to the barrier. The set of buffer objects affected by this bit is derived from the buffer object bindings used for arrays of generic vertex attributes (VERTEX_ATTRIB_ARRAY_BUFFER bindings). • ELEMENT_ARRAY_BARRIER_BIT: If set, vertex array indices sourced from buffer objects after the barrier will reflect data written by shaders prior to the barrier. The buffer objects affected by this bit are derived from the ELEMENT_ARRAY_BUFFER binding. OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

124

• UNIFORM_BARRIER_BIT: Shader uniforms sourced from buffer objects after the barrier will reflect data written by shaders prior to the barrier. • TEXTURE_FETCH_BARRIER_BIT: Texture fetches from shaders, including fetches from buffer object memory via buffer textures, after the barrier will reflect data written by shaders prior to the barrier. • SHADER_IMAGE_ACCESS_BARRIER_BIT: Memory accesses using shader built-in image load and store functions issued after the barrier will reflect data written by shaders prior to the barrier. Additionally, image stores issued after the barrier will not execute until all memory accesses (e.g., loads, stores, texture fetches, vertex fetches) initiated prior to the barrier complete. • COMMAND_BARRIER_BIT: Command data sourced from buffer objects by Draw*Indirect and DispatchComputeIndirect commands after the barrier will reflect data written by shaders prior to the barrier. The buffer objects affected by this bit are derived from the DRAW_INDIRECT_BUFFER and DISPATCH_INDIRECT_BUFFER bindings. • PIXEL_BUFFER_BARRIER_BIT: Reads/writes of buffer objects via the PIXEL_PACK_BUFFER and PIXEL_UNPACK_BUFFER bindings (ReadPixels, TexSubImage, etc.) after the barrier will reflect data written by shaders prior to the barrier. Additionally, buffer object writes issued after the barrier will wait on the completion of all shader writes initiated prior to the barrier. • TEXTURE_UPDATE_BARRIER_BIT: Writes to a texture via Tex(Sub)Image*, CopyTex*, or CompressedTex* after the barrier will reflect data written by shaders prior to the barrier. Additionally, texture writes from these commands issued after the barrier will not execute until all shader writes initiated prior to the barrier complete. • BUFFER_UPDATE_BARRIER_BIT: Reads and writes to buffer object memory after the barrier using the commands in sections 6.2, 6.3, and 6.5 will reflect data written by shaders prior to the barrier. Additionally, writes via these commands issued after the barrier will wait on the completion of any shader writes to the same memory initiated prior to the barrier. • FRAMEBUFFER_BARRIER_BIT: Reads and writes via framebuffer object attachments after the barrier will reflect data written by shaders prior to the barrier. Additionally, framebuffer writes issued after the barrier will wait on the completion of all shader writes issued prior to the barrier.

OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

125

• TRANSFORM_FEEDBACK_BARRIER_BIT: Writes via transform feedback bindings after the barrier will reflect data written by shaders prior to the barrier. Additionally, transform feedback writes issued after the barrier will wait on the completion of all shader writes issued prior to the barrier. • ATOMIC_COUNTER_BARRIER_BIT: Accesses to atomic counters after the barrier will reflect writes prior to the barrier. • SHADER_STORAGE_BARRIER_BIT: Memory accesses using shader buffer variables issued after the barrier will reflect data written by shaders prior to the barrier. Additionally, assignments to and atomic operations performed on shader buffer variables after the barrier will not execute until all memory accesses (e.g., loads, stores, texture fetches, vertex fetches) initiated prior to the barrier complete. If barriers is ALL_BARRIER_BITS, shader memory accesses will be synchronized relative to all the operations described above. Errors An INVALID_VALUE error is generated if barriers is not the special value ALL_BARRIER_BITS, and has any bits set other than those described above. Implementations may cache buffer object or texture image memory that could be written by shaders in multiple caches; for example, there may be separate caches for texture, vertex fetching, and one or more caches for shader memory accesses. Implementations are not required to keep these caches coherent with shader memory writes. Stores issued by one invocation may not be immediately observable by other pipeline stages or other shader invocations because the value stored may remain in a cache local to the processor executing the store, or because data overwritten by the store is still in a cache elsewhere in the system. When MemoryBarrier is called, the GL flushes and/or invalidates any caches relevant to the operations specified by the barriers parameter to ensure consistent ordering of operations across the barrier. To allow for independent shader invocations to communicate by reads and writes to a common memory address, image variables in the OpenGL ES Shading Language may be declared as coherent. Buffer object or texture image memory accessed through such variables may be cached only if caches are automatically updated due to stores issued by any other shader invocation. If the same address is accessed using both coherent and non-coherent variables, the accesses using variables declared as coherent will observe the results stored using coherent variables in other invocations. Using variables declared as coherent guarantees only OpenGL ES 3.2 (May 5, 2022)

7.11. SHADER MEMORY ACCESS

126

that the results of stores will be immediately visible to shader invocations using similarly-declared variables; calling MemoryBarrier is required to ensure that the stores are visible to other operations. The following guidelines may be helpful in choosing when to use coherent memory accesses and when to use barriers. • Data that are read-only or constant may be accessed without using coherent variables or calling MemoryBarrier. Updates to the read-only data via commands such as BufferSubData will invalidate shader caches implicitly as required. • Data that are shared between shader invocations at a fine granularity (e.g., written by one invocation, consumed by another invocation) should use coherent variables to read and write the shared data. • Data written to image variables in one rendering pass and read by the shader in a later pass need not use coherent variables or memoryBarrier. Calling MemoryBarrier with the SHADER_IMAGE_ACCESS_BARRIER_BIT set in barriers between passes is necessary. • Data written by the shader in one rendering pass and read by another mechanism (e.g., vertex or index buffer pulling) in a later pass need not use coherent variables or memoryBarrier. Calling MemoryBarrier with the appropriate bits set in barriers between passes is necessary. The command void MemoryBarrierByRegion( bitfield barriers ); behave as described above for MemoryBarrier, with two differences: First, it narrows the region under consideration so that only reads/writes of prior fragment shaders that are invoked for a smaller region of the framebuffer will be completed/reflected prior to subsequent reads/write of following fragment shaders. The size of the region is implementation dependent and may be as small as one framebuffer pixel. Second, it only applies to memory transactions that may be read by or written by a fragment shader. Therefore, only the barrier bits • ATOMIC_COUNTER_BARRIER_BIT • FRAMEBUFFER_BARRIER_BIT • SHADER_IMAGE_ACCESS_BARRIER_BIT OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

127

• SHADER_STORAGE_BARRIER_BIT • TEXTURE_FETCH_BARRIER_BIT • UNIFORM_BARRIER_BIT are supported. When barriers is ALL_BARRIER_BITS, shader memory accesses will be synchronized relative to all these barrier bits, but not to other barrier bits specific to MemoryBarrier. This implies that reads/writes for scatter/gather-like algorithms may or may not be completed/reflected after a MemoryBarrierByRegion command. However, for uses such as deferred shading, where a linked list of visible surfaces with the head at a framebuffer address may be constructed, and the entirety of the list is only dependent on previous executions at that framebuffer address, MemoryBarrierByRegion may be significantly more efficient than MemoryBarrier. Errors An INVALID_VALUE error is generated if barriers is not the special value ALL_BARRIER_BITS, and has any bits set other than those described above.

7.12

Shader, Program, and Program Pipeline Queries

The command void GetShaderiv( uint shader, enum pname, int *params ); returns properties of the shader object named shader in params. The parameter value to return is specified by pname. If pname is SHADER_TYPE, one of the values from table 7.1 corresponding to the type of shader is returned. If pname is DELETE_STATUS, TRUE is returned if the shader has been flagged for deletion and FALSE is returned otherwise. If pname is COMPILE_STATUS, TRUE is returned if the shader was last compiled successfully, and FALSE is returned otherwise. If pname is INFO_LOG_LENGTH, the length of the info log, including a null terminator, is returned. If there is no info log, zero is returned. If pname is SHADER_SOURCE_LENGTH, the length of the concatenation of the source strings making up the shader source, including a null terminator, is returned. If no source has been defined, zero is returned. OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

128

Errors An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. An INVALID_ENUM error is generated if pname is not SHADER_TYPE, DELETE_STATUS, COMPILE_STATUS, INFO_LOG_LENGTH, or SHADER_SOURCE_LENGTH. The command void GetProgramiv( uint program, enum pname, int *params ); returns properties of the program object named program in params. The parameter value to return is specified by pname. Most properties set within program objects are specified not to take effect until the next call to LinkProgram or ProgramBinary. Some properties further require a successful call to either of these commands before taking effect. GetProgramiv returns the properties currently in effect for program, which may differ from the properties set within program since the most recent call to LinkProgram or ProgramBinary, which have not yet taken effect. If there has been no such call putting changes to pname into effect, initial values are returned. If pname is DELETE_STATUS, TRUE is returned if the program has been flagged for deletion, and FALSE is returned otherwise. If pname is LINK_STATUS, TRUE is returned if the program was last linked successfully, and FALSE is returned otherwise. If pname is VALIDATE_STATUS, TRUE is returned if the last call to ValidateProgram (see section 11.1.3.11) with program was successful, and FALSE is returned otherwise. If pname is INFO_LOG_LENGTH, the length of the info log, including a null terminator, is returned. If there is no info log, zero is returned. If pname is ATTACHED_SHADERS, the number of objects attached is returned. If pname is ACTIVE_ATTRIBUTES, the number of active attributes (see section 7.3.1) in program is returned. If no active attributes exist, zero is returned. If pname is ACTIVE_ATTRIBUTE_MAX_LENGTH, the length of the longest active attribute name, including a null terminator, is returned. If no active attributes exist, zero is returned. If pname is ACTIVE_UNIFORMS, the number of active uniforms is returned. If no active uniforms exist, zero is returned. OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

129

If pname is ACTIVE_UNIFORM_MAX_LENGTH, the length of the longest active uniform name, including a null terminator, is returned. If no active uniforms exist, zero is returned. If pname is TRANSFORM_FEEDBACK_BUFFER_MODE, the buffer mode used when transform feedback (see section 11.1.2.1) is active is returned. It can be one of SEPARATE_ATTRIBS or INTERLEAVED_ATTRIBS. If pname is TRANSFORM_FEEDBACK_VARYINGS, the number of output variables to capture in transform feedback mode for the program is returned. If pname is TRANSFORM_FEEDBACK_VARYING_MAX_LENGTH, the length of the longest output variable name specified to be used for transform feedback, including a null terminator, is returned. If no outputs are used for transform feedback, zero is returned. If pname is ACTIVE_UNIFORM_BLOCKS, the number of uniform blocks for program containing active uniforms is returned. If pname is ACTIVE_UNIFORM_BLOCK_MAX_NAME_LENGTH, the length of the longest active uniform block name, including the null terminator, is returned. If pname is GEOMETRY_VERTICES_OUT, the maximum number of vertices the geometry shader (see section 11.3) will output is returned. If pname is GEOMETRY_INPUT_TYPE, the geometry shader input type, which must be one of POINTS, LINES, LINES_ADJACENCY, TRIANGLES or TRIANGLES_ADJACENCY, is returned. If pname is GEOMETRY_OUTPUT_TYPE, the geometry shader output type, which must be one of POINTS, LINE_STRIP or TRIANGLE_STRIP, is returned. If pname is GEOMETRY_SHADER_INVOCATIONS, the number of geometry shader invocations per primitive will be returned. If pname is TESS_CONTROL_OUTPUT_VERTICES, the number of vertices in the tessellation control shader (see section 11.2.1) output patch is returned. If pname is TESS_GEN_MODE, QUADS, TRIANGLES, or ISOLINES is returned, depending on the primitive mode declaration in the tessellation evaluation shader (see section 11.2.3). If pname is TESS_GEN_SPACING, EQUAL, FRACTIONAL_EVEN, or FRACTIONAL_ODD is returned, depending on the spacing declaration in the tessellation evaluation shader. If pname is TESS_GEN_VERTEX_ORDER, CCW or CW is returned, depending on the vertex order declaration in the tessellation evaluation shader. If pname is TESS_GEN_POINT_MODE, TRUE is returned if point mode is enabled in a tessellation evaluation shader declaration; FALSE is returned otherwise. If pname is COMPUTE_WORK_GROUP_SIZE, an array of three integers containing the workgroup size of the compute program (see chapter 17), as specified by its input layout qualifier(s), is returned OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

130

If pname is PROGRAM_SEPARABLE, TRUE is returned if the program has been flagged for use as a separable program object that can be bound to individual shader stages with UseProgramStages. If pname is PROGRAM_BINARY_RETRIEVABLE_HINT, the value of whether the binary retrieval hint is enabled for program is returned. If pname is ACTIVE_ATOMIC_COUNTER_BUFFERS, the number of active atomic counter buffers used by program is returned. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_ENUM error is generated if pname is not one of the values listed above. An INVALID_OPERATION error is generated if GEOMETRY_VERTICES_OUT, GEOMETRY_INPUT_TYPE, GEOMETRY_OUTPUT_TYPE, or GEOMETRY_SHADER_INVOCATIONS are queried for a program which has not been linked successfully, or which does not contain objects to form a geometry shader. An INVALID_OPERATION error is generated if TESS_CONTROL_OUTPUT_VERTICES is queried for a program which has not been linked successfully, or which does not contain objects to form a tessellation control shader. An INVALID_OPERATION error is generated if TESS_GEN_MODE, TESS_GEN_SPACING, TESS_GEN_VERTEX_ORDER, or TESS_GEN_POINT_MODE are queried for a program which has not been linked successfully, or which does not contain objects to form a tessellation evaluation shader, An INVALID_OPERATION error is generated if COMPUTE_WORK_GROUP_SIZE is queried for a program which has not been linked successfully, or which does not contain objects to form a compute shader, The command void GetProgramPipelineiv( uint pipeline, enum pname, int *params ); returns properties of the program pipeline object named pipeline in params. The parameter value to return is specified by pname. If pipeline is a name that has been generated (without subsequent deletion) by GenProgramPipelines, but refers to a program pipeline object that has not been OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

131

previously bound, the GL first creates a new state vector in the same manner as when BindProgramPipeline creates a new program pipeline object. If pname is ACTIVE_PROGRAM, the name of the active program object (used for uniform updates) of pipeline is returned. If pname is one of the shader stage type arguments in table 7.1, the name of the program object current for the corresponding shader stage of pipeline returned. If pname is VALIDATE_STATUS, the validation status of pipeline, as determined by ValidateProgramPipeline (see section 11.1.3.11) is returned. If pname is INFO_LOG_LENGTH, the length of the info log for pipeline, including a null terminator, is returned. If there is no info log, zero is returned. Errors An INVALID_OPERATION error is generated if pipeline is not a name returned from a previous call to GenProgramPipelines or if such a name has since been deleted by DeleteProgramPipelines. An INVALID_ENUM error is generated if pname is not ACTIVE_PROGRAM, INFO_LOG_LENGTH, VALIDATE_STATUS, or one of the type arguments in table 7.1. The command void GetAttachedShaders( uint program, sizei maxCount, sizei *count, uint *shaders ); returns the names of shader objects attached to program in shaders. The actual number of shader names written into shaders is returned in count. If no shaders are attached, count is set to zero. If count is NULL then it is ignored. The maximum number of shader names that may be written into shaders is specified by maxCount. The number of objects attached to program is given by can be queried by calling GetProgramiv with ATTACHED_SHADERS. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if maxCount is negative. A string that contains information about the last compilation attempt on a shader object, last link or validation attempt on a program object, or last validaOpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

132

tion attempt on a program pipeline object, called the info log, can be obtained with the commands void GetShaderInfoLog( uint shader, sizei bufSize, sizei *length, char *infoLog ); void GetProgramInfoLog( uint program, sizei bufSize, sizei *length, char *infoLog ); void GetProgramPipelineInfoLog( uint pipeline, sizei bufSize, sizei *length, char *infoLog ); These commands return an info log string for the corresponding type of object in infoLog. This string will be null-terminated even if the INFO_LOG_LENGTH query returns zero. The actual number of characters written into infoLog, excluding the null terminator, is returned in length. If length is NULL, then no length is returned. The maximum number of characters that may be written into infoLog, including the null terminator, is specified by bufSize. The number of characters in the info log for a shader object, program object, or program pipeline object can be queried respectively with GetShaderiv, GetProgramiv, or GetProgramPipelineiv with pname INFO_LOG_LENGTH. If shader is a shader object, GetShaderInfoLog will return either an empty string or information about the last compilation attempt for that object. If program is a program object, GetProgramInfoLog will return either an empty string or information about the last link attempt or last validation attempt (see section 11.1.3.11) for that object. If pipeline is a program pipeline object, GetProgramPipelineInfoLog will return either an empty string or information about the last validation attempt for that object. The info log is typically only useful during application development and an application should not expect different GL implementations to produce identical info logs. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a

OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

133

program object. An INVALID_VALUE error is generated if pipeline is not the name of an existing program pipeline object. An INVALID_VALUE error is generated if bufSize is negative. The command void GetShaderSource( uint shader, sizei bufSize, sizei *length, char *source ); returns in source the string making up the source code for the shader object shader. The string source will be null-terminated. The actual number of characters written into source, excluding the null terminator, is returned in length. If length is NULL, no length is returned. The maximum number of characters that may be written into source, including the null terminator, is specified by bufSize. The string source is a concatenation of the strings passed to the GL using ShaderSource. The length of this concatenation is given by SHADER_SOURCE_LENGTH, which can be queried with GetShaderiv. Errors An INVALID_VALUE error is generated if shader is not the name of either a program or shader object. An INVALID_OPERATION error is generated if shader is the name of a program object. An INVALID_VALUE error is generated if bufSize is negative. The command void GetShaderPrecisionFormat( enum shadertype, enum precisiontype, int *range, int *precision ); returns the range and precision for different numeric formats supported by the shader compiler. shadertype must be VERTEX_SHADER or FRAGMENT_SHADER. precisiontype must be one of LOW_FLOAT, MEDIUM_FLOAT, HIGH_FLOAT, LOW_INT, MEDIUM_INT or HIGH_INT. range points to an array of two integers in which encodings of the format’s numeric range are returned. If min and max are the smallest and largest values representable in the format, then the values returned are defined to be range[0] = blog2 (|min|)c OpenGL ES 3.2 (May 5, 2022)

7.12. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

134

range[1] = blog2 (|max|)c precision points to an integer in which the log2 value of the number of bits of precision of the format is returned. If the smallest representable value greater than 1 is 1 + , then *precision will contain b−log2 ()c, and every value in the range [−2range[0] , 2range[1] ] can be represented to at least one part in 2∗precision . For example, an IEEE singleprecision floating-point format would return range[0] = 127, range[1] = 127, and ∗precision = 23, while a 32-bit two’s-complement integer format would return range[0] = 31, range[1] = 30, and ∗precision = 0. The minimum required precision and range for formats corresponding to the different values of precisiontype are described in section 4.5 (“Precision and Precision Qualifiers”) of the OpenGL ES Shading Language Specification. Errors An INVALID_ENUM error is generated if shadertype is not VERTEX_SHADER or FRAGMENT_SHADER. The commands void GetUniformfv( uint program, int location, float *params ); void GetUniformiv( uint program, int location, int *params ); void GetUniformuiv( uint program, int location, uint *params ); void GetnUniformfv( uint program, int location, sizei bufSize, float *params ); void GetnUniformiv( uint program, int location, sizei bufSize, int *params ); void GetnUniformuiv( uint program, int location, sizei bufSize, uint *params ); return the value or values of the uniform at location location of the default uniform block for program object program in the array params. The type of the uniform at location determines the number of values returned. In order to query the values of an array of uniforms, a GetUniform* command needs to be issued for each array element. If the uniform queried is a matrix, the values of the matrix are returned in column major order. OpenGL ES 3.2 (May 5, 2022)

7.13. REQUIRED STATE

135

Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated if program has not been linked successfully, or if location is not a valid location for program. An INVALID_OPERATION error is generated by GetnUniform* if the buffer size required to store the requested data is greater than bufSize.

7.13

Required State

The GL maintains state to indicate which shader and program object names are in use. Initially, no shader or program objects exist, and no names are in use. The state required per shader object consists of: • An unsigned integer specifying the shader object name. • An integer holding the value of SHADER_TYPE. • A boolean holding the delete status, initially FALSE. • A boolean holding the status of the last compile, initially FALSE. • An array of type char containing the information log, initially empty. • An integer holding the length of the information log. • An array of type char containing the concatenated shader string, initially empty. • An integer holding the length of the concatenated shader string. The state required per program object consists of: • An unsigned integer indicating the program object name. • A boolean holding the delete status, initially FALSE. • A boolean holding the status of the last link attempt, initially FALSE. • A boolean holding the status of the last validation attempt, initially FALSE. OpenGL ES 3.2 (May 5, 2022)

7.13. REQUIRED STATE

136

• An integer holding the number of attached shader objects. • A list of unsigned integers to keep track of the names of the shader objects attached. • An array of type char containing the information log, initially empty. • An integer holding the length of the information log. • An integer holding the number of active uniforms. • For each active uniform, three integers, holding its location, size, and type, and an array of type char holding its name. • An array holding the values of each active uniform. • An integer holding the number of active attributes. • For each active attribute, three integers holding its location, size, and type, and an array of type char holding its name. • A boolean holding the hint to the retrievability of the program binary, initially FALSE. Additional state required to support transform feedback consists of: • An integer holding the transform feedback mode, initially INTERLEAVED_ATTRIBS. • An integer holding the number of outputs to be captured, initially zero. • An integer holding the length of the longest output name being captured, initially zero. • For each output being captured, two integers holding its size and type, and an array of type char holding its name. Additionally, one unsigned integer is required to hold the name of the current program object, if any. This list of program object state is not complete. Tables 21.21-21.29 describe additional program object state specific to program binaries, geometry shaders, tessellation control and evaluation shaders, and uniform blocks. Table 21.30 describes state related to vertex and geometry shaders that is not program object state. OpenGL ES 3.2 (May 5, 2022)

Chapter 8

Textures and Samplers Texturing maps a portion of one or more specified images onto a fragment or vertex. This mapping is accomplished in shaders by sampling the color of an image at the location indicated by specified (s, t, r) texture coordinates. Texture lookups are typically used to modify a fragment’s RGBA color but may be used for any purpose in a shader. This chapter first describes how pixel rectangles, texture images, and texture and sampler object parameters are specified and queried, in sections 8.1-8.11. The remainder of the chapter in sections 8.12-8.23 describe how texture sampling is performed in shaders. The internal data type of a texture may be signed or unsigned normalized fixedpoint, signed or unsigned integer, or floating-point, depending on the internal format of the texture. The correspondence between the internal format and the internal data type is given in tables 8.10-8.11. Fixed-point and floating-point textures return a floating-point value and integer textures return signed or unsigned integer values. The fragment shader is responsible for interpreting the result of a texture lookup as the correct data type, otherwise the result is undefined. Each of the supported types of texture is a collection of texture images built from two-or three-dimensional arrays of texels (see section 2.6.6). Two- and threedimensional textures consist respectively of two-or three-dimensional texture images. Two-dimensional array textures are arrays of two-dimensional images. Each image consists of one or more layers. Two-dimensional multisample textures are special two-dimensional textures containing multiple samples in each texel. Cube maps are special two-dimensional array textures with six layers that represent the faces of a cube. When accessing a cube map, the texture coordinates are projected onto one of the six faces of the cube. A cube map array is a collection of cube map layers stored as a two-dimensional array texture. When accessing a cube map ar-

137

8.1. TEXTURE OBJECTS

138

ray, the texture coordinates s, t, and r are applied similarly as cube maps while the last texture coordinate q is used as the index of one of the cube map slices. Buffer textures are special one-dimensional textures whose texture images are stored in separate buffer objects. Implementations must support texturing using multiple images. The following subsections (up to and including section 8.14) specify the GL operation with a single texture. Multiple texture images may be sampled and combined by shaders as described in section 11.1.3.5. The coordinates used for texturing in a fragment shader are defined by the OpenGL ES Shading Language Specification. The command void ActiveTexture( enum texture ); specifies the active texture unit selector. The selector may be queried by calling GetIntegerv with pname set to ACTIVE_TEXTURE. Each texture image unit consists of all the texture state defined in chapter 8. The active texture unit selector selects the texture image unit accessed by commands involving texture image processing. Such commands include TexParameter, TexImage, BindTexture, and queries of all such state. Errors An INVALID_ENUM error is generated if an invalid texture is specified. texture is a symbolic constant of the form TEXTUREi, indicating that texture unit i is to be modified. Each TEXTUREi adheres to TEXTUREi = TEXTURE0 + i, where i is in the range zero to k −1, and k is the value of MAX_COMBINED_TEXTURE_IMAGE_UNITS)a . a

The OpenGL ES header file only defines symbolic constants TEXTURE0 through TEXTURE31. Applications accessing more than 32 texture image units must use the TEXTURE0 + i approach rather than an explicit symbolic constant.

The state required for the active texture image unit selector is a single integer. The initial value is TEXTURE0.

8.1

Texture Objects

Textures in GL are represented by named objects. The name space for texture objects is the unsigned integers, with zero reserved by the GL to represent the default texture object. The default texture object is bound to each of the TEXTURE_2D, OpenGL ES 3.2 (May 5, 2022)

8.1. TEXTURE OBJECTS

139

TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_BUFFER, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_2D_MULTISAMPLE, and TEXTURE_2D_MULTISAMPLE_ARRAY targets during context initialization.

A new texture object is created by binding an unused name to one of these texture targets. The command void GenTextures( sizei n, uint *textures );; returns n previously unused texture names in textures. These names are marked as used, for the purposes of GenTextures only, but they acquire texture state and a dimensionality only when they are first bound, just as if they were unused. Errors An INVALID_VALUE error is generated if n is negative. The binding is effected by calling void BindTexture( enum target, uint texture ); with target set to the desired texture target and texture set to the unused name. The resulting texture object is a new state vector, comprising all the state and with the same initial values listed in section 8.19 The new texture object bound to target is, and remains a texture of the dimensionality and type specified by target until it is deleted. BindTexture may also be used to bind an existing texture object to any of these targets. If the bind is successful no change is made to the state of the bound texture object, and any previous binding to target is broken. While a texture object is bound, GL operations on the target to which it is bound affect the bound object, and queries of the target to which it is bound return state from the bound object. If texture mapping of the dimensionality of the target to which a texture object is bound is enabled, the state of the bound texture object directs the texturing operation. Errors An INVALID_ENUM error is generated if target is not one of the texture targets described in the introduction to section 8.1. An INVALID_OPERATION error is generated if an attempt is made to bind a texture object of different dimensionality than the specified target. Texture objects are deleted by calling OpenGL ES 3.2 (May 5, 2022)

8.2. SAMPLER OBJECTS

140

void DeleteTextures( sizei n, const uint *textures ); textures contains n names of texture objects to be deleted. After a texture object is deleted, it has no contents or dimensionality, and its name is again unused. If a texture that is currently bound to any of the target bindings of BindTexture is deleted, it is as though BindTexture had been executed with the same target and texture zero. Additionally, special care must be taken when deleting a texture if any of the images of the texture are attached to a framebuffer object. See section 9.2.8 for details. Unused names in textures that have been marked as used for the purposes of GenTextures are marked as unused again. Unused names in textures are silently ignored, as is the name zero. Errors An INVALID_VALUE error is generated if n is negative. The command boolean IsTexture( uint texture ); returns TRUE if texture is the name of a texture object. If texture is zero, or is a nonzero value that is not the name of a texture object, or if an error condition occurs, IsTexture returns FALSE. The texture object name space, including the initial two- and three- dimensional, two-dimensional array, buffer, cube map, cube map array, two-dimensional multisample, and two-dimensional multisample array texture objects, is shared among all texture units. A texture object may be bound to more than one texture unit simultaneously. After a texture object is bound, any GL operations on that target object affect any other texture units to which the same texture object is bound. Texture binding is affected by the setting of the state ACTIVE_TEXTURE. If a texture object is deleted, it as if all texture units which are bound to that texture object are rebound to texture object zero.

8.2

Sampler Objects

The state necessary for texturing can be divided into two categories as described in section 8.19. A GL texture object includes both categories. The first category represents dimensionality and other image parameters, and the second category OpenGL ES 3.2 (May 5, 2022)

8.2. SAMPLER OBJECTS

141

represents sampling state. Additionally, a sampler object may be created to encapsulate only the second category - the sampling state – of a texture object. A new sampler object is created by binding an unused name to a texture unit. The command void GenSamplers( sizei count, uint *samplers ); returns count previously unused sampler object names in samplers. The name zero is reserved by the GL to represent no sampler being bound to a sampler unit. The names are marked as used, for the purposes of GenSamplers only, but they acquire state only when they are first used as a parameter to BindSampler, SamplerParameter*, GetSamplerParameter*, or IsSampler. When a sampler object is first used in one of these functions, the resulting sampler object is initialized with a new state vector, comprising all the state and with the same initial values listed in table 21.121 . Errors An INVALID_VALUE error is generated if count is negative. When a sampler object is bound to a texture unit, its state supersedes that of the texture object bound to that texture unit. If the sampler name zero is bound to a texture unit, the currently bound texture’s sampler state becomes active. A single sampler object may be bound to multiple texture units simultaneously. A sampler object binding is effected with the command void BindSampler( uint unit, uint sampler ); with unit set to the zero-based index of the texture unit to which to bind the sampler and sampler set to the name of a sampler object returned from a previous call to GenSamplers. If the bind is successful no change is made to the state of the bound sampler object, and any previous binding to unit is broken. The currently bound sampler may be queried by calling GetIntegerv with pname set to SAMPLER_BINDING. When a sampler object is unbound from the texture unit (by binding another sampler object, or the sampler object named zero, to that texture unit) the modified state is again replaced with the sampler state associated with the texture object bound to that texture unit. 1 Note that unlike texture objects, the initial sampler object state for TEXTURE_MIN_FILTER and TEXTURE_WRAP_* are fixed, rather than dependent on the type of texture image.

OpenGL ES 3.2 (May 5, 2022)

8.2. SAMPLER OBJECTS

142

Errors An INVALID_VALUE error is generated if unit is greater than or equal to the value of MAX_COMBINED_TEXTURE_IMAGE_UNITS. An INVALID_OPERATION error is generated if sampler is not zero or a name returned from a previous call to GenSamplers, or if such a name has since been deleted with DeleteSamplers. The parameters represented by a sampler object are a subset of those described in section 8.10. Each parameter of a sampler object is set by calling void SamplerParameter{if}( uint sampler, enum pname, T param ); void SamplerParameter{if}v( uint sampler, enum pname, const T *params ); void SamplerParameterI{i ui}v( uint sampler, enum pname, const T *params ); sampler is the name of a sampler object previously reserved by a call to GenSamplers. pname is the name of a parameter to modify, and must be one of the sampler state names in table 21.12. In the scalar forms of the command, param is a value to which to set a single-valued parameter; in the vector forms, params is an array of parameters whose type depends on the parameter being set. Texture state listed in table 21.11 but not listed here and in the sampler state in table 21.12 is not part of the sampler state, and remains in the texture object. Data conversions are performed as specified in section 2.2.1, with these exceptions: • If the values for TEXTURE_BORDER_COLOR are specified with SamplerParameterIiv or SamplerParameterIuiv, they are unmodified and stored with an internal data type of integer. If specified with SamplerParameteriv, they are converted to floating-point using equation 2.2. Otherwise, the values are unmodified and stored as floating-point. Modifying a parameter of a sampler object affects all texture units to which that sampler object is bound. Calling TexParameter has no effect on the sampler object bound to the active texture unit. It will modify the parameters of the texture object bound to that unit.

OpenGL ES 3.2 (May 5, 2022)

8.3. SAMPLER OBJECT QUERIES

143

Errors An INVALID_OPERATION error is generated if sampler is not the name of a sampler object previously returned from a call to GenSamplers. An INVALID_ENUM error is generated if pname is not one of the sampler state names in table 21.12. An INVALID_ENUM error is generated if SamplerParameter{if} is called for a non-scalar parameter (pname TEXTURE_BORDER_COLOR). If the value of param is not an acceptable value for the parameter specified in pname, an error is generated as specified in the description of TexParameter*. Sampler objects are deleted by calling void DeleteSamplers( sizei count, const uint *samplers ); samplers contains count names of sampler objects to be deleted. After a sampler object is deleted, its name is again unused. If a sampler object that is currently bound to one or more texture units is deleted, it is as though BindSampler is called once for each texture unit to which the sampler is bound, with unit set to the texture unit and sampler set to zero. Unused names in samplers that have been marked as used for the purposes of GenSamplers are marked as unused again. Unused names in samplers are silently ignored, as is the reserved name zero. Errors An INVALID_VALUE error is generated if count is negative. The command boolean IsSampler( uint sampler ); may be called to determine whether sampler is the name of a sampler object. IsSampler will return TRUE if sampler is the name of a sampler object previously returned from a call to GenSamplers and FALSE otherwise. Zero is not the name of a sampler object.

8.3

Sampler Object Queries

The current values of the parameters of a sampler object may be queried by calling

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

144

void GetSamplerParameter{if}v( uint sampler, enum pname, T *params ); void GetSamplerParameterI{i ui}v( uint sampler, enum pname, T *params ); sampler is the name of the sampler object from which to retrieve parameters. pname is the name of the parameter to be queried, and must be one of the sampler state names in table 21.12. params is the address of an array into which the current value of the parameter will be placed. Querying TEXTURE_BORDER_COLOR with GetSamplerParameterIiv or GetSamplerParameterIuiv returns the border color values as signed integers or unsigned integers, respectively; otherwise the values are returned as described in section 2.2.2. If the border color is queried with a type that does not match the original type with which it was specified, the result is undefined. Errors An INVALID_OPERATION error is generated if sampler is not the name of a sampler object previously returned from a call to GenSamplers. An INVALID_ENUM error is generated if pname is not one of the sampler state names in table 21.12.

8.4

Pixel Rectangles

Rectangles of color, depth, and certain other values may be specified to the GL using TexImage*D (see section 8.5). Some of the parameters and operations governing the operation of these commands are shared by ReadPixels (used to obtain pixel values from the framebuffer); the discussion of ReadPixels, however, is deferred until chapter 9 after the framebuffer has been discussed in detail. Nevertheless, we note in this section when parameters and state pertaining to these commands also pertain to ReadPixels. A number of parameters control the encoding of pixels in buffer object or client memory (for reading and writing) and how pixels are processed before being placed in or after being read from the framebuffer (for reading, writing, and copying). These parameters are set with PixelStorei.

8.4.1

Pixel Storage Modes and Pixel Buffer Objects

Pixel storage modes affect the operation of TexImage*D, TexSubImage*D, and ReadPixels when one of these commands is issued. Pixel storage modes are set OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES Parameter Name UNPACK_ROW_LENGTH UNPACK_SKIP_ROWS UNPACK_SKIP_PIXELS UNPACK_ALIGNMENT UNPACK_IMAGE_HEIGHT UNPACK_SKIP_IMAGES

145 Type integer integer integer integer integer integer

Initial Value 0 0 0 4 0 0

Valid Range [0, ∞) [0, ∞) [0, ∞) 1,2,4,8 [0, ∞) [0, ∞)

Table 8.1: PixelStorei parameters pertaining to one or more of TexImage2D, TexImage3D, TexSubImage2D, and TexSubImage3D.

with void PixelStorei( enum pname, int param ); pname is a symbolic constant indicating a parameter to be set, and param is the value to set it to. Tables 8.1 and 16.1 summarize the pixel storage parameters, their types, their initial values, and their allowable ranges. Errors An INVALID_ENUM error is generated if pname is not one of the paramater names in table 8.1 or 16.1. An INVALID_VALUE error is generated if param is outside the given range for the corresponding pname in table 8.1 or 16.1. Data conversions are performed as specified in section 2.2.1. In addition to storing pixel data in client memory, pixel data may also be stored in buffer objects (described in section 6). The current pixel unpack and pack buffer objects are designated by the PIXEL_UNPACK_BUFFER and PIXEL_PACK_BUFFER targets respectively. Initially, zero is bound for the PIXEL_UNPACK_BUFFER, indicating that image specification commands such as TexImage*D source their pixels from client memory pointer parameters. However, if a non-zero buffer object is bound as the current pixel unpack buffer, then the pointer parameter is treated as an offset into the designated buffer object.

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

146

byte, short, or packed pixel component data stream Unpack

Convert to Float

Pixel Storage Operations

Convert L to RGB

Clamp to [0,1]

Final Conversion

RGBA pixel data out

Figure 8.1. Transfer of pixel rectangles to the GL. Output is RGBA pixels. Depth and stencil pixel paths are not shown.

8.4.2

Transfer of Pixel Rectangles

The process of transferring pixels encoded in buffer object or client memory is diagrammed in figure 8.1. We describe the stages of this process in the order in which they occur. Commands accepting or returning pixel rectangles take the following arguments (as well as additional arguments specific to their function): format is a symbolic constant indicating what the values in memory represent. internalformat is a symbolic constant indicating with what format and minimum precision the values should be stored by the GL. width and height are the width and height, respectively, of the pixel rectangle to be transferred. data refers to the data to be drawn. These data are represented with one of several GL data types, specified by type. The correspondence between the type token values and the GL data types they indicate is given in table 8.4. Not all combinations of format, type, and internalformat are valid. The combinations accepted by the GL are defined in tables 8.2 and 8.3. Some additional constraints on the combinations of format and type values that are accepted are OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

147

discussed below. Additional restrictions may be imposed by specific commands.

Format

Type

RGBA

UNSIGNED_BYTE

External Bytes per Pixel 4

RGBA RGBA RGBA RGBA RGBA RGBA RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGB

BYTE UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_5_5_5_1 UNSIGNED_INT_2_10_10_10_REV HALF_FLOAT FLOAT UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT UNSIGNED_INT_2_10_10_10_REV UNSIGNED_BYTE

4 2 2 4 8 16 4 4 8 8 16 16 4 3

RGB RGB RGB RGB RGB

BYTE UNSIGNED_SHORT_5_6_5 UNSIGNED_INT_10F_11F_11F_REV UNSIGNED_INT_5_9_9_9_REV HALF_FLOAT

3 2 4 4 6

RGB

FLOAT

12

Internal Format RGBA8, RGB5_A1, RGBA4, SRGB8_ALPHA8 RGBA8_SNORM RGBA4 RGB5_A1 RGB10_A2, RGB5_A1 RGBA16F RGBA32F, RGBA16F RGBA8UI RGBA8I RGBA16UI RGBA16I RGBA32UI RGBA32I RGB10_A2UI RGB8, RGB565, SRGB8 RGB8_SNORM RGB565 R11F_G11F_B10F RGB9_E5 RGB16F, R11F_G11F_B10F, RGB9_E5 RGB32F, RGB16F, R11F_G11F_B10F, RGB9_E5 RGB8UI RGB8I RGB16UI RGB16I

3 3 6 6 Valid combinations of format, type, and sized internalformat continued on next page

RGB_INTEGER RGB_INTEGER RGB_INTEGER RGB_INTEGER

UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

148

Valid combinations of format, type, and sized internalformat continued from previous page External Bytes Internal Format Type per Pixel Format RGB_INTEGER UNSIGNED_INT 12 RGB32UI RGB_INTEGER INT 12 RGB32I RG UNSIGNED_BYTE 2 RG8 RG BYTE 2 RG8_SNORM RG HALF_FLOAT 4 RG16F RG FLOAT 8 RG32F, RG16F RG_INTEGER UNSIGNED_BYTE 2 RG8UI RG_INTEGER BYTE 2 RG8I RG_INTEGER UNSIGNED_SHORT 4 RG16UI RG_INTEGER SHORT 4 RG16I RG_INTEGER UNSIGNED_INT 8 RG32UI RG_INTEGER INT 8 RG32I RED UNSIGNED_BYTE 1 R8 RED BYTE 1 R8_SNORM RED HALF_FLOAT 2 R16F RED FLOAT 4 R32F, R16F RED_INTEGER UNSIGNED_BYTE 1 R8UI RED_INTEGER BYTE 1 R8I RED_INTEGER UNSIGNED_SHORT 2 R16UI RED_INTEGER SHORT 2 R16I RED_INTEGER UNSIGNED_INT 4 R32UI RED_INTEGER INT 4 R32I DEPTH_COMPONENT UNSIGNED_SHORT 2 DEPTH_COMPONENT16 DEPTH_COMPONENT UNSIGNED_INT 4 DEPTH_COMPONENT24, DEPTH_COMPONENT16 DEPTH_COMPONENT32F DEPTH24_STENCIL8 DEPTH32F_STENCIL8 STENCIL_INDEX8

4 4 8 1 Table 8.2: Valid combinations of format, type, and sized internalformat.

DEPTH_COMPONENT DEPTH_STENCIL DEPTH_STENCIL STENCIL_INDEX

FLOAT UNSIGNED_INT_24_8 FLOAT_32_UNSIGNED_INT_24_8_REV UNSIGNED_BYTE

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

149

External Bytes Internal Format Type per Pixel Format RGBA UNSIGNED_BYTE 4 RGBA RGBA UNSIGNED_SHORT_4_4_4_4 2 RGBA RGBA UNSIGNED_SHORT_5_5_5_1 2 RGBA RGB UNSIGNED_BYTE 3 RGB RGB UNSIGNED_SHORT_5_6_5 2 RGB LUMINANCE_ALPHA UNSIGNED_BYTE 2 LUMINANCE_ALPHA LUMINANCE UNSIGNED_BYTE 1 LUMINANCE ALPHA UNSIGNED_BYTE 1 ALPHA Table 8.3: Valid combinations of format, type, and unsized internalformat.

8.4.2.1

Unpacking

Data are taken from the currently bound pixel unpack buffer or client memory as a sequence of signed or unsigned bytes (GL data types byte and ubyte), signed or unsigned short integers (GL data types short and ushort), signed or unsigned integers (GL data types int and uint), or floating-point values (GL data types half and float). These elements are grouped into sets of one, two, three, or four values, depending on the format, to form a group. Table 8.5 summarizes the format of groups obtained from memory; it also indicates those formats that yield indices and those that yield floating-point or integer components. If a pixel unpack buffer is bound (as indicated by a non-zero value of PIXEL_UNPACK_BUFFER_BINDING), data is an offset into the pixel unpack buffer and the pixels are unpacked from the buffer relative to this offset; otherwise, data is a pointer to client memory and the pixels are unpacked from client memory relative to the pointer. Errors An INVALID_OPERATION error is generated if a pixel unpack buffer object is bound and unpacking the pixel data according to the process described below would access memory beyond the size of the pixel unpack buffer’s

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

type Parameter Token Name UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT HALF_FLOAT FLOAT UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_5_5_5_1 UNSIGNED_INT_2_10_10_10_REV UNSIGNED_INT_24_8 UNSIGNED_INT_10F_11F_11F_REV UNSIGNED_INT_5_9_9_9_REV FLOAT_32_UNSIGNED_INT_24_8_REV

150

Corresponding GL Data Type ubyte byte ushort short uint int half float ushort ushort ushort uint uint uint uint n/a

Special Interpretation No No No No No No No No Yes Yes Yes Yes Yes Yes Yes Yes

Table 8.4: Pixel data type parameter values and the corresponding GL data types. Refer to table 2.2 for definitions of GL data types. Special interpretations are described in section 8.4.2.2.

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

Format Name DEPTH_COMPONENT DEPTH_STENCIL STENCIL_INDEX RED RG RGB RGBA LUMINANCE ALPHA LUMINANCE_ALPHA RED_INTEGER RG_INTEGER RGB_INTEGER RGBA_INTEGER

Element Meaning and Order Depth Depth and Stencil Stencil Index R R, G R, G, B R, G, B, A Luminance A Luminance, A iR iR, iG iR, iG, iB iR, iG, iB, iA

151

Target Buffer Depth Depth and Stencil Stencil Color Color Color Color Color Color Color Color Color Color Color

Table 8.5: Pixel data formats. The second column gives a description of and the number and order of elements in a group. Except for stencil, formats yield components. Components are floating-point unless prefixed with the letter ’i’, which indicates they are integer.

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

152

memory size. An INVALID_OPERATION error is generated if a pixel unpack buffer object is bound and data is not evenly divisible by the number of basic machine units needed to store in memory the corresponding GL data type from table 8.4 for the type parameter (or not evenly divisible by 4 for type FLOAT_32_UNSIGNED_INT_24_8_REV, which does not have a corresponding GL data type). The values of each GL data type are interpreted as they would be specified in the language of the client’s GL binding. The groups in memory are treated as being arranged in a rectangle. This rectangle consists of a series of rows, with the first element of the first group of the first row pointed to by data. If the value of UNPACK_ROW_LENGTH is zero, then the number of groups in a row is width; otherwise the number of groups is the value of UNPACK_ROW_LENGTH. If p indicates the location in memory of the first element of the first row, then the first element of the N th row is indicated by p + Nk where N is the row number (counting from zero) and k is defined as  nl  s ≥ a, k= a snl s<a s a

(8.1)

(8.2)

where n is the number of elements in a group, l is the number of groups in the row, a is the value of UNPACK_ALIGNMENT, and s is the size, in units of GL ubytes, of an element. If the number of bits per element is not 1, 2, 4, or 8 times the number of bits in a GL ubyte, then k = nl for all values of a. There is a mechanism for selecting a sub-rectangle of groups from a larger containing rectangle. This mechanism relies on three integer parameters: UNPACK_ROW_LENGTH, UNPACK_SKIP_ROWS, and UNPACK_SKIP_PIXELS. Before obtaining the first group from memory, the data pointer is advanced by (UNPACK_SKIP_PIXELS)n + (UNPACK_SKIP_ROWS)k elements. Then width groups are obtained from contiguous elements in memory (without advancing the pointer), after which the pointer is advanced by k elements. height sets of width groups of values are obtained this way. See figure 8.2. 8.4.2.2

Special Interpretations

A type matching one of the types in table 8.6 is a special case in which all the components of each group are packed into a single unsigned byte, unsigned short,

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

153

ROW_LENGTH

subimage SKIP_PIXELS

SKIP_ROWS

Figure 8.2. Selecting a subimage from an image. The indicated parameter names are prefixed by UNPACK_ TexImage* and by PACK_ for ReadPixels.

or unsigned int, depending on the type. If type is FLOAT_32_UNSIGNED_INT_24_8_REV, the components of each group are contained within two 32-bit words; the first word contains the float component, and the second word contains a packed 24-bit unused field, followed by an 8-bit component. The number of components per packed pixel is fixed by the type, and must match the number of components per group indicated by the format parameter, as listed in table 8.6. An INVALID_OPERATION error is generated by any command processing pixel rectangles if a mismatch occurs. Bitfield locations of the first, second, third, and fourth components of each packed pixel type are illustrated in figures 8.3- 8.5. Each bitfield is interpreted as an unsigned integer value. If the base GL type is supported with more than the minimum precision (e.g. a 9-bit byte) the packed components are right-justified in the pixel. Components are normally packed with the first component in the most significant bits of the bitfield, and successive component occupying progressively less significant locations. Types whose token names end with _REV reverse the component packing order from least to most significant locations. In all cases, the most significant bit of each component is packed in the most significant bit location of its location in the bitfield.

OpenGL ES 3.2 (May 5, 2022)

8.4. PIXEL RECTANGLES

154

type Parameter Token Name

GL Data Type ushort ushort ushort uint uint uint uint n/a

UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_5_5_5_1 UNSIGNED_INT_2_10_10_10_REV UNSIGNED_INT_24_8 UNSIGNED_INT_10F_11F_11F_REV UNSIGNED_INT_5_9_9_9_REV FLOAT_32_UNSIGNED_INT_24_8_REV

Number of Components 3 4 4 4 2 3 4 2

Matching Pixel Formats RGB RGBA RGBA RGBA, RGBA_INTEGER DEPTH_STENCIL RGB RGB DEPTH_STENCIL

Table 8.6: Packed pixel formats.

15

14

13

12

11

10

9

7

8

1st Component

6

5

3

4

2nd

2

1

0

1

0

1

0

3rd

UNSIGNED_SHORT_5_6_5 15

14

13

12

11

1st Component

10

9

8

7

6

2nd

5

3

4

2

3rd

4th

UNSIGNED_SHORT_4_4_4_4 15

14

13

12

1st Component

11

10

9

8

7

6

5

3

4

2nd

3rd

UNSIGNED_SHORT_5_5_5_1

Figure 8.3: UNSIGNED_SHORT formats

OpenGL ES 3.2 (May 5, 2022)

2

4th

8.4. PIXEL RECTANGLES

155

31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

3rd

4th

7

8

6

5

4

3

2

1

0

1st Component

2nd

UNSIGNED_INT_2_10_10_10_REV 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

7

8

6

5

1st Component

4

3

2

1

0

2

1

0

1

0

2nd

UNSIGNED_INT_24_8 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

3rd

7

8

6

5

4

3

1st Component

2nd

UNSIGNED_INT_10F_11F_11F_REV 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

4th

3rd

7

8

6

5

4

3

2

1st Component

2nd

UNSIGNED_INT_5_9_9_9_REV

Figure 8.4: UNSIGNED_INT formats

31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

8

7

6

5

4

3

1st Component 2nd FLOAT_32_UNSIGNED_INT_24_8_REV

Figure 8.5: FLOAT_UNSIGNED_INT formats

OpenGL ES 3.2 (May 5, 2022)

2

1

0

8.4. PIXEL RECTANGLES Format RGB RGBA DEPTH_STENCIL

156

First Component red red depth

Second Component green green stencil

Third Component blue blue

Fourth Component alpha

Table 8.7: Packed pixel field assignments.

The assignment of component to fields in the packed pixel is as described in table 8.7. The above discussions of row length and image extraction are valid for packed pixels, if “group” is substituted for “component” and the number of components per group is understood to be one. A type of UNSIGNED_INT_10F_11F_11F_REV and format of RGB is a special case in which the data are a series of GL uint values. Each uint value specifies 3 packed components as shown in figure 8.4. The 1st, 2nd, and 3rd components are called fred (11 bits), fgreen (11 bits), and fblue (10 bits) respectively. fred and fgreen are treated as unsigned 11-bit floating-point values and converted to floating-point red and green components respectively as described in section 2.3.4.3. fblue is treated as an unsigned 10-bit floating-point value and converted to a floating-point blue component as described in section 2.3.4.4. A type of UNSIGNED_INT_5_9_9_9_REV and format of RGB is a special case in which the data are a series of GL uint values. Each uint value specifies 4 packed components as shown in figure 8.4. The 1st, 2nd, 3rd, and 4th components are called pred , pgreen , pblue , and pexp respectively and are treated as unsigned integers. These are then used to compute floating-point RGB components (ignoring the “Conversion to floating-point” section below in this case) as follows: red = pred 2pexp −B−N green = pgreen 2pexp −B−N blue = pblue 2pexp −B−N where B = 15 (the exponent bias) and N = 9 (the number of mantissa bits). 8.4.2.3

Conversion to floating-point

This step applies only to groups of floating-point components. It is not performed on indices or integer components. For groups containing both components and OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

157

indices, such as DEPTH_STENCIL, the indices are not converted. Each element in a group is converted to a floating-point value. For unsigned or signed normalized fixed-point elements, equations 2.1 or 2.2, respectively, are used. 8.4.2.4

Conversion to RGB

This step is applied only if the format is LUMINANCE or LUMINANCE_ALPHA. If the format is LUMINANCE, then each group of one element is converted to a group of R, G, and B (three) elements by copying the original single element into each of the three new elements. If the format is LUMINANCE_ALPHA, then each group of two elements is converted to a group of R, G, B, and A (four) elements by copying the first original element into each of the first three new elements and copying the second original element to the A (fourth) new element. 8.4.2.5

Final Expansion to RGBA

This step is performed only for non-depth component groups. Each group is converted to a group of 4 elements as follows: if a group does not contain an A element, then A is added and set to one for integer components or 1.0 for floating-point components. If any of R, G, or B is missing from the group, each missing element is added and assigned a value of 0 for integer components or 0.0 for floating-point components.

8.5

Texture Image Specification

The command void TexImage3D( enum target, int level, int internalformat, sizei width, sizei height, sizei depth, int border, enum format, enum type, const void *data ); is used to specify a three-dimensional texture image. target must be one of TEXTURE_3D for a three-dimensional texture, TEXTURE_2D_ARRAY for a twodimensional array texture, or TEXTURE_CUBE_MAP_ARRAY for a cube map array texture. format, type, and data specify the format of the image data, the type of those data, and a reference to the image data in the currently bound pixel unpack buffer or client memory, as described in section 8.4.2.

OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

158

The groups in memory are treated as being arranged in a sequence of adjacent rectangles. Each rectangle is a two-dimensional image, whose size and organization are specified by the width and height parameters to TexImage3D. The values of UNPACK_ROW_LENGTH and UNPACK_ALIGNMENT control the row-to-row spacing in these images as described in section 8.4.2. If the value of the integer parameter UNPACK_IMAGE_HEIGHT is not positive, then the number of rows in each two-dimensional image is height; otherwise the number of rows is UNPACK_IMAGE_HEIGHT. Each two-dimensional image comprises an integral number of rows, and is exactly adjacent to its neighbor images. The mechanism for selecting a sub-volume of a three-dimensional image relies on the integer parameter UNPACK_SKIP_IMAGES. If UNPACK_SKIP_IMAGES is positive, the pointer is advanced by UNPACK_SKIP_IMAGES times the number of elements in one two-dimensional image before obtaining the first group from memory. Then depth two-dimensional images are processed, each having a subimage extracted as described in section 8.4.2. The selected groups are transferred to the GL as described in section 8.4.2 and then clamped to the representable range of the internal format as follows: • If the internalformat of the texture is signed or unsigned integer, components are clamped to [−2n−1 , 2n−1 − 1] or [0, 2n − 1], respectively, where n is the number of bits per component. • For color component groups, if the internalformat of the texture is signed or unsigned normalized fixed-point: – If the type of the data is a floating-point type (as defined in table 8.4), it is clamped to [−1, 1] or [0, 1], respectively. – Otherwise, it is clamped to to [−2n−1 , 2n−1 − 1] or [0, 2n − 1], respectively, where n is the number of bits in the normalized representation. • For depth component groups, the depth value is clamped to [0, 1]. • Otherwise, values are not modified. Components are then selected from the resulting R, G, B, A, depth, or stencil values to obtain a texture with the base internal format specified by (or derived from) internalformat. Table 8.8 summarizes the mapping of R, G, B, A, depth, or stencil values to texture components, as a function of the base internal format of the texture image. Textures with a base internal format of DEPTH_COMPONENT, DEPTH_STENCIL or STENCIL_INDEX are supported by texture image specification commands only if target is TEXTURE_2D, TEXTURE_2D_MULTISAMPLE, OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION Base Internal Format DEPTH_COMPONENT DEPTH_STENCIL LUMINANCE ALPHA LUMINANCE_ALPHA RED RG RGB RGBA STENCIL_INDEX

RGBA, Depth, and Stencil Values Depth Depth,Stencil R A R,A R R,G R,G,B R,G,B,A Stencil

159 Internal Components D D,S L A L,A R R,G R,G,B R,G,B,A S

Table 8.8: Conversion from RGBA, depth, and stencil pixel components to internal texture components. Texture components L, R, G, B, and A are converted back to RGBA colors during filtering as shown in table 14.1.

TEXTURE_2D_ARRAY, TEXTURE_2D_MULTISAMPLE_ARRAY, TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY. Using these formats in conjunction with any other target will result in an INVALID_OPERATION error.

The internal component resolution is the number of bits allocated to each value in a texture image. If internalformat is specified as a base internal format, the GL stores the resulting texture with internal component resolutions of its own choosing. If internalformat is a sized internal format, the effective internal format is the specified sized internal format. Otherwise, if internalformat is a base internal format, the effective internal format is a sized internal format that is derived from the format and type for internal use by the GL. Table 8.9 specifies the mapping of format and type to effective internal formats. The effective internal format is used by the GL for purposes such as texture completeness or type checks for CopyTex* and TexSubImage* commands. In these cases, the GL is required to operate as if the effective internal format was used as the internalformat when specifying the texture data. Note that unless specified elsewhere, the effective internal format values described in table 8.9 are not legal for an application to pass directly to the GL. If a sized internal format is specified, the mapping of the R, G, B, A, depth, and stencil values to texture components is equivalent to the mapping of the corresponding base internal format’s components, as specified in table 8.8; the type (unsigned int, float, etc.) is assigned the same type specified by internalformat; OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

160

Format

Type

Effective Internal Format

RGBA RGBA RGBA RGB RGB LUMINANCE_ALPHA LUMINANCE ALPHA

UNSIGNED_BYTE UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_5_5_5_1 UNSIGNED_BYTE UNSIGNED_SHORT_5_6_5 UNSIGNED_BYTE UNSIGNED_BYTE UNSIGNED_BYTE

RGBA8 RGBA4 RGB5_A1 RGB8 RGB565

Luminance8Alpha8 Luminance8 Alpha8

Table 8.9: Effective internal format corresponding to external format and type. Formats in italics do not correspond to GL constants.

and the memory allocation per texture component is assigned by the GL to match or exceed the allocations listed in tables 8.10- 8.11.

8.5.1

Required Texture Formats

Implementations are required to support the following sized internal formats. Requesting one of these sized internal formats for any texture type will allocate at least the internal component sizes, and exactly the component types shown for that format in tables 8.10- 8.11: • Color formats which are checked in the “Req. tex.” column of table 8.10. • All of the specific compressed texture formats in table 8.17. • Depth, depth+stencil, and stencil formats which are checked in the “Req. format” column of table 8.11.

8.5.2

Encoding of Special Internal Formats

If internalformat is R11F_G11F_B10F, the red, green, and blue bits are converted to unsigned 11-bit, unsigned 11-bit, and unsigned 10-bit floating-point values as described in sections 2.3.4.3 and 2.3.4.4. If internalformat is RGB9_E5, the red, green, and blue bits are converted to a shared exponent format according to the following procedure:

OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

161

Components red, green, and blue are first clamped (in the process, mapping NaN to zero) as follows: redc = max(0, min(sharedexpmax , red)) greenc = max(0, min(sharedexpmax , green)) bluec = max(0, min(sharedexpmax , blue)) where

(2N − 1) Emax −B 2 . 2N N is the number of mantissa bits per component (9), B is the exponent bias (15), and Emax is the maximum allowed biased exponent value (31). The largest clamped component, maxc , is determined: sharedexpmax =

maxc = max(redc , greenc , bluec ) A preliminary shared exponent expp is computed: expp = max(−B − 1, blog2 (maxc )c) + 1 + B A refined shared exponent exps is computed: k j max c maxs = expp −B−N + 0.5 2 ( expp , 0 ≤ maxs < 2N exps = expp + 1, maxs = 2N Finally, three integer values in the range 0 to 2N − 1 are computed:  reds =

redc



+ 0.5 2exps −B−N j green k c greens = exps −B−N + 0.5 2  bluec blues = exps −B−N + 0.5 2 The resulting reds , greens , blues , and exps are stored in the red, green, blue, and shared bits respectively of the texture image. An implementation accepting pixel data of type UNSIGNED_INT_5_9_9_9_REV with format RGB is allowed to store the components “as is”.

OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION Sized Internal Format

Base Internal Format

Bits/component S are shared bits G B A

162 CR

TF

R S R8 RED 8   R8_SNORM RED s8  RG8 RG 8 8   RG8_SNORM RG s8 s8  RGB8 RGB 8 8 8   RGB8_SNORM RGB s8 s8 s8  RGB565 RGB 5 6 5   RGBA4 RGBA 4 4 4 4   RGB5_A1 RGBA 5 5 5 1   RGBA8 RGBA 8 8 8 8   RGBA8_SNORM RGBA s8 s8 s8 s8  RGB10_A2 RGBA 10 10 10 2   RGB10_A2UI RGBA ui10 ui10 ui10 ui2  SRGB8 RGB 8 8 8  SRGB8_ALPHA8 RGBA 8 8 8 8   R16F RED f16   RG16F RG f16 f16   RGB16F RGB f16 f16 f16  RGBA16F RGBA f16 f16 f16 f16   R32F RED f32  RG32F RG f32 f32  RGB32F RGB f32 f32 f32 RGBA32F RGBA f32 f32 f32 f32  R11F_G11F_B10F RGB f11 f11 f10   RGB9_E5 RGB 9 9 9 5  R8I RED i8  R8UI RED ui8  R16I RED i16  R16UI RED ui16  R32I RED i32  R32UI RED ui32  RG8I RG i8 i8  RG8UI RG ui8 ui8  RG16I RG i16 i16  Sized internal color formats continued on next page

OpenGL ES 3.2 (May 5, 2022)

Req. rend.

Req. tex.



                                 

                        

8.5. TEXTURE IMAGE SPECIFICATION

163

Sized internal color formats continued from previous page Sized Base Bits/component CR TF Req. Internal Internal S are shared bits rend. Format Format R G B A S RG16UI RG ui16 ui16   RG32I RG i32 i32   RG32UI RG ui32 ui32   RGB8I RGB i8 i8 i8 RGB8UI RGB ui8 ui8 ui8 RGB16I RGB i16 i16 i16 RGB16UI RGB ui16 ui16 ui16 RGB32I RGB i32 i32 i32 RGB32UI RGB ui32 ui32 ui32 RGBA8I RGBA i8 i8 i8 i8   RGBA8UI RGBA ui8 ui8 ui8 ui8   RGBA16I RGBA i16 i16 i16 i16   RGBA16UI RGBA ui16 ui16 ui16 ui16   RGBA32I RGBA i32 i32 i32 i32   RGBA32UI RGBA ui32 ui32 ui32 ui32   Table 8.10: Correspondence of sized internal color formats to base internal formats, internal data type, minimum component resolutions, and use cases for each sized internal format. The component resolution prefix indicates the internal data type: f is floating point, i is signed integer, ui is unsigned integer, s is signed normalized fixed-point, and no prefix is unsigned normalized fixed-point. The “CR” (color-renderable), “TF” (texture-filterable), “Req. rend.” and “Req. tex.” columns are described in sections 9.4, 8.17, 9.2.5, and 8.5.1, respectively.

A GL implementation may vary its allocation of internal component resolution based on any TexImage3D or TexImage2D (see below) parameter (except target), but the allocation must not be a function of any other state and cannot be changed once they are established. Allocations must be invariant; the same allocation must be chosen each time a texture image is specified with the same parameter values.

OpenGL ES 3.2 (May 5, 2022)

Req. tex.               

8.5. TEXTURE IMAGE SPECIFICATION Sized Internal Format

Base Internal Format

DEPTH_COMPONENT16 DEPTH_COMPONENT24 DEPTH_COMPONENT32F DEPTH24_STENCIL8 DEPTH32F_STENCIL8 STENCIL_INDEX8

DEPTH_COMPONENT DEPTH_COMPONENT DEPTH_COMPONENT DEPTH_STENCIL DEPTH_STENCIL STENCIL_INDEX

164 D bits 16 24 f32 24 f32

S bits

ui8 ui8 ui8

Req. format      

Table 8.11: Correspondence of sized internal depth and stencil formats to base internal formats, internal data type, and minimum component resolutions for each sized internal format. The component resolution prefix indicates the internal data type: f is floating point, ui is unsigned integer, and no prefix is fixed-point. The “Req. format” column is described in section 8.5.1.

8.5.3

Texture Image Structure

The image itself (referred to by data) is a sequence of groups of values. The first group is the lower left back corner of the texture image. Subsequent groups fill out rows of width width from left to right; height rows are stacked from bottom to top forming a single two-dimensional image slice; and depth slices are stacked from back to front. When the final R, G, B, and A components have been computed for a group, they are assigned to components of a texel as described by table 8.8. Counting from zero, each resulting N th texel is assigned internal integer coordinates (i, j, k), where i = (N mod width)   N j=( mod height) width   N k=( mod depth) width × height Thus the last two-dimensional image slice of the three-dimensional image is indexed with the highest value of k. When target is TEXTURE_CUBE_MAP_ARRAY. specifying a cube map array texture, k refers to a layer-face. The layer is given by   k layer = , 6 OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

165

and the face is given by f ace = k mod 6. The face number corresponds to the cube map faces as shown in table 8.24. If the internal data type of the texture image is signed or unsigned normalized fixed-point, each color component is converted using equation 2.4 or 2.3, respectively. If the internal type is floating-point or integer, components are clamped to the representable range of the corresponding internal component, but are not converted. The level argument to TexImage3D is an integer level-of-detail number. Levels of detail are discussed in section 8.14.3. The main texture image has a level of detail number of zero. The maximum allowable size, in any relevant dimension, of a texture image is an implementation-dependent function of the texture target, the level-of-detail, and the internal format of the image. For most texture types, it must satisfy the relationship maxsize ≥ 2k−level

(8.3)

for images of level-of-detail (level) 0 through k, where k is a texture targetdependent maximum level of detail. The maximum size may be zero for any images where level > k. The maximum allowable width, height, or depth of a texure image for a threedimensional texture is determined by equation 8.3, where k is log2 of the value of MAX_3D_TEXTURE_SIZE. In a similar fashion, the maximum allowable width or height of a texture image for a two-dimensional, two-dimensional array, two-dimensional multisample, or two-dimensional multisample array texture is determined by equation 8.3, where k is log2 of the value of MAX_TEXTURE_SIZE. The maximum allowable width or height of a cube map or cube map array texture image must be the same, and is determined by equation 8.3, where k is log2 of the value of MAX_CUBE_MAP_TEXTURE_SIZE. The maximum number of layers (depth) for two-dimensional array textures, or or the maximum number of layer-faces for cube map array textures (depth), must each be at least the value of the implementation-dependent constant MAX_RECTANGLE_TEXTURE_SIZE. As described in section 8.17, these implementation-dependent limits may be configured to reject textures at level one or greater unless a mipmap complete set of texture images consistent with the specified sizes can be supported.

OpenGL ES 3.2 (May 5, 2022)

8.5. TEXTURE IMAGE SPECIFICATION

166

Regardless of the values of these implementation-dependent constants, an implementation may not succeed in creating a texture of the maximum sizes due to resource limits, resulting in memory exhaustion. Errors An INVALID_ENUM error is generated if target is not one of the valid targets listed for each TexImage*D command. An INVALID_VALUE error is generated if level is negative. An INVALID_VALUE error is generated if width, height, or depth (if each argument is present) exceed the corresponding target-dependent maximum size, as described above. An INVALID_VALUE error is generated if width, height, or depth are negative. An INVALID_VALUE error is generated by TexImage3D if target is TEXTURE_CUBE_MAP_ARRAY and width and height are not equal, or if depth is not a multiple of six, indicating 6N layer-faces in the cube map array. An INVALID_VALUE error is generated if target is one of the cube map face targets from table 8.20, and width and height are not equal. An INVALID_VALUE error is generated if border is not zero. An INVALID_VALUE error is generated if internalformat is not one of the valid formats in tables 8.2 or 8.3. An INVALID_OPERATION error is generated if format, type, and internalformat are each valid, but the combination of values specified is not listed as a valid combination in tables 8.2 or 8.3. An INVALID_OPERATION error is generated if a pixel unpack buffer object is bound and storing texture data would access memory beyond the end of the pixel unpack buffer. The command void TexImage2D( enum target, int level, int internalformat, sizei width, sizei height, int border, enum format, enum type, const void *data ); is used to specify a two-dimensional texture image.

target must be one of

TEXTURE_2D for a two-dimensional texture, or one of the cube map face targets

from table 8.20 for a cube map texture. The other parameters match the corresponding parameters of TexImage3D. For the purposes of decoding the texture image, TexImage2D is equivalent to calling TexImage3D with corresponding arguments and depth of 1, except that OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS UNPACK_SKIP_IMAGES is ignored.

A two-dimensional texture consists of a single two-dimensional texture image. A cube map texture is a set of six two-dimensional texture images. The six cube map texture face targets from table 8.20 form a single cube map texture. These targets each update the corresponding cube map face two-dimensional texture image. Note that the cube map face targets are used when specifying, updating, or querying one of a cube map’s six two-dimensional images, but when binding to a cube map texture object (that is, when the cube map is accessed as a whole as opposed to a particular two-dimensional image), the TEXTURE_CUBE_MAP target is specified. The image indicated to the GL by the image pointer is decoded and copied into the GL’s internal memory. We shall refer to the decoded image as the texture image. A three-dimensional texture image has width, height, and depth wt , ht , and dt . A two-dimensional texture image has depth dt = 1, with height ht and width wt as above. An element (i, j, k) of the texture image is called a texel (for a two-dimensional texture, k is irrelevant). The texture value used in texturing a fragment is determined by sampling the texture in a shader, but may not correspond to any actual texel. See figure 8.6. If target is TEXTURE_CUBE_MAP_ARRAY, the texture value is determined by (s, t, r, q) coordinates where s, t, and r are defined to be the same as for TEXTURE_CUBE_MAP and q is defined as the index of a specific cube map in the cube map array. If the data argument of TexImage2D or TexImage3D is NULL, and the pixel unpack buffer object is zero, a two-or three-dimensional texture image is created with the specified target, level, internalformat, border, width, height, and depth, but with unspecified image contents. In this case no pixel values are accessed in client memory, and no pixel processing is performed. Errors are generated, however, exactly as though the data pointer were valid. Otherwise if the pixel unpack buffer object is non-zero, the data argument is treatedly normally to refer to the beginning of the pixel unpack buffer object’s data.

8.6

Alternate Texture Image Specification Commands

Two-dimensional texture images may also be specified using image data taken directly from the framebuffer, and rectangular subregions of existing texture images may be respecified. The command void CopyTexImage2D( enum target, int level, OpenGL ES 3.2 (May 5, 2022)

167

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

Figure 8.6. A texture image and the coordinates used to access it. This is a twodimensional texture with width 8 and height 4. α and β, values used in blending adjacent texels to obtain a texture value are also shown.

OpenGL ES 3.2 (May 5, 2022)

168

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS Read Buffer Format Normalized Fixed-point 10-bit Normalized Fixed-point Signed Integer Unsigned Integer

169

format

type

RGBA RGBA RGBA_INTEGER RGBA_INTEGER

UNSIGNED_BYTE UNSIGNED_INT_2_10_10_10_REV INT UNSIGNED_INT

Table 8.12: ReadPixels format and type used during CopyTex*.

enum internalformat, int x, int y, sizei width, sizei height, int border ); defines a two-dimensional texture image in exactly the manner of TexImage2D, except that the image data are taken from the framebuffer rather than from client memory. target must be one of TEXTURE_2D or one of the cube map face targets from table 8.20. x, y, width, and height correspond precisely to the corresponding arguments to ReadPixels (refer to section 16.1); they specify the image’s width and height, and the lower left (x, y) coordinates of the framebuffer region to be copied. The image is taken from the current color buffer exactly as if these arguments were passed to ReadPixels with arguments format and type set according to table 8.12, though the final copy into memory (see section 16.1.6) is skipped. stopping after conversion of RGBA values. Subsequent processing is identical to that described for TexImage2D, beginning with clamping of the R, G, B, and A values from the resulting pixel groups. Parameters level, internalformat, and border are specified using the same values, with the same meanings, as the equivalent arguments of TexImage2D. internalformat is further constrained such that color buffer components can be dropped during the conversion to internalformat, but new components cannot be added. For example, an RGB color buffer can be used to create LUMINANCE or RGB textures, but not ALPHA, LUMINANCE_ALPHA, or RGBA textures. Table 8.13 summarizes the valid framebuffer and texture base internal format combinations. The constraints on width, height, and border are exactly those for the corresponding arguments of TexImage2D. If internalformat is sized, the internal format of the new texture image is internalformat, and this is also the new image’s effective internal format. If internalformat is unsized, the internal format of the new image is determined by the following rules, applied in order. If an effective internal format exists that has 1. the same component sizes as, OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

Framebuffer R RG RGB RGBA D DS S

A



L    

LA



R    

Texture Format RG RGB RGBA   

 

D

DS

170

S



Table 8.13: Valid CopyTexImage source framebuffer/destination texture base internal format combinations.

2. component sizes greater than or equal to, or 3. component sizes smaller than or equal to those of the source buffer’s effective internal format (for all matching components in internalformat), that format is chosen for the new texture image, and this is also the new image’s effective internal format. When matching formats that involve a luminance component, a luminance component is considered to match with a red component. If multiple possible matches exist in the same rule, the one with the closest component sizes is chosen. Note that the above rules disallow matches where some components sizes are smaller and others are larger (such as RGB10_A2). The effective internal format of the source buffer is determined with the following rules applied in order: • If the source buffer is a texture or renderbuffer that was created with a sized internal format then the effective internal format is the source buffer’s sized internal format. • If the source buffer is a texture that was created with an unsized base internal format, then the effective internal format is the source image’s effective internal format, as specified by table 8.9, which is determined from the format and type that were used when the source image was specified by TexImage*. • If the source buffer contains any floating point components, then the effective internal format is taken from the first (highest) row in table 8.14 for which the source buffer’s red, green, blue, and alpha component OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS Destination Internal Format any sized any sized any sized any sized any sized any sized any sized any sized

Source Red Size 1 ≤ R ≤ 16 1 ≤ R ≤ 16 16 < R 16 < R 1 ≤ R ≤ 16 1 ≤ R ≤ 16 16 < R 16 < R

Source Green Size G=0 1 ≤ G ≤ 16 G=0 16 < G 1 ≤ G ≤ 16 1 ≤ G ≤ 16 16 < G 16 < G

Source Blue Size B=0 B=0 B=0 B=0 1 ≤ B ≤ 16 1 ≤ B ≤ 16 16 < B 16 < B

171

Source Alpha Size A=0 A=0 A=0 A=0 A=0 1 ≤ A ≤ 16 A=0 16 < A

Table 8.14: Effective internal format corresponding to floating-point framebuffers.

sizes (the values of FRAMEBUFFER_RED_SIZE, FRAMEBUFFER_GREEN_SIZE, FRAMEBUFFER_BLUE_SIZE, and FRAMEBUFFER_ALPHA_SIZE respectively) are consistent with the rules in that row for R, G, B, and A respectively. • Otherwise the effective internal format is determined by the first (highest) row in table 8.15 or table 8.16 for which the Destination Internal Format column matches internalformat, and for which the source buffer’s red, green, blue, and alpha component sizes are consistent with the rules in that row for R, G, B, and A respectively. Table 8.15 is used if the framebuffer encoding (the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING) is LINEAR and table 8.16 is used if the framebuffer encoding is SRGB. In tables 8.14, 8.15, and 8.16, ”any sized” matches any specified sized internal format. ”N/A” means the source buffer’s component size is ignored. If the effective internal format of the source buffer does not match the effective internal format of the new image, and if there are no components which exist in the destination format which don’t exist in the source, this function may still complete successfully using the conversion functions defined in this specification. However, this behaviour is not guaranteed, and if a given conversion is not supported, an INVALID_OPERATION error will be generated as described below. Errors

OpenGL ES 3.2 (May 5, 2022)

Effective Internal Format R16F RG16F R32F RG32F RGB16F RGBA16F RGB32F RGBA32F

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

Destination Internal Format any sized any sized any sized any sized any sized any sized any sized any sized any sized ALPHA LUMINANCE LUMINANCE_ALPHA RGB RGB RGBA RGBA RGBA

172

Source Red Size R=0 1≤R≤8 1≤R≤8 1≤R≤5 5<R≤8 1≤R≤4 4<R≤5 4<R≤8 8 < R ≤ 10 N/A 1≤R≤8 1≤R≤8

Source Green Size G=0 G=0 1≤G≤8 1≤G≤6 6<G≤8 1≤G≤4 4<G≤5 4<G≤8 8 < G ≤ 10 N/A N/A N/A

Source Blue Size B=0 B=0 B=0 1≤B≤5 5<B≤8 1≤B≤4 4<B≤5 4<B≤8 8 < B ≤ 10 N/A N/A N/A

Source Alpha Size 1≤A≤8 A=0 A=0 A=0 A=0 1≤A≤4 A=1 1<A≤8 1<A≤2 1≤A≤8 N/A 1≤A≤8

Effective Internal Format Alpha8

Alpha8 Luminance8 Luminance8Alpha8

1≤R≤5 5<R≤8 1≤R≤4 4<R≤5 4<R≤8

1≤G≤6 6<G≤8 1≤G≤4 4<G≤5 4<G≤8

1≤B≤5 5<B≤8 1≤B≤4 4<B≤5 4<B≤8

N/A N/A 1≤A≤4 A=1 1<A≤8

RGB565 RGB8 RGBA4 RGB5_A1 RGBA8

R8 RG8 RGB565 RGB8 RGBA4 RGB5_A1 RGBA8 RGBA10_A2

Table 8.15: Effective internal format corresponding to destination internalformat and linear source buffer component sizes. Effective internal formats in italics do not correspond to GL constants.

Destination Internal Format any sized

Source Red Size 1≤R≤8

Source Green Size 1≤G≤8

Source Blue Size 1≤B≤8

Source Alpha Size 1≤A≤8

Effective Internal Format SRGB_ALPHA8

Table 8.16: Effective internal format corresponding to destination internalformat and sRGB source buffer component sizes.

OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS An INVALID_ENUM error is generated if target is not TEXTURE_2D or one of the cube map face targets from table 8.20. An INVALID_ENUM error is generated if an invalid value is specified for internalformat. An INVALID_VALUE error is generated if target is one of the six cube map two-dimensional image targets, and width and height are not equal. An INVALID_OPERATION error is generated under any of the following conditions: • • if the effective internal format of the source buffer does not match the effective internal format of the new image. • if the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the read buffer (see section 16.1.1) is LINEAR (see section 9.2.3) and internalformat is one of the sRGB formats in table 8.23 • if the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the read buffer is SRGB and internalformat is not one of the sRGB formats in table 8.23. An INVALID_VALUE error is generated if width or height is negative. An INVALID_VALUE error is generated if border is non-zero. An INVALID_OPERATION error is generated if the framebuffer and texture base internal format are not compatible, as defined in table 8.13. An INVALID_OPERATION error is generated if internalformat is unsized and no effective internal format exists which matches the rules described above. An INVALID_OPERATION error is generated if the component sizes of internalformat do not exactly match the corresponding component sizes of the source buffer’s effective internal format. An INVALID_OPERATION error is generated if there are no rows in tables 8.14, 8.15 or 8.16 which match internalformat and the source buffer component types and sizes. In this case, the source buffer does not have an effective internal format. To respecify only a rectangular subregion of the texture image of a texture object, use the commands void TexSubImage3D( enum target, int level, int xoffset, OpenGL ES 3.2 (May 5, 2022)

173

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

174

int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, enum type, const void *data ); void TexSubImage2D( enum target, int level, int xoffset, int yoffset, sizei width, sizei height, enum format, enum type, const void *data ); void CopyTexSubImage3D( enum target, int level, int xoffset, int yoffset, int zoffset, int x, int y, sizei width, sizei height ); void CopyTexSubImage2D( enum target, int level, int xoffset, int yoffset, int x, int y, sizei width, sizei height ); No change is made to the internalformat, width, height, depth, or border parameters of the specified texture image, nor is any change made to texel values outside the specified subregion. The target arguments of TexSubImage2D and CopyTexSubImage2D must be one of TEXTURE_2D or one of the cube map face targets from table 8.20, and the target arguments of TexSubImage3D and CopyTexSubImage3D must be TEXTURE_3D, TEXTURE_2D_ARRAY or TEXTURE_CUBE_MAP_ARRAY. The level parameter of each command specifies the level of the texture image that is modified. Errors An INVALID_VALUE error is generated if level is negative or greater than the log2 of the maximum texture width, height, or depth. TexSubImage3D arguments width, height, depth, format, and type match the corresponding arguments to TexImage3D, meaning that they accept the same values, and have the same meanings. Likewise, TexSubImage2D arguments width, height, format, and type match the corresponding arguments to TexImage2D. The data argument of TexSubImage3D and TexSubImage2D matches the corresponding argument of TexImage3D and TexImage2D, respectively, except that a NULL pointer does not represent unspecified image contents. CopyTexSubImage3D and CopyTexSubImage2D arguments x, y, width, and height match the corresponding arguments to CopyTexImage2D2 . Each of the TexSubImage commands interprets and processes pixel groups in exactly the manner of its TexImage counterpart, except that the assignment of R, G, B, A, depth, 2

Because the framebuffer is inherently two-dimensional, there is no CopyTexImage3D command.

OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

175

and stencil index pixel group values to the texture components is controlled by the internalformat of the texture image, not by an argument to the command. The same constraints and errors apply to the TexSubImage commands’ argument format and the internalformat of the texture image being respecified as apply to the format and internalformat arguments of its TexImage counterparts. It is implementationdependent whether the internal format used for error checking is the internalformat given when the texture was created, or the effective internal format of the texture. If it is the former, then texels will be converted to the effective internal format on upload. Arguments xoffset, yoffset, and zoffset of TexSubImage3D and CopyTexSubImage3D specify the lower left texel coordinates of a width-wide by heighthigh by depth-deep rectangular subregion of the texture image. For cube map array textures, zoffset is the first layer-face to update, and depth is the number of layerfaces to update. The depth argument associated with CopyTexSubImage3D is always 1, because framebuffer memory is two-dimensional - only a portion of a single (s, t) slice of a three-dimensional texture is replaced by CopyTexSubImage3D. Taking wt , ht , and dt to be the specified width, height, and depth of the texture image, and taking x, y, z, w, h, and d to be the xoffset, yoffset, zoffset, width, height, and depth argument values, any of the following relationships generates an INVALID_VALUE error: x<0 x + w > wt y<0 y + h > ht z<0 z + d > dt Counting from zero, the nth pixel group is assigned to the texel with internal integer coordinates [i, j, k], where i = x + (n mod w) n j = y + (b c mod h) w n c mod d k = z + (b width ∗ height

OpenGL ES 3.2 (May 5, 2022)

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

176

Arguments xoffset and yoffset of TexSubImage2D and CopyTexSubImage2D specify the lower left texel coordinates of a width-wide by height-high rectangular subregion of the texture image. Taking wt and ht to be the specified width and height of the image, and taking x, y, w, and h to be the xoffset, yoffset, width, and height argument values, any of the following relationships generates an INVALID_VALUE error: x<0 x + w > wt y<0 y + h > ht Counting from zero, the nth pixel group is assigned to the texel with internal integer coordinates [i, j], where i = x + (n mod w) n j = y + (b c mod h) w Errors An INVALID_FRAMEBUFFER_OPERATION error is generated by CopyTexSubImage3D, CopyTexImage2D, or CopyTexSubImage2D if the object bound to READ_FRAMEBUFFER_BINDING is not framebuffer complete (see section 9.4.2) An INVALID_OPERATION error is generated by CopyTexSubImage3D, CopyTexImage2D, or CopyTexSubImage2D if • the read buffer is NONE, or • the internalformat of the texture image being (re)specified is RGB9_E5, or • the value of READ_FRAMEBUFFER_BINDING is non-zero, and – the read buffer selects an attachment that has no image attached, or – the effective value of SAMPLE_BUFFERS for the read framebuffer (see section 9.2.3.1) is one.

OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

8.6.1

177

Texture Copying Feedback Loops

Calling CopyTexSubImage3D, CopyTexImage2D, or CopyTexSubImage2D will result in undefined behavior if the destination texture image level is also bound to to the selected read buffer (see section 16.1.1) of the read framebuffer. This situation is discussed in more detail in the description of feedback loops in section 9.3.2.

8.7

Compressed Texture Images

Texture images may also be specified or modified using image data already stored in a known compressed image format, including the formats defined in appendix C as well as any additional formats defined by extensions. The GL provides a mechanism to obtain token values for all compressed formats supported by the implementation. The number of specific compressed internal formats supported by the renderer can be obtained by querying the value of NUM_COMPRESSED_TEXTURE_FORMATS. The set of specific compressed internal formats supported by the renderer can be obtained by querying the value of COMPRESSED_TEXTURE_FORMATS. All implementations support at least the formats listed in table 8.17. The commands void CompressedTexImage2D( enum target, int level, enum internalformat, sizei width, sizei height, int border, sizei imageSize, const void *data ); void CompressedTexImage3D( enum target, int level, enum internalformat, sizei width, sizei height, sizei depth, int border, sizei imageSize, const void *data ); define two-and three-dimensional texture images, respectively, with incoming data stored in a compressed image format. The target, level, internalformat, width, height, depth, and border parameters have the same meaning as in TexImage2D and TexImage3D. data refers to compressed image data stored in the specific compressed image format corresponding to internalformat. If a pixel unpack buffer is bound (as indicated by a non-zero value of PIXEL_UNPACK_BUFFER_BINDING), data is an offset into the pixel unpack buffer and the compressed data is read from the buffer relative to this offset; otherwise, data is a pointer to client memory and the compressed data is read from client memory relative to the pointer. The compressed image will be decoded according to the specification defining the internalformat token. Compressed texture images are treated as an array of imageSize ubytes relative to data. OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

178

If the compressed image is not encoded according to the defined image format, the results of the call are undefined. All pixel storage modes are ignored when decoding a compressed texture image.

Compressed Internal Format

COMPRESSED_R11_EAC COMPRESSED_SIGNED_R11_EAC COMPRESSED_RG11_EAC COMPRESSED_SIGNED_RG11_EAC COMPRESSED_RGB8_ETC2 COMPRESSED_SRGB8_ETC2 COMPRESSED_RGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_RGBA8_ETC2_EAC COMPRESSED_SRGB8_ALPHA8_ETC2_EAC COMPRESSED_RGBA_ASTC_4x4 COMPRESSED_RGBA_ASTC_5x4 COMPRESSED_RGBA_ASTC_5x5 COMPRESSED_RGBA_ASTC_6x5 COMPRESSED_RGBA_ASTC_6x6 COMPRESSED_RGBA_ASTC_8x5 COMPRESSED_RGBA_ASTC_8x6 COMPRESSED_RGBA_ASTC_8x8 COMPRESSED_RGBA_ASTC_10x5 COMPRESSED_RGBA_ASTC_10x6 COMPRESSED_RGBA_ASTC_10x8 COMPRESSED_RGBA_ASTC_10x10 COMPRESSED_RGBA_ASTC_12x10 COMPRESSED_RGBA_ASTC_12x12 COMPRESSED_SRGB8_ALPHA8_ASTC_4x4

Base

Block

Border

3D

Cube Map

Internal Format

Type

Tex.

RED RED RG RG RGB RGB RGBA

Width x Height 4×4 4×4 4×4 4×4 4×4 4×4 4×4

unorm snorm unorm snorm unorm unorm unorm

Array Tex.       

RGBA

4×4

unorm



RGBA RGBA

4×4 4×4

unorm unorm

 

RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA RGBA

4×4 5×4 5×5 6×5 6×6 8×5 8×6 8×8 10 × 5 10 × 6 10 × 8 10 × 10 12 × 10 12 × 12 4×4

unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm unorm

              

(Continued on next page) OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

179

Commpressed internal formats (continued) Compressed Internal Format Base Block Border Internal Width x Type Format Height COMPRESSED_SRGB8_ALPHA8_RGBA 5×4 unorm ASTC_5x4 COMPRESSED_SRGB8_ALPHA8_ASTC_5x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x5 COMPRESSED_SRGB8_ALPHA8_ASTC_8x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x5 COMPRESSED_SRGB8_ALPHA8_ASTC_10x6 COMPRESSED_SRGB8_ALPHA8_ASTC_10x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x12

3D Tex.

Cube Map Array Tex. 

RGBA

5×5

unorm



RGBA

6×5

unorm



RGBA

6×6

unorm



RGBA

8×5

unorm



RGBA

8×6

unorm



RGBA

8×8

unorm



RGBA

10 × 5

unorm



RGBA

10 × 6

unorm



RGBA

10 × 8

unorm



RGBA

10 × 10

unorm



RGBA

12 × 10

unorm



RGBA

12 × 12

unorm



Table 8.17: Compressed internal formats. The formats are described in appendix C. The “Block Size” column specifies the compressed block size of the format. Modifying compressed images along aligned block boundaries is possible, as described in this section. The “Border Type” column determines how border colors are clamped, as described in section 8.14.2. The “3D Tex.” and “Cube Map Array Tex.” columns determine if 3D images composed of compressed 2D slices and cube map array textures, respectively, can be specified using CompressedTexImage3D. OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

180

Compressed internal formats may impose format-specific restrictions on the use of the compressed image specification calls or parameters. For example, the compressed image format might be supported only for 2D textures. Any such restrictions will be documented in the extension specification defining the compressed internal format, and will be invariant with respect to image contents. This means that if the GL accepts and stores a texture image in compressed form, CompressedTexImage2D or CompressedTexImage3D will accept any properly encoded compressed texture image of the same width, height, depth, compressed image size, and compressed internal format for storage at the same texture level. If internalformat is one of the specific compressed formats described in table 8.17, the compressed image data is stored using the corresponding texture image encoding (see appendix C). The corresponding texture compression algorithms supports only two-dimensional images. However, if the “3D Tex” column of table 8.17 is checked, CompressedTexImage3D will accept a three-dimensional image specified as an array of compressed data consisting of multiple rows of compressed blocks laid out as described in section 8.5. The width and height of each sub-image must be a multiple of the block size for the format, shown in the same table. Errors An INVALID_ENUM error is generated by CompressedTexImage2D if target is not TEXTURE_2D or one of the cube map face targets from table 8.20. An INVALID_VALUE error is generated by • CompressedTexImage2D if target is one of the cube map face targets from table 8.20, and • CompressedTexImage3D if target is TEXTURE_CUBE_MAP_ARRAY, and width and height are not equal. An INVALID_OPERATION error is generated by CompressedTexImage3D if internalformat is one of the the formats in table 8.17 and target is not TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or TEXTURE_3D. An INVALID_OPERATION error is generated by CompressedTexImage3D if target is TEXTURE_CUBE_MAP_ARRAY and the “Cube Map Array” column of table 8.17 is not checked, or if target is TEXTURE_3D and the “3D Tex.” column of table 8.17 is not checked. An INVALID_VALUE error is generated if border is non-zero.

OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

181

An INVALID_ENUM error is generated if internalformat is not a supported specific compressed internal format from table 8.17 or one of the additional formats defined by OpenGL ES extensions. An INVALID_VALUE error is generated if width, height, depth, or imageSize is negative. An INVALID_OPERATION error is generated if a pixel unpack buffer object is bound and data+imageSize is greater than the size of the pixel buffer. An INVALID_VALUE error is generated if the imageSize parameter is not consistent with the format, dimensions, and contents of the compressed image. An INVALID_OPERATION error is generated if any format-specific restrictions imposed by specific compressed internal formats are violated by the compressed image specification calls or parameters. If the data argument of CompressedTexImage2D or CompressedTexImage3D is NULL, and the pixel unpack buffer object is zero, a texture image with unspecified image contents is created, just as when a NULL pointer is passed to TexImage2D or TexImage3D. To respecify only a rectangular subregion of the texture image of a texture object, with incoming data stored in a specific compressed image format, use the commands void CompressedTexSubImage2D( enum target, int level, int xoffset, int yoffset, sizei width, sizei height, enum format, sizei imageSize, const void *data ); void CompressedTexSubImage3D( enum target, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, sizei imageSize, const void *data ); The target, level, xoffset, yoffset, zoffset, width, height, and depth parameters have the same meaning as in TexSubImage2D, and TexSubImage3D. data points to compressed image data stored in the compressed image format corresponding to format. The image pointed to by data and the imageSize parameter are interpreted as though they were provided to CompressedTexImage2D and CompressedTexImage3D. Any restrictions imposed by specific compressed internal formats will be invariant with respect to image contents, meaning that if the GL accepts and stores a texture image in compressed form, CompressedTexSubImage2D or CompressedTexSubImage3D will accept any properly encoded compressed texture OpenGL ES 3.2 (May 5, 2022)

8.7. COMPRESSED TEXTURE IMAGES

182

image of the same width, height, compressed image size, and compressed internal format for storage at the same texture level. If the internal format of the image being modified is one of the specific compressed formats described in table 8.17, the texture is stored using the corresponding texture image encoding (see appendix C). Since these specific compressed formats are easily edited along texel block boundaries, the limitations on subimage location and size are relaxed for CompressedTexSubImage2D and CompressedTexSubImage3D. The block width and height varies for different formats, as described in table 8.17. The contents of any block of texels of a compressed texture image in these specific compressed formats that does not intersect the area being modified are preserved during CompressedTexSubImage* calls. Errors An INVALID_ENUM error is generated by CompressedTexSubImage2D if target is not TEXTURE_2D or one of the cube map face targets from table 8.20. An INVALID_OPERATION error is generated by CompressedTexSubImage3D if format is one of the formats in table 8.17 and target is not TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or TEXTURE_3D. An INVALID_OPERATION error is generated by CompressedTexSubImage3D if format is TEXTURE_CUBE_MAP_ARRAY and the “Cube Map Array” column of table 8.17 is not checked, or if format is TEXTURE_3D and the “3D Tex.” column of table 8.17 is not checked. An INVALID_OPERATION error is generated if format does not match the internal format of the texture image being modified, since these commands do not provide for image format conversion. An INVALID_VALUE error is generated if width, height, depth, or imageSize is negative. An INVALID_VALUE error is generated if imageSize is not consistent with the format, dimensions, and contents of the compressed image (too little or too much data), An INVALID_OPERATION error is generated if any format-specific restrictions are violated, as with CompressedTexImage calls. Any such restrictions will be documented in the specification defining the compressed internal format. An INVALID_OPERATION error is generated if xoffset, yoffset, or zoffset are not equal to zero, or if width, height, and depth do not match the corre-

OpenGL ES 3.2 (May 5, 2022)

8.8. MULTISAMPLE TEXTURES

183

sponding dimensions of the texture level. The contents of any texel outside the region modified by the call are undefined. These restrictions may be relaxed for specific compressed internal formats whose images are easily modified. An INVALID_OPERATION error is generated if format is one of the formats in table 8.17 and any of the following conditions occurs. The block width and height refer to the values in the corresponding column of the table. • width is not a multiple of the format’s block width, and width + xoffset is not equal to the value of TEXTURE_WIDTH. • height is not a multiple of the format’s block height, and height +yoffset is not equal to the value of TEXTURE_HEIGHT. • xoffset or yoffset is not a multiple of the block width or height, respectively.

8.8

Multisample Textures

In addition to the texture types described in previous sections, two additional types of texture are supported. Multisample textures are similar to two-dimensional or two-dimensional array textures, except that they contain multiple samples per texel. Multisample textures do not have multiple image levels, and are immutable. The commands void TexStorage2DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, boolean fixedsamplelocations ); void TexStorage3DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, sizei depth, boolean fixedsamplelocations ); establishes the data storage, format, dimensions, and number of samples of a multisample texture’s image. For TexStorage2DMultisample target must be TEXTURE_2D_MULTISAMPLE, and for TexStorage3DMultisample target must be TEXTURE_2D_MULTISAMPLE_ARRAY. width and height are the dimensions in texels of the texture, and depth is the number of array layers. samples represents a request for a desired minimum number of samples. Since different implementations may support different sample counts for multisampled textures, the actual number of samples allocated for the texture image is implementation-dependent. However, the resulting value for TEXTURE_SAMPLES

OpenGL ES 3.2 (May 5, 2022)

8.8. MULTISAMPLE TEXTURES

184

is guaranteed to be greater than or equal to samples and no more than the next larger sample count supported by the implementation. If fixedsamplelocations is TRUE, the image will use identical sample locations and the same number of samples for all texels in the image, and the sample locations will not depend on the internalformat or size of the image. Upon success, TexStorage*DMultisample delete any existing image for target and the contents of texels are undefined. The values of TEXTURE_WIDTH, TEXTURE_HEIGHT, TEXTURE_SAMPLES, TEXTURE_INTERNAL_FORMAT and TEXTURE_FIXED_SAMPLE_LOCATIONS are set to width, height, the actual number of samples allocated, internalformat, and fixedsamplelocations respectively. When a multisample texture is accessed in a shader, the access takes one vector of integers describing which texel to fetch and an integer corresponding to the sample numbers described in section 13.4 describing which sample within the texel to fetch. No standard sampling instructions are allowed on the multisample texture targets, and no filtering is performed by the fetch. Fetching a sample number less than zero, or greater than or equal to the number of samples in the texture, produces undefined results. Errors An INVALID_ENUM error is generated if target is not an accepted multisample target as described above. An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_VALUE error is generated if width, height or depth is less than 1. An INVALID_VALUE error is generated if samples is zero. An INVALID_VALUE error is generated if width or height is greater than the value of MAX_TEXTURE_SIZE. An INVALID_VALUE error is generated by TexStorage3DMultisample if depth is greater than the value of MAX_ARRAY_TEXTURE_LAYERS. An INVALID_ENUM error is generated if internalformat, is not colorrenderable, depth-renderable, or stencil-renderable (as defined in section 9.4). An INVALID_ENUM error is generated if internalformat is one of the unsized base internal formats listed in table 8.8. An INVALID_OPERATION error is generated if samples is greater than the maximum number of samples supported for this target and internalformat. The maximum number of samples supported can be determined by calling GetInternalformativ with a pname of SAMPLES (see section 20.3).

OpenGL ES 3.2 (May 5, 2022)

8.9. BUFFER TEXTURES

185

An INVALID_OPERATION error is generated if the value of TEXTURE_IMMUTABLE_FORMAT for the texture currently bound to target on the active texture unit is TRUE.

8.9

Buffer Textures

In addition to the types of textures described in previous sections, one additional type of texture is supported. A buffer texture is similar to a one-dimensional texture. However, unlike other texture types, the texture image is not stored as part of the texture. Instead, a buffer object is attached to a buffer texture and the texture image is taken from that buffer object’s data store. When the contents of a buffer object’s data store are modified, those changes are reflected in the contents of any buffer texture to which the buffer object is attached. Buffer textures do not have multiple image levels; only a single data store is available. The command void TexBufferRange( enum target, enum internalformat, uint buffer, intptr offset, sizeiptr size ); attaches the range of the storage for the buffer object named buffer for size basic machine units, starting at offset (also in basic machine units) to the buffer texture currently bound to target. target must be TEXTURE_BUFFER. If buffer is zero, then any buffer object attached to the buffer texture is detached, the values offset and size are ignored and the state for offset and size for the buffer texture are reset to zero. internalformat specifies the storage format for the texture image found in the range of the attached buffer object, and must be one of the sized internal formats found in table 8.18. Errors An INVALID_ENUM error is generated if target is not TEXTURE_BUFFER. An INVALID_ENUM error is generated if internalformat is not one of the sized internal formats in table 8.18. An INVALID_OPERATION error is generated if buffer is non-zero and is not the name of a buffer object. An INVALID_VALUE error is generated if offset is negative, if size is less than or equal to zero, or if offset + size is greater than the value of BUFFER_SIZE for the buffer bound to target. An INVALID_VALUE error is generated if offset is not an integer multiple of the value of TEXTURE_BUFFER_OFFSET_ALIGNMENT. OpenGL ES 3.2 (May 5, 2022)

8.9. BUFFER TEXTURES

186

The command void TexBuffer( enum target, enum internalformat, uint buffer ); is equivalent to TexBufferRange(target, internalf ormat, buf f er, 0, size); where size is the value of BUFFER_SIZE for buffer. When a range of the storage of a buffer object is attached to a buffer texture, the range of the buffer’s data store is taken as the texture’s texture image. The number of texels in the buffer texture’s texture image is given by   size . components × sizeof (base type) where components and base type are the element count and base type for elements, as specified in table 8.18. The number of texels in the texture image is then clamped to an implementation-dependent limit, the value of MAX_TEXTURE_BUFFER_SIZE. When a buffer texture is accessed in a shader, the results of a texel fetch are undefined if the specified texel coordinate is negative, or greater than or equal to the clamped number of texels in the texture image. When a buffer texture is accessed in a shader, an integer is provided to indicate the texel coordinate being accessed. If no buffer object is bound to the buffer texture, the results of the texel access are undefined. Otherwise, the attached buffer object’s data store is interpreted as an array of elements of the GL data type corresponding to internalformat. Each texel consists of one to four elements that are mapped to texture components (R, G, B, and A). Element m of the texel numbered n is taken from element n × components + m of the attached buffer object’s data store. Elements and texels are both numbered starting with zero. For texture formats with signed or unsigned normalized fixed-point components, the extracted values are converted to floating-point using equations 2.2 or 2.1, respectively. The components of the texture are then converted to a (R, G, B, A) vector according to table 8.18, and returned to the shader as a four-component result vector with components of the appropriate data type for the texture’s internal format. The base data type, component count, normalized component information, and mapping of data store elements to texture components is specified in table 8.18.

OpenGL ES 3.2 (May 5, 2022)

8.9. BUFFER TEXTURES Sized Internal Format

Base Type

187 Components

Norm

Component 0 1 2 3 R8 ubyte 1 Yes R 0 0 1 R16F half 1 No R 0 0 1 R32F float 1 No R 0 0 1 R8I byte 1 No R 0 0 1 R16I short 1 No R 0 0 1 R32I int 1 No R 0 0 1 R8UI ubyte 1 No R 0 0 1 R16UI ushort 1 No R 0 0 1 R32UI uint 1 No R 0 0 1 RG8 ubyte 2 Yes R G 0 1 RG16F half 2 No R G 0 1 RG32F float 2 No R G 0 1 RG8I byte 2 No R G 0 1 RG16I short 2 No R G 0 1 RG32I int 2 No R G 0 1 RG8UI ubyte 2 No R G 0 1 RG16UI ushort 2 No R G 0 1 RG32UI uint 2 No R G 0 1 RGB32F float 3 No R G B 1 RGB32I int 3 No R G B 1 RGB32UI uint 3 No R G B 1 RGBA8 ubyte 4 Yes R G B A RGBA16F half 4 No R G B A RGBA32F float 4 No R G B A RGBA8I byte 4 No R G B A RGBA16I short 4 No R G B A RGBA32I int 4 No R G B A RGBA8UI ubyte 4 No R G B A RGBA16UI ushort 4 No R G B A RGBA32UI uint 4 No R G B A Table 8.18: Internal formats for buffer textures. For each format, the data type of each element is indicated in the “Base Type” column and the element count is in the “Components” column. The “Norm” column indicates whether components should be treated as normalized floating-point values. The “Component 0, 1, 2, and 3” columns indicate the mapping of each element of a texel to texture components. OpenGL ES 3.2 (May 5, 2022)

8.10. TEXTURE PARAMETERS

188

In addition to attaching buffer objects to textures, buffer objects can be bound to the buffer object target named TEXTURE_BUFFER, in order to specify, modify, or read the buffer object’s data store. The buffer object bound to TEXTURE_BUFFER has no effect on rendering. A buffer object is bound to TEXTURE_BUFFER by calling BindBuffer with target set to TEXTURE_BUFFER, as described in section 6.

8.10

Texture Parameters

Texture parameters control how the texture image of a texture object is treated when specified or changed, and when applied to a fragment. Each parameter is set with the commands void TexParameter{if}( enum target, enum pname, T param ); void TexParameter{if}v( enum target, enum pname, const T *params ); void TexParameterI{i ui}v( uint texture, enum pname, const T *params ); target is the target, and must be one of TEXTURE_2D, TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_2D_MULTISAMPLE, or TEXTURE_2D_MULTISAMPLE_ARRAY. pname is a symbolic constant indicating the parameter to be set; the possible constants and corresponding parameters are summarized in table 8.19. In the scalar forms of the command, param is a value to which to set a single-valued parameter; in the vector forms, params is an array of parameters whose type depends on the parameter being set. Data conversions are performed as specified in section 2.2.1, with these exceptions: • If the values for TEXTURE_BORDER_COLOR are specified with TexParameterIiv or TexParameterIuiv, they are unmodified and stored with an internal data type of integer. If specified with TexParameteriv, they are converted to floating-point using equation 2.2. Otherwise, the values are unmodified and stored as floating-point.

OpenGL ES 3.2 (May 5, 2022)

8.10. TEXTURE PARAMETERS Name DEPTH_STENCIL_TEXTURE_MODE

189 Type enum

TEXTURE_COMPARE_MODE

int 4 floats, ints, or uints enum

TEXTURE_COMPARE_FUNC

enum

TEXTURE_MAG_FILTER TEXTURE_MAX_LEVEL TEXTURE_MAX_LOD TEXTURE_MIN_FILTER

enum int float enum

TEXTURE_BASE_LEVEL TEXTURE_BORDER_COLOR

Legal Values DEPTH_COMPONENT, STENCIL_INDEX

any non-negative integer any 4 values NONE, COMPARE_REF_TO_TEXTURE LEQUAL, GEQUAL, LESS, GREATER, EQUAL, NOTEQUAL, ALWAYS, NEVER NEAREST, LINEAR

any non-negative integer any value NEAREST, LINEAR, NEAREST_MIPMAP_NEAREST, NEAREST_MIPMAP_LINEAR, LINEAR_MIPMAP_NEAREST, LINEAR_MIPMAP_LINEAR,

TEXTURE_MIN_LOD TEXTURE_SWIZZLE_R

float enum

TEXTURE_SWIZZLE_G

enum

TEXTURE_SWIZZLE_B

enum

TEXTURE_SWIZZLE_A

enum

TEXTURE_WRAP_S

enum

TEXTURE_WRAP_T

enum

TEXTURE_WRAP_R

enum

any value RED, GREEN, BLUE, ALPHA, ZERO, ONE RED, GREEN, BLUE, ALPHA, ZERO, ONE RED, GREEN, BLUE, ALPHA, ZERO, ONE RED, GREEN, BLUE, ALPHA, ZERO, ONE CLAMP_TO_EDGE, REPEAT, MIRRORED_REPEAT, CLAMP_TO_BORDER CLAMP_TO_EDGE, REPEAT, MIRRORED_REPEAT, CLAMP_TO_BORDER CLAMP_TO_EDGE, REPEAT, MIRRORED_REPEAT, CLAMP_TO_BORDER

Table 8.19: Texture parameters and their values.

OpenGL ES 3.2 (May 5, 2022)

8.10. TEXTURE PARAMETERS

190

0

In the remainder of chapter 8, denote by lodmin , lodmax , levelbase , and 0 levelmax the values of the texture parameters TEXTURE_MIN_LOD, TEXTURE_MAX_LOD, TEXTURE_BASE_LEVEL, and TEXTURE_MAX_LEVEL respectively. Let  levelbase =

0

min(levelbase , levelimmut − 1), for immutable-format textures 0 otherwise levelbase , (8.4)

and 

0

min(max(levelbase , levelmax ), levelimmut − 1), for immutable-format textures 0 levelmax , otherwise (8.5) levelimmut is the levels parameter passed to TexStorage* for the texture object (the value of TEXTURE_IMMUTABLE_LEVELS; see section 8.18). Texture parameters for a cube map texture apply to the cube map as a whole; the six distinct two-dimensional texture images use the texture parameters of the cube map itself.

levelmax =

Errors An INVALID_ENUM error is generated if target is not one of the valid targets listed above. An INVALID_ENUM error is generated if pname is not one of the parameter names in table 8.19. An INVALID_ENUM error is generated if the type of the parameter specified by pname is enum, and the value(s) specified by param or params are not among the legal values shown in table 8.19. An INVALID_VALUE error is generated if pname is TEXTURE_BASE_LEVEL or TEXTURE_MAX_LEVEL, and param or params is negative. An INVALID_VALUE error is generated if pname is TEXTURE_BASE_LEVEL or TEXTURE_MAX_LEVEL, and param or params is negative. An INVALID_ENUM error is generated if Tex*Parameter{if} is called for a non-scalar parameter (pname TEXTURE_BORDER_COLOR). An INVALID_ENUM error is generated if target is TEXTURE_2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY, and pname is any sampler state from table 21.12. An INVALID_OPERATION error is generated if target is TEXTURE_-

OpenGL ES 3.2 (May 5, 2022)

8.11. TEXTURE QUERIES

191

2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY, and pname TEXTURE_BASE_LEVEL is set to a value other than zero.

8.11

Texture Queries

8.11.1

Active Texture

As discussed in section 2.2.2, queries of most texture state variables are qualified by the value of ACTIVE_TEXTURE to determine which server texture state vector is queried.

8.11.2

Texture Parameter Queries

Parameters of a texture object may be queried with the commands void GetTexParameter{if}v( enum target, enum pname, T *params ); void GetTexParameterI{i ui}v( enum target, enum pname, T *params ); The texture object is that which is bound to target. The value of texture parameter pname for the texture is returned in params. target must be one of TEXTURE_2D, TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_2D_MULTISAMPLE, or TEXTURE_2D_MULTISAMPLE_ARRAY, indicating the currently bound two-dimensional, three-dimensional, two-dimensional array, cube map, cube map array, two-dimensional multisample, or two-dimensional multisample array texture object, respectively. pname must be one of IMAGE_FORMAT_COMPATIBILITY_TYPE, TEXTURE_IMMUTABLE_FORMAT, TEXTURE_IMMUTABLE_LEVELS, or one of the symbolic values in table 8.19. Querying pname TEXTURE_BORDER_COLOR with GetTexParameterIiv or GetTexParameterIuiv returns the border color values as signed integers or unsigned integers, respectively; otherwise the values are returned as described in section 2.2.2. If the border color is queried with a type that does not match the original type with which it was specified, the result is undefined. Errors An INVALID_ENUM error is generated if target is not one of the texture

OpenGL ES 3.2 (May 5, 2022)

8.11. TEXTURE QUERIES

192

targets described above. An INVALID_ENUM error is generated if pname is not one of the texture parameters described above.

8.11.3

Texture Level Parameter Queries

The commands void GetTexLevelParameter{if}v( enum target, int level, enum pname, T *params ); place information about texture image parameter pname for level-of-detail level of the specified target into params. pname must be one of the symbolic values in table 21.11. target may be one of TEXTURE_2D, TEXTURE_3D, TEXTURE_2D_ARRAY, one of the cube map face targets from table 8.20, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_BUFFER, TEXTURE_2D_MULTISAMPLE, or TEXTURE_2D_MULTISAMPLE_ARRAY, indicating the two-or three-dimensional texture, twodimensional array texture, one of the six distinct 2D images making up the cube map texture object, cube map array texture, buffer texture, two-dimensional multisample texture, or two-dimensional multisample array texture. level determines which level-of-detail’s state is returned. The maximum value of level depends on the texture target: • For cube map face targets, the maximum value is log2 of the value of MAX_CUBE_MAP_TEXTURE_SIZE. • For target TEXTURE_3D, the maximum value is log2 of the value of MAX_3D_TEXTURE_SIZE. • For targets TEXTURE_BUFFER, TEXTURE_2D_MULTISAMPLE, and TEXTURE_2D_MULTISAMPLE_ARRAY, which do not support mipmaps, the maximum value is zero. • For all other texture targets supported by GetTexLevelParameter*, the maximum value is log2 of the value of MAX_TEXTURE_SIZE. Note that TEXTURE_CUBE_MAP is not a valid target parameter for GetTexLevelParameter, because it does not specify a particular cube map face. For texture images with uncompressed internal formats, queries of pname TEXTURE_RED_TYPE, TEXTURE_GREEN_TYPE, TEXTURE_BLUE_TYPE, TEXTURE_ALPHA_TYPE, and TEXTURE_DEPTH_TYPE return the data type used OpenGL ES 3.2 (May 5, 2022)

8.12. DEPTH COMPONENT TEXTURES

193

to store the component. Types NONE, SIGNED_NORMALIZED, UNSIGNED_NORMALIZED, FLOAT, INT, and UNSIGNED_INT respectively indicate missing, signed normalized fixed-point, unsigned normalized fixed-point, floating-point, signed unnormalized integer, and unsigned unnormalized integer components. Queries of pname TEXTURE_RED_SIZE, TEXTURE_GREEN_SIZE, TEXTURE_BLUE_SIZE, TEXTURE_ALPHA_SIZE, TEXTURE_DEPTH_SIZE, TEXTURE_STENCIL_SIZE, and TEXTURE_SHARED_SIZE return the actual resolutions of the stored image components, not the resolutions specified when the image was defined. For texture images with compressed internal formats, the types returned specify how components are interpreted after decompression, while the resolutions returned specify the component resolution of an uncompressed internal format that produces an image of roughly the same quality as the compressed image in question. Since the quality of the implementation’s compression algorithm is likely data-dependent, the returned component sizes should be treated only as rough approximations. Queries of pname TEXTURE_INTERNAL_FORMAT, TEXTURE_WIDTH, TEXTURE_HEIGHT, and TEXTURE_DEPTH return the internal format, width, height, and depth, respectively, as specified when the texture image was created. Queries of pname TEXTURE_SAMPLES, and TEXTURE_FIXED_SAMPLE_LOCATIONS on multisample textures return the number of samples and whether texture sample fixed locations are enabled, respectively. For non-multisample textures, the default values in table 21.11 are returned. Errors An INVALID_ENUM error is generated if target is not one of the texture targets described above. An INVALID_ENUM error is generated if pname is not one of the symbolic values in tables 21.11. An INVALID_VALUE error is generated if level is negative or larger than the maximum allowable level-of-detail for target as described above.

8.12

Depth Component Textures

Depth textures and the depth components of depth/stencil textures can be treated as RED textures during texture filtering and application (see section 8.20).

OpenGL ES 3.2 (May 5, 2022)

8.13. CUBE MAP TEXTURE SELECTION Major Axis Direction +rx −rx +ry −ry +rz −rz

Target TEXTURE_CUBE_MAP_POSITIVE_X TEXTURE_CUBE_MAP_NEGATIVE_X TEXTURE_CUBE_MAP_POSITIVE_Y TEXTURE_CUBE_MAP_NEGATIVE_Y TEXTURE_CUBE_MAP_POSITIVE_Z TEXTURE_CUBE_MAP_NEGATIVE_Z

194 sc −rz rz rx rx rx −rx

tc −ry −ry rz −rz −ry −ry

ma rx rx ry ry rz rz

Table 8.20: Selection of cube map images based on major axis direction of texture coordinates.

8.13

Cube Map Texture Selection

When cube map texturing is enabled,
the s t r texture coordinates are treated as a direction vector rx ry rz emanating from the center of a cube. At texture application time, the interpolated per-fragment direction vector selects one of the cube map face’s two-dimensional images based on the largest magnitude coordinate direction (the major axis direction). If two or more coordinates have the identical magnitude, the implementation may define the rule to disambiguate
this situation. The rule must be deterministic and depend only on rx ry rz . The target column in table 8.20 explains how the major axis direction maps to the twodimensional image of a particular cube map target. Using the sc , tc , and ma
determined by the major axis direction as specified in table 8.20, an updated s t is calculated as follows:   1 sc s= +1 2 |ma |   1 tc t= +1 2 |ma |

8.13.1

Seamless Cube Map Filtering

The rules for texel selection in sections 8.14 through 8.15 are modified for cube maps so that texture wrap modes are ignored3 . Instead, • If NEAREST filtering is done within a miplevel, always apply apply wrap mode CLAMP_TO_EDGE. 3

This is a behavior change in OpenGL ES 3.0. In previous versions, texture wrap modes were respected and neighboring cube map faces were not used for border texels.

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

195

• If LINEAR filtering is done within a miplevel, always apply wrap mode CLAMP_TO_BORDER. Then, – If a texture sample location would lie in the texture border in either u or v, instead select the corresponding texel from the appropriate neighboring face. – If a texture sample location would lie in the texture border in both u and v (in one of the corners of the cube), there is no unique neighboring face from which to extract one texel. The recommended method to generate this texel is to average the values of the three available samples. However, implementations are free to construct this fourth texel in another way, so long as, when the three available samples have the same value, this texel also has that value.

8.14

Texture Minification

Applying a texture to a primitive implies a mapping from texture image space to framebuffer image space. In general, this mapping involves a reconstruction of the sampled texture image, followed by a homogeneous warping implied by the mapping to framebuffer space, then a filtering, followed finally by a resampling of the filtered, warped, reconstructed image before applying it to a fragment. In the GL this mapping is approximated by one of two simple filtering schemes. One of these schemes is selected based on whether the mapping from texture space to framebuffer space is deemed to magnify or minify the texture image.

8.14.1

Scale Factor and Level of Detail

The choice is governed by a scale factor ρ(x, y) and the level-of-detail parameter λ(x, y), defined as λbase (x, y) = log2 [ρ(x, y)]

(8.6)

λ0 (x, y) = λbase (x, y) + clamp(biasshader )

(8.7)

λ0 > lodmax lodmin ≤ λ0 ≤ lodmax λ0 < lodmin lodmin > lodmax

(8.8)

  max ,  lod  λ0 , λ= lodmin ,    undef ined,

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

196

biasshader is the value of the optional bias parameter in the texture lookup functions available to fragment shaders. If the texture access is performed in a fragment shader without a provided bias, or outside a fragment shader, then biasshader is zero. The sum of these values is clamped to the range [−biasmax , biasmax ] where biasmax is the value of the implementation defined constant MAX_TEXTURE_LOD_BIAS. If λ(x, y) is less than or equal to zero the texture is said to be magnified; if it is greater, the texture is minified. Sampling of minified textures is described in the remainder of this section, while sampling of magnified textures is described in section 8.15. The initial values of lodmin and lodmax are chosen so as to never clamp the normal range of λ. Let s(x, y) be the function that associates an s texture coordinate with each set of window coordinates (x, y) that lie within a primitive; define t(x, y) and r(x, y) analogously. Let u(x, y) = wt × s(x, y) + δu v(x, y) = ht × t(x, y) + δv

(8.9)

w(x, y) = dt × r(x, y) + δw where wt , ht , and dt are the width, height, and depth of the texture image whose level is levelbase . For a two-dimensional, two-dimensional array, cube map, or cube map array texture, define w(x, y) = 0. (δu , δv , δw ) are the texel offsets specified in the OpenGL ES Shading Language texture lookup functions that support offsets. If the texture function used does not support offsets, all three shader offsets are taken to be zero. If the value of any non-ignored component of the offset vector operand is outside implementation-dependent limits, the results of the texture lookup are undefined. For all instructions except textureGather, the limits are the values of MIN_PROGRAM_TEXEL_OFFSET and MAX_PROGRAM_TEXEL_OFFSET. For the textureGather instruction, the limits are the values of MIN_PROGRAM_TEXTURE_GATHER_OFFSET and MAX_PROGRAM_TEXTURE_GATHER_OFFSET. The value of MIN_PROGRAM_TEXTURE_GATHER_OFFSET must be less than or equal to the value of MIN_PROGRAM_TEXEL_OFFSET. The value of MAX_PROGRAM_TEXTURE_GATHER_OFFSET must be greater than or equal to the value of MAX_PROGRAM_TEXEL_OFFSET. A point sampled in screen space has an elliptical footprint in texture space. The ideal scale factor ρ should be the major axis of this ellipse. ∂u ∂v ∂v ∂w ∂w The derivatives ∂u ∂x , ∂y , ∂x , ∂y , ∂x , and ∂y are used to calculate the value of ρ for a fragment with window coordinates (x, y). OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

197

Computing the ellipse major axis using the derivatives in x and y can be impractical to implement. Therefore, an implementation may approximate the ideal ρ with a function f (x, y) subject to these conditions: 1. f (x, y) is continuous and monotonically increasing in each of

∂u ∂x ,

∂u ∂y ,

∂v ∂v ∂w ∂w ∂x , ∂y , ∂x , and ∂y ∂u ∂v ∂w ∂v ∂w 2. max( ∂u ∂x , ∂y , ∂x , ∂y , ∂x , ∂y ) ≤ f (x, y) ≤ ∂v ∂x

8.14.2

√

2 max( ∂u ∂x +

∂u ∂v ∂w + ∂w ∂x , ∂y + ∂y + ∂y )

Coordinate Wrapping and Texel Selection

After generating u(x, y), v(x, y), and w(x, y), they may be clamped and wrapped before sampling the texture, depending on the corresponding texture wrap modes. Let u0 (x, y) = u(x, y), v 0 (x, y) = v(x, y), and w0 (x, y) = w(x, y). The value assigned to TEXTURE_MIN_FILTER is used to determine how the texture value for a fragment is selected. When the value of TEXTURE_MIN_FILTER is NEAREST, the texel in the texture image of level levelbase that is nearest (in Manhattan distance) to (u0 , v 0 , w0 ) is obtained. Let (i, j, k) be integers such that i = wrap(bu0 (x, y)c) j = wrap(bv 0 (x, y)c) k = wrap(bw0 (x, y)c) and the value returned by wrap() is defined in table 8.21. For a three-dimensional texture, the texel at location (i, j, k) becomes the texture value. For twodimensional, two-dimensional array, or cube map textures, k is irrelevant, and the texel at location (i, j) becomes the texture value. For two-dimensional array textures, the texel is obtained from image layer l, where l = clamp(RN E(r), 0, dt − 1)4 and RN E() is the round-to-nearest-even operation defined by IEEE arithmetic. 4

Implementations may instead round the texture layer using the nearly equivalent computation br + 12 c.

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION Wrap mode CLAMP_TO_EDGE CLAMP_TO_BORDER REPEAT MIRRORED_REPEAT

198

Result of wrap(coord) clamp(coord, 0, size − 1) clamp(coord, −1, size) f mod(coord, size) (size − 1) − mirror(f mod(coord, 2 × size) − size)

Table 8.21: Texel location wrap mode application. f mod(a, b) returns a − b × b ab c. mirror(a) returns a if a ≥ 0, and −(1 + a) otherwise. The values of mode and size are TEXTURE_WRAP_S and wt , TEXTURE_WRAP_T and ht , and TEXTURE_WRAP_R and dt when wrapping i, j, or k coordinates, respectively.

If the selected (i, j, k), (i, j), or i location refers to a border texel that satisfies any of the conditions

i<0

i ≥ wt

j<0

j ≥ ht

k<0

k ≥ dt

then the border values defined by TEXTURE_BORDER_COLOR are used in place of the non-existent texel. If the texture contains color components, the values of TEXTURE_BORDER_COLOR are interpreted as an RGBA color to match the texture’s internal format in a manner consistent with table 8.8. The internal data type of the border values must be consistent with the type returned by the texture as described in chapter 8, or the result is undefined. Border values are clamped before they are used, according to the format in which texture components are stored. For signed and unsigned normalized fixed-point formats, border values are clamped to [−1, 1] and [0, 1], respectively. For floating-point and integer formats, border values are clamped to the representable range of the format. For compressed formats, border values are clamped as signed normalized (“snorm”), unsigned normalized (“unorm”), or floating-point as described in table 8.17 for each format. If the texture contains depth components, the first component of TEXTURE_BORDER_COLOR is interpreted as a depth value. When the value of TEXTURE_MIN_FILTER is LINEAR, a 2 × 2 × 2 cube of texels in the texture image of level levelbase is selected. Let

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

199

i0 = wrap(bu0 − 0.5c) j0 = wrap(bv 0 − 0.5c) k0 = wrap(bw0 − 0.5c) i1 = wrap(bu0 − 0.5c + 1) j1 = wrap(bv 0 − 0.5c + 1) k1 = wrap(bw0 − 0.5c + 1) α = f rac(u0 − 0.5) β = f rac(v 0 − 0.5) γ = f rac(w0 − 0.5) where f rac(x) denotes the fractional part of x. For a three-dimensional texture, the texture value τ is found as τ = (1 − α)(1 − β)(1 − γ)τi0 j0 k0 + α(1 − β)(1 − γ)τi1 j0 k0 + (1 − α)β(1 − γ)τi0 j1 k0 + αβ(1 − γ)τi1 j1 k0 + (1 − α)(1 − β)γτi0 j0 k1 + α(1 − β)γτi1 j0 k1

(8.10)

+ (1 − α)βγτi0 j1 k1 + αβγτi1 j1 k1 where τijk is the texel at location (i, j, k) in the three-dimensional texture image. For a two-dimensional, two-dimensional array, or cube map texture, τ =(1 − α)(1 − β)τi0 j0 + α(1 − β)τi1 j0 + (1 − α)βτi0 j1 + αβτi1 j1 where τij is the texel at location (i, j) in the two-dimensional texture image. For two-dimensional array textures, all texels are obtained from layer l, where l = clamp(br + 0.5c, 0, dt − 1). The textureGather and textureGatherOffset built-in shader functions return a vector derived from sampling a 2 × 2 block of texels in the texture image of level levelbase . The rules for the LINEAR minification filter are applied to identify the four selected texels. Each texel is then converted to a texture source color (Rs , Gs , Bs , As ) according to table 14.1 and then swizzled as described in section 14.2.1. A four-component vector is then assembled by taking a single component from the swizzled texture source colors of the four texels, in the order τi0 j1 , OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

200

Figure 8.7. An example of an 8 × 8 texture image and the components returned for textureGather. The vector (X, Y, Z, W ) is returned, where each component is taken from the post-swizzle component selected by comp of the corresponding texel.

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

201

τi1 j1 , τi1 j0 , and τi0 j0 (see figure 8.7). The selected component is identified by the by the optional comp argument, where the values zero, one, two, and three identify the Rs , Gs , Bs , or As component, respectively. If comp is omitted, it is treated as identifying the Rs component. Incomplete textures (see section 8.17) are considered to return a texture source color of (0.0, 0.0, 0.0, 1.0) in floating-point format for all four source texels. For any texel in the equation above that refers to a border texel outside the defined range of the image, the texel value is taken from the texture border color as with NEAREST filtering. The textureGatherOffsets functions operate identically to textureGather, except that the array of two-component integer vectors offsets is used to determine the location of the four texels to sample. Each of the four texels is obtained by applying the corresponding offset in the four-element array offsets as a (u, v) coordinate offset to the coordinates coord, identifying the four-texel LINEAR footprint, and then selecting the texel τi0 j0 of that footprint. The specified values in offsets must be constant. A limited range of offset values are supported; the minimum and maximum offset values are implementation-dependent and given by the values of MIN_PROGRAM_TEXTURE_GATHER_OFFSET and MAX_PROGRAM_TEXTURE_GATHER_OFFSET, respectively. Note that offset does not apply to the layer coordinate for array textures. 8.14.2.1

Rendering Feedback Loops

If all of the following conditions are satisfied, then the value of the selected τijk or τij in the above equations is undefined instead of referring to the value of the texel at location (i, j, k) or (i, j), respectively. This situation is discussed in more detail in the description of feedback loops in section 9.3.1. • The current DRAW_FRAMEBUFFER_BINDING names a framebuffer object F. • The texture is attached to one of the attachment points, A, of framebuffer object F. • The value of TEXTURE_MIN_FILTER is NEAREST or LINEAR, and the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL for attachment point A is equal to levelbase -orThe value of TEXTURE_MIN_FILTER is NEAREST_MIPMAP_NEAREST, NEAREST_MIPMAP_LINEAR, LINEAR_MIPMAP_NEAREST, or LINEAR_OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

202

MIPMAP_LINEAR, and the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL for attachment point A is within the inclusive range from

levelbase to q.

8.14.3

Mipmapping

TEXTURE_MIN_FILTER values NEAREST_MIPMAP_NEAREST, NEAREST_MIPMAP_LINEAR, LINEAR_MIPMAP_NEAREST, and LINEAR_MIPMAP_LINEAR

each require the use of a mipmap. A mipmap is an ordered set of arrays representing the same image; each array has a resolution lower than the previous one. If the texture image of level levelbase has dimensions wt × ht × dt , then there are blog2 (maxsize)c + 1 levels in the mipmap. where ( max(wt , ht ), maxsize = max(wt , ht , dt ),

for 2D, 2D array, cube map, and cube map array textures for 3D textures

Numbering the levels such that level levelbase is the 0th level, the ith array has dimensions       wt ht dt max(1, ) × max(1, ) × max(1, ) wd hd dd where wd = 2i hd = 2i ( 2i , for 3D textures dd = 1, otherwise until the last array is reached with dimension 1 × 1 × 1. Each array in a mipmap is defined using TexImage3D, TexImage2D, CopyTexImage2D, or by functions that are defined in terms of these functions. The array being set is indicated with the level-of-detail argument level. Level-of-detail numbers proceed from levelbase for the original texture image through the maximum level p, with each unit increase indicating an array of half the dimensions of the previous one (rounded down to the next integer if fractional) as already described. For immutable-format textures, p is one less than levelimmut . Otherwise, p = blog2 (maxsize)c + levelbase , and all arrays from levelbase through q = min{p, levelmax } must be defined, as discussed in section 8.17. OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

203

The mipmap is used in conjunction with the level of detail to approximate the application of an appropriately filtered texture to a fragment. Since this discussion pertains to minification, we are concerned only with values of λ where λ > 0. For mipmap filters NEAREST_MIPMAP_NEAREST and LINEAR_MIPMAP_NEAREST, the dth mipmap array is selected, where   λ≤0 levelbase , d = nearest(λ), λ > 0, levelbase + λ ≤ q + 12 (8.11)   1 q, λ > 0, levelbase + λ > q + 2 where (  levelbase + λ + 21 − 1, preferred  nearest(λ) =  levelbase + λ + 12 , alternative The rules for NEAREST or LINEAR filtering are then applied to the selected array. Specifically, the coordinate (u, v, w) is computed as in equation 8.9, with ws , hs , and ds equal to the width, height, and depth of the texture image whose level is d. For mipmap filters NEAREST_MIPMAP_LINEAR and LINEAR_MIPMAP_LINEAR, the level d1 and d2 mipmap arrays are selected, where ( q, levelbase + λ ≥ q d1 = blevelbase + λc, otherwise ( q, levelbase + λ ≥ q d2 = d1 + 1, otherwise

(8.12) (8.13)

The rules for NEAREST or LINEAR filtering are then applied to each of the selected arrays, yielding two corresponding texture values τ1 and τ2 . Specifically, for level d1 , the coordinate (u, v, w) is computed as in equation 8.9, with ws , hs , and ds equal to the width, height, and depth of the texture image whose level is d1 . For level d2 the coordinate (u0 , v 0 , w0 ) is computed as in equation 8.9, with ws , hs , and ds equal to the width, height, and depth of the texture image whose level is d2 . The final texture value is then found as τ = [1 − frac(λ)]τ1 + frac(λ)τ2 .

OpenGL ES 3.2 (May 5, 2022)

8.14. TEXTURE MINIFICATION

8.14.4

204

Manual Mipmap Generation

Mipmaps can be generated manually with the command void GenerateMipmap( enum target ); where target is one of TEXTURE_2D, TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY. Mipmap generation affects the texture image attached to target. If target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, the texture bound to target must be cube complete or cube array complete, respectively, as defined in section 8.17. Otherwise, if levelbase is not defined, or if any dimension is zero, all mipmap levels are left unchanged. This is not an error. Mipmap generation replaces texture image levels levelbase + 1 through q with images derived from the levelbase image, regardless of their previous contents. All other mipmap levels, including levelbase , are left unchanged by this computation. If levelbase is not defined, all mipmap levels are left unchanged. This is not an error. The internal formats and effective internal formats of the derived mipmap images all match those of the levelbase image, and the dimensions of the derived images follow the requirements described in section 8.17. The contents of the derived images are computed by repeated, filtered reduction of the levelbase image. For two-dimensional array and cube map array textures, each layer is filtered independently. No particular filter algorithm is required, though a box filter is recommended. Any synchronization required before performing this reduction will be done within the Generate*Mipmap commands themselves5 . Errors An INVALID_ENUM error is generated if target is not TEXTURE_2D, TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY. An INVALID_OPERATION error is generated if target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture bound to target is not cube complete or cube array complete, respectively. An INVALID_OPERATION error is generated if the levelbase array was not 5

For example, if mipmaps were generated by texture fetches, a fetch barrier would be issued prior to reduction; or if mipmaps were generated on the CPU, a texture update barrier would be issued prior to reduction.

OpenGL ES 3.2 (May 5, 2022)

8.15. TEXTURE MAGNIFICATION

205

specified with an unsized internal format from table 8.3 or a sized internal format that is both color-renderable and texture-filterable according to table 8.10.

8.15

Texture Magnification

When λ indicates magnification, the value assigned to TEXTURE_MAG_FILTER determines how the texture value is obtained. There are two possible values for TEXTURE_MAG_FILTER: NEAREST and LINEAR. NEAREST behaves exactly as NEAREST for TEXTURE_MIN_FILTER and LINEAR behaves exactly as LINEAR for TEXTURE_MIN_FILTER as described in section 8.14, including the texture coordinate wrap modes specified in table 8.21. The level-of-detail levelbase texture image is always used for magnification.

8.16

Combined Depth/Stencil Textures

If the texture image has a base internal format of DEPTH_STENCIL, then the stencil index texture component is ignored by default. The texture value τ does not include a stencil index component, but includes only the depth component. In order to access the stencil index texture component, the DEPTH_STENCIL_TEXTURE_MODE texture parameter should be set to STENCIL_INDEX. When this mode is set the depth component is ignored and the texture value includes only the stencil index component. The stencil index value is treated as an unsigned integer texture and returns an unsigned integer value when sampled. When sampling the stencil index only NEAREST filtering is supported. The DEPTH_STENCIL_TEXTURE_MODE is ignored for non depth/stencil textures.

8.17

Texture Completeness

A texture is said to be complete if the texture images and (where applicable) texture parameters required to utilize the texture are consistently defined. Different operations on the texture require different levels of completeness and the definitions vary depending on texture dimensionality and type. A cube map texture is cube complete if: • The levelbase images of each of the six texture images making up the cube map have identical, positive, and square dimensions. • The levelbase images were each specified with the same effective internal format. OpenGL ES 3.2 (May 5, 2022)

8.17. TEXTURE COMPLETENESS

206

A cube map array texture is cube array complete if every slice within the array is cube complete. A texture is base complete if: • All dimensions of the levelbase images are positive. For a multisample texture, levelbase is always zero. • If the texture is a cube map texture, then it is cube complete. • If the texture is a cube map array texture, then it is cube array complete. A texture is mipmap complete if: • It is base complete • For every base level image, the set of mipmap images levelbase through q (where q is defined in section 8.14.3) were each specified with the same effective internal format. • The dimensions of the images follow the sequence described in section 8.14.3. • levelbase ≤ levelmax Image levels k where k < levelmax or k > q are insignificant to the definition of mipmap completeness. A texture is complete unless any of the following conditions hold true: • The texture is not base complete. • The minification filter requires a mimap (is neither NEAREST nor LINEAR), the texture is not multisample, and the texture is not mipmap complete. • The texture is not multisample; either the magnification filter is not NEAREST, or the minification filter is neither NEAREST nor NEAREST_MIPMAP_NEAREST; and any of – The effective internal format specified for the texture images is a sized internal color format that is not texture-filterable – The effective internal format specified for the texture images is a sized internal depth or depth and stencil format (see table 8.11), and the value of TEXTURE_COMPARE_MODE is NONE. – The internal format of the texture is DEPTH_STENCIL, and the value of DEPTH_STENCIL_TEXTURE_MODE for the texture is STENCIL_INDEX. – The internal format is STENCIL_INDEX. OpenGL ES 3.2 (May 5, 2022)

8.18. IMMUTABLE-FORMAT TEXTURE IMAGES

8.17.1

207

Effects of Sampler Objects on Texture Completeness

If a sampler object and a texture object are simultaneously bound to the same texture unit, then the sampling state for that unit is taken from the sampler object (see section 8.2). This can have an effect on the effective completeness of the texture. In particular, if the texture is not mipmap complete and the sampler object specifies a TEXTURE_MIN_FILTER requiring mipmaps, the texture will be considered incomplete for the purposes of that texture unit. However, if the sampler object does not require mipmaps, the texture object will be considered complete. This means that a texture can be considered both complete and incomplete simultaneously if it is bound to two or more texture units along with sampler objects with different states.

8.17.2

Effects of Completeness on Texture Application

Texture lookup and texture fetch operations performed in shaders are affected by completeness of the texture being sampled as described in sections 11.1.3.5 and 14.2.1.

8.17.3

Effects of Completeness on Texture Image Specification

The implementation-dependent maximum sizes for texture images depend on the texture level. In particular, an implementation may allow a texture texture image of level one or greater to be created only if a mipmap complete set of images consistent with the requested array can be supported where the values of TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL are 0 and 1000 respectively. As a result, implementations may permit a texture image at level zero that will never be mipmap complete and can only be used with non-mipmapped minification filters.

8.18

Immutable-Format Texture Images

An alternative set of commands is provided for specifying the properties of all levels of a texture at once. Once a texture is specified with such a command, the format and dimensions of all levels becomes immutable. The contents of the images and the parameters can still be modified. Such a texture is referred to as an immutable-format texture. The immutability status of a texture can be determined by calling GetTexParameter with pname TEXTURE_IMMUTABLE_FORMAT. Each of the commands below is described by pseudocode which indicates the effect on the dimensions and format of the texture. For each command the following apply in addition to the pseudocode:

OpenGL ES 3.2 (May 5, 2022)

8.18. IMMUTABLE-FORMAT TEXTURE IMAGES

208

• If executing the pseudocode would result in any other error, the error is generated and the command will have no effect. • Any existing levels that are not replaced are reset to their initial state. • The pixel unpack buffer should be considered to be zero; i.e., the image contents are unspecified. • Since no pixel data are provided, the format and type values used in the pseudocode are irrelevant; they can be considered to be any values that are legal to use with internalformat. • If the command is successful, TEXTURE_IMMUTABLE_FORMAT becomes TRUE and TEXTURE_IMMUTABLE_LEVELS becomes levels. • If internalformat is a compressed texture format, then references to TexImage* should be replaced by CompressedTexImage*, with format, type and data replaced by any valid imageSize and data. For each command, the following errors are generated in addition to the errors described specific to that command: Errors An INVALID_OPERATION error is generated if zero is bound to target. If executing the pseudo-code would result in a OUT_OF_MEMORY error, the error is generated and the results of executing the command are undefined. An INVALID_VALUE error is generated if width, height, depth or levels are less than 1, for commands with the corresponding parameters. An INVALID_OPERATION error is generated if internalformat is a compressed texture format and there is no imageSize for which the corresponding CompressedTexImage* command would have been valid. An INVALID_ENUM error is generated if internalformat is one of the unsized base internal formats listed in table 8.8. The command void TexStorage2D( enum target, sizei levels, enum internalformat, sizei width, sizei height ); specifies all the levels of a two-dimensional or cube map, texture at the same time. The pseudocode depends on target: target TEXTURE_2D: OpenGL ES 3.2 (May 5, 2022)

8.18. IMMUTABLE-FORMAT TEXTURE IMAGES

209

for (i = 0; i < levels; i++) { TexImage2D(target, i, internalf ormat, width, height, 0, f ormat, type,  NULL); width width = max(1, j 2 ); k height = max(1,

height 2

);

} target TEXTURE_CUBE_MAP: for (i = 0; i < levels; i++) { for face in (+X, -X, +Y, -Y, +Z, -Z) { TexImage2D(face, i, internalf ormat, width, height, 0, f ormat, type, NULL); }   ); width = max(1, jwidth 2 k height = max(1,

height 2

);

}

Errors An INVALID_ENUM error is generated if target is not TEXTURE_2D or TEXTURE_CUBE_MAP. An INVALID_OPERATION error is generated if levels is greater than blog2 (max(width, height))c + 1 The command void TexStorage3D( enum target, sizei levels, enum internalformat, sizei width, sizei height, sizei depth ); specifies all the levels of a three-dimensional, two-dimensional array or cube map array texture. The pseudocode depends on target: target TEXTURE_3D: for (i = 0; i < levels; i++) { TexImage3D(target, i, internalf ormat, width, height, depth, 0, f ormat, type, NULL); OpenGL ES 3.2 (May 5, 2022)

8.18. IMMUTABLE-FORMAT TEXTURE IMAGES

210

  width = max(1, jwidth ); 2 k

height = max(1, height ); j 2k depth = max(1, depth ); 2 } target TEXTURE_2D_ARRAY or TEXTURE_CUBE_MAP_ARRAY: for (i = 0; i < levels; i++) { TexImage3D(target, i, internalf ormat, width, height, depth, 0, f ormat, type,  widthNULL);  width = max(1, j 2 ); k height = max(1,

height 2

);

} Errors An INVALID_ENUM error is generated if target is not TEXTURE_3D, TEXTURE_2D_ARRAY or TEXTURE_CUBE_MAP_ARRAY. An INVALID_OPERATION error is generated if any of the following conditions hold: • target is TEXTURE_3D and levels blog2 (max(width, height, depth)))c + 1

is

greater

than

• target is TEXTURE_2D_ARRAY or TEXTURE_CUBE_MAP_ARRAY and levels is greater than blog2 (max(width, height))c + 1 • After a successful call to any TexStorage* command, no further changes to the dimensions or format of the texture object may be made. Other commands may only alter the texel values and texture parameters. Errors An INVALID_OPERATION error is generated by any of the following commands with the same texture, even if it does not affect the dimensions or format: • TexImage*

OpenGL ES 3.2 (May 5, 2022)

8.19. TEXTURE STATE

211

• CompressedTexImage* • CopyTexImage* • TexStorage*

8.19

Texture State

The state necessary for texture can be divided into two categories. First, there are the multiple sets of texture images (one set of mipmap images each for the two-and three-dimensional texture and two-dimensional array texture targets; and six sets of mipmap images each for the cube map and cube map array texture targets) and their number. Each image has associated with it a width, height (except for buffer textures) and depth (three-dimensional, two-dimensional array and cube map array only), an integer describing the internal format of the image, integer values describing the resolutions of each of the red, green, blue, alpha, depth, and stencil components of the image, integer values describing the type (unsigned normalized, integer, floating-point, etc.) of each of the components, a boolean describing whether the image is compressed or not, an integer size of a compressed image, and an integer containing the name of a buffer object bound as the data store of the image. Each initial texture image is null. It has zero width, height, and depth, internal format RGBA, component sizes set to zero and component types set to NONE, the compressed flag set to FALSE, a zero compressed size, and the bound buffer object name is zero. Multisample textures also contain an integer identifying the number of samples in each texel, and a boolean indicating whether identical sample locations and number of samples will be used for all texels in the image. Buffer textures also contain two pointer sized integers containing the offset and range of the buffer object’s data store. Next, there are the five sets of texture properties, corresponding to the twodimensional, two-dimensional array, three-dimensional, cube map, and cube map array texture targets. Each set consists of the selected minification and magnification filters, the wrap modes for s, t, and r (three-dimensional only), the texture border color, two floating-point numbers describing the minimum and maximum level of detail, two integers describing the base and maximum mipmap array, a boolean flag indicating whether the format and dimensions of the texture are immutable, two integers describing the compare mode and compare function (see section 8.20), an integer describing the depth stencil texture mode, and four integers describing the red, green, blue, and alpha swizzle modes (see section 14.2.1). OpenGL ES 3.2 (May 5, 2022)

8.20. TEXTURE COMPARISON MODES

212

In the initial state, the value assigned to TEXTURE_MIN_FILTER is NEAREST_MIPMAP_LINEAR and the value for TEXTURE_MAG_FILTER is LINEAR. s, t, and r wrap modes are all set to REPEAT. The values of TEXTURE_MIN_LOD and TEXTURE_MAX_LOD are -1000 and 1000 respectively. The values of TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL are 0 and 1000 respectively. The value of TEXTURE_BORDER_COLOR is (0, 0, 0, 0). The value of TEXTURE_IMMUTABLE_FORMAT is FALSE. The value of TEXTURE_IMMUTABLE_LEVELS is 0. The values of TEXTURE_COMPARE_MODE and TEXTURE_COMPARE_FUNC are NONE and LEQUAL respectively. The value of DEPTH_STENCIL_TEXTURE_MODE is DEPTH_COMPONENT. The values of TEXTURE_SWIZZLE_R, TEXTURE_SWIZZLE_G, TEXTURE_SWIZZLE_B, and TEXTURE_SWIZZLE_A are RED, GREEN, BLUE, and ALPHA, respectively.

8.20

Texture Comparison Modes

Texture values can also be computed according to a specified comparison function. Texture parameter TEXTURE_COMPARE_MODE specifies the comparison operands, and parameter TEXTURE_COMPARE_FUNC specifies the comparison function.

8.20.1

Depth Texture Comparison Mode

If the currently bound texture’s base internal format is DEPTH_COMPONENT or DEPTH_STENCIL, then TEXTURE_COMPARE_MODE and TEXTURE_COMPARE_FUNC control the output of the texture unit as described below. Otherwise, the texture unit operates in the normal manner and texture comparison is bypassed. Let Dt be the depth texture value and St be the stencil index component of a depth/stencil texture. If there is no stencil component, the value of St is undefined. Let Dref be the reference value, provided by the shader’s texture lookup function. If the texture’s internal format indicates a fixed-point depth texture, then Dt and Dref are clamped to the range [0, 1]; otherwise no clamping is performed. Then the effective texture value is computed as follows: • If the base internal format is STENCIL_INDEX, then r = St . • If the base internal format is DEPTH_STENCIL and the value of DEPTH_STENCIL_TEXTURE_MODE is STENCIL_INDEX, then r = St • Otherwise, if the value of TEXTURE_COMPARE_MODE is NONE, then r = Dt

OpenGL ES 3.2 (May 5, 2022)

8.21. SRGB TEXTURE COLOR CONVERSION

213

• Otherwise, if the value of TEXTURE_COMPARE_MODE is COMPARE_REF_TO_TEXTURE, then r depends on the texture comparison function as shown in table 8.22 Texture Comparison Function LEQUAL

GEQUAL

LESS

GREATER

EQUAL

NOTEQUAL ALWAYS NEVER

Computed result r ( 1.0, Dref ≤ Dt r= 0.0, Dref > Dt ( 1.0, Dref ≥ Dt r= 0.0, Dref < Dt ( 1.0, Dref < Dt r= 0.0, Dref ≥ Dt ( 1.0, Dref > Dt r= 0.0, Dref ≤ Dt ( 1.0, Dref = Dt r= 0.0, Dref 6= Dt ( 1.0, Dref 6= Dt r= 0.0, Dref = Dt r = 1.0 r = 0.0

Table 8.22: Depth texture comparison functions.

The resulting r is assigned to Rt . If the value of TEXTURE_MAG_FILTER is not NEAREST, or the value of TEXTURE_MIN_FILTER is not NEAREST or NEAREST_MIPMAP_NEAREST, then r may be computed by comparing more than one depth texture value to the texture reference value. The details of this are implementation-dependent, but r should be a value in the range [0, 1] which is proportional to the number of comparison passes or failures.

8.21

sRGB Texture Color Conversion

If the currently bound texture’s internal format is one of the sRGB formats in table 8.23, the red, green, and blue components are converted from an sRGB color space to a linear color space as part of filtering described in sections 8.14 and 8.15.

OpenGL ES 3.2 (May 5, 2022)

8.21. SRGB TEXTURE COLOR CONVERSION

214

Internal Format SRGB8 SRGB8_ALPHA8 COMPRESSED_SRGB8_ETC2 COMPRESSED_SRGB8_ALPHA8_ETC2_EAC COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_SRGB8_ALPHA8_ASTC_4x4 COMPRESSED_SRGB8_ALPHA8_ASTC_5x4 COMPRESSED_SRGB8_ALPHA8_ASTC_5x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x5 COMPRESSED_SRGB8_ALPHA8_ASTC_8x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x5 COMPRESSED_SRGB8_ALPHA8_ASTC_10x6 COMPRESSED_SRGB8_ALPHA8_ASTC_10x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x12

Table 8.23: sRGB texture internal formats.

Any alpha component is left unchanged. Ideally, implementations should perform this color conversion on each sample prior to filtering but implementations are allowed to perform this conversion after filtering (though this post-filtering approach is inferior to converting from sRGB prior to filtering). The conversion from an sRGB encoded component cs , to a linear component cl is as follows. ( cs , cs ≤ 0.04045
cl = 12.92 (8.14) cs +0.055 2.4 , cs > 0.04045 1.055 Assume cs is the sRGB component in the range [0, 1].

OpenGL ES 3.2 (May 5, 2022)

8.22. SHARED EXPONENT TEXTURE COLOR CONVERSION

8.22

215

Shared Exponent Texture Color Conversion

If the currently bound texture’s internal format is RGB9_E5, the red, green, blue, and shared bits are converted to color components (prior to filtering) using shared exponent decoding. The component reds , greens , blues , and exps values (see section 8.5.2) are treated as unsigned integers and are converted to floating-point red, green, and blue as follows: red = reds 2exps −B−N green = greens 2exps −B−N blue = blues 2exps −B−N

8.23

Texture Image Loads and Stores

The contents of a texture may be made available for shaders to read and write by binding the texture to one of a collection of image units. The GL implementation provides an array of image units numbered beginning with zero, with the total number of image units provided determined by the implementation-dependent value of MAX_IMAGE_UNITS. Unlike texture image units, image units do not have a separate attachment for each texture target texture; each image unit may have only one texture bound at a time. An immutable texture may be bound to an image unit for use by image loads and stores by calling: void BindImageTexture( uint unit, uint texture, int level, boolean layered, int layer, enum access, enum format ); where unit identifies the image unit, texture is the name of the texture, and level selects a single level of the texture. If texture is zero, any texture currently bound to image unit unit is unbound. If the texture identified by texture is a two-dimensional array, threedimensional, cube map or cube map array texture, it is possible to bind either the entire texture level or a single layer or face of the texture level. If layered is TRUE, the entire level is bound. If layered is FALSE, only the single layer identified by layer will be bound, and the layer is treated as a different texture target for image accesses; two-dimensional

OpenGL ES 3.2 (May 5, 2022)

8.23. TEXTURE IMAGE LOADS AND STORES Layer Number 0 1 2 3 4 5

216

Cube Map Face TEXTURE_CUBE_MAP_POSITIVE_X TEXTURE_CUBE_MAP_NEGATIVE_X TEXTURE_CUBE_MAP_POSITIVE_Y TEXTURE_CUBE_MAP_NEGATIVE_Y TEXTURE_CUBE_MAP_POSITIVE_Z TEXTURE_CUBE_MAP_NEGATIVE_Z

Table 8.24: Layer numbers for cube map texture faces. The layers are numbered in the same sequence as the cube map face token values.

array, three-dimensional, cube map, and cube map array texture layers are treated as two-dimensional textures. For cube map textures where layered is FALSE, the face is taken by mapping the layer number to a face according to table 8.24. For cube map array textures where layered is FALSE, the selected layer number is mapped to a texture layer and cube face using the following equations and mapping face to a face according to table 8.24.   layerorig layer = 6 f ace = layerorig − (layer × 6) If the texture identified by texture does not have multiple layers or faces, the entire texture level is bound, regardless of the values specified by layered and layer. format specifies the format that the elements of the image will be treated as when doing formatted stores, as described later in this section. This is referred to as the image unit format. access specifies whether the texture bound to the image will be treated as READ_ONLY, WRITE_ONLY, or READ_WRITE. If a shader reads from an image unit with a texture bound as WRITE_ONLY, or writes to an image unit with a texture bound as READ_ONLY, the results of that shader operation are undefined and may lead to application termination. If a texture object bound to one or more image units is deleted by DeleteTextures, it is detached from each such image unit, as though BindImageTexture were called with unit identifying the image unit and texture set to zero.

OpenGL ES 3.2 (May 5, 2022)

8.23. TEXTURE IMAGE LOADS AND STORES Texture target TEXTURE_2D TEXTURE_3D TEXTURE_CUBE_MAP TEXTURE_BUFFER TEXTURE_2D_ARRAY TEXTURE_CUBE_MAP_ARRAY

i x x x x x x

j y y y y y

217 k z -

Face / layer z z z

Table 8.25: Mapping of image load, store and atomic texel coordinate components to texel numbers.

Errors An INVALID_VALUE error is generated if unit is greater than or equal to the value of MAX_IMAGE_UNITS, if level or layer is negative, or if texture is not the name of an existing texture object. An INVALID_VALUE error is generated if format is not one of the formats listed in table 8.26. An INVALID_ENUM error is generated if the texture identified by texture is a two-dimensional multisample or two-dimensional multisample array texture. An INVALID_OPERATION error is generated if texture is neither the name of a buffer texture, nor the name of an immutable texture object. When a shader accesses the texture bound to an image unit using a built-in image load, store or atomic function, it identifies a single texel by providing a twoor three-dimensional coordinate. A coordinate vector is mapped to an individual texel τij or τijk according to the target of the texture bound to the image unit using table 8.25. As noted above, single-layer bindings of array or cube map textures are considered to use a texture target corresponding to the bound layer, rather than that of the full texture. If the texture target has layers or cube map faces, the layer or face number is taken from the layer argument of BindImageTexture if the texture is bound with layered set to FALSE, or from the coordinate identified by table 8.25 otherwise. For cube map and cube map array textures with layered set to TRUE, the coordinate is mapped to a layer and face in the same manner as the layer argument of BindImageTexture. If the individual texel identified for an image load, store or atomic operation doesn’t exist, the access is treated as invalid. Invalid image loads will return a OpenGL ES 3.2 (May 5, 2022)

8.23. TEXTURE IMAGE LOADS AND STORES

218

vector where the value of R, G, and B components is 0 and the value of the A component is undefined. Invalid image stores will have no effect. Invalid image atomics will not update any texture bound to the image unit and will return zero. An access is considered invalid if: • no texture is bound to the selected image unit; • the texture bound to the selected image unit is incomplete; • the texture level bound to the image unit is less than the base level or greater than the maximum level of the texture; • the internal format of the texture bound to the image unit is not found in table 8.26; • the internal format of the texture bound to the image unit is incompatible with the specified format according to table 8.27; • the texture bound to the image unit has layers, and the selected layer or cube map face doesn’t exist; • the selected texel τij or τijk doesn’t exist; Additionally, there are a number of cases where image load, store, size or atomic operations are considered to involve a format mismatch. In such cases, undefined values will be returned by image load, size and atomic operations and undefined values will be written by image store and atomic operations. A format mismatch will occur if: • the type of image variable used to access the image unit does not match the target of a texture bound to the image unit with layered set to TRUE; • the type of image variable used to access the image unit does not match the target corresponding to a single layer of a multi-layer texture target bound to the image unit with layered set to FALSE; • the type of image variable used to access the image unit has a component data type (floating-point, signed integer, unsigned integer) incompatible with the format of the image unit; • the format layout qualifier for an image variable used for an image load or atomic operation does not match the format of the image unit, according to table 8.26; or OpenGL ES 3.2 (May 5, 2022)

8.23. TEXTURE IMAGE LOADS AND STORES

219

• the image variable used for an image store has a format layout qualifier, and that qualifier does not match the format of the image unit, according to table 8.26. Accesses to textures bound to image units do format conversions based on the format argument specified when the image is bound. Loads always return a value as a vec4, ivec4, or uvec4, and stores always take the source data as a vec4, ivec4, or uvec4. Data are converted to/from the specified format according to the process described for a TexImage2D or ReadPixels command with format and type as RGBA and FLOAT for vec4 data, as RGBA_INTEGER and INT for ivec4 data, or as RGBA_INTEGER and UNSIGNED_INT for uvec4 data, respectively. Unused components are filled in with (0, 0, 0, 1) (where 0 and 1 are either floatingpoint or integer values, depending on the format). Any image variable used for shader loads or atomic operations must be declared with a format layout qualifier matching the format of its associated image unit, as enumerated in table 8.26. Otherwise, the access is considered to involve a format mismatch, as described above6 . Image Unit Format

Format Qualifer

RGBA32F RGBA16F R32F RGBA32UI RGBA16UI RGBA8UI R32UI RGBA32I RGBA16I RGBA8I R32I RGBA8 RGBA8_SNORM

rgba32f rgba16f r32f rgba32ui rgba16ui rgba8ui r32ui rgba32i rgba16i rgba8i r32i rgba8 rgba8_snorm

Table 8.26: Supported image unit formats, with equivalent format layout qualifiers.

6 The OpenGL Specification does not require that format qualifiers be declared for image variables used exclusively for image stores, unlike this Specification. This is an intentional behavior difference.

OpenGL ES 3.2 (May 5, 2022)

8.23. TEXTURE IMAGE LOADS AND STORES

220

When a texture is bound to an image unit, the format parameter for the image unit need not exactly match the texture internal format as long as the formats are considered compatible. A pair of formats is considered to match in size if the corresponding entries in the Size column of table 8.27 are identical. A pair of formats is considered to match by class if the corresponding entries in the Class column of table 8.27 are identical. For textures allocated by the GL, an image unit format is compatible with a texture internal format if they match by size. For textures allocated outside the GL, format compatibility is determined by matching by size or by class, in an implementation dependent manner. The matching criterion used for a given texture may be determined by calling GetTexParameter with pname set to IMAGE_FORMAT_COMPATIBILITY_TYPE, with return values of IMAGE_FORMAT_COMPATIBILITY_BY_SIZE and IMAGE_FORMAT_COMPATIBILITY_BY_CLASS, specifying matches by size and class, respectively. When the format associated with an image unit does not exactly match the internal format of the texture bound to the image unit, image loads, stores and atomic operations re-interpret the memory holding the components of an accessed texel according to the format of the image unit. The re-interpretation for image loads and the read portion of atomic operations is performed as though data were copied from the texel of the bound texture to a similar texel represented in the format of the image unit. Similarly, the re-interpretation for image stores and the write portion of atomic operations is performed as though data were copied from a texel represented in the format of the image unit to the texel in the bound texture. In both cases, this copy operation would be performed by: • reading the texel from the source format to scratch memory according to the process described for ReadPixels (see section 16), using default pixel storage modes and format and type parameters corresponding to the source format in table 8.27; and • writing the texel from scratch memory to the destination format according to the process described for TexSubImage3D (see section 8.6), using default pixel storage modes and format and type parameters corresponding to the destination format in table 8.27. Image Format RGBA32F

Size 128

Class Pixel format 4x32 RGBA (Continued on next page)

OpenGL ES 3.2 (May 5, 2022)

Pixel type FLOAT

8.23. TEXTURE IMAGE LOADS AND STORES

221

Texel sizes, compatibility classes ... (continued) Image Format Size Class Pixel format Pixel type RGBA16F 64 4x16 RGBA HALF_FLOAT R32F 32 1x32 RED FLOAT RGBA32UI 128 4x32 RGBA_INTEGER UNSIGNED_INT RGBA16UI 64 4x16 RGBA_INTEGER UNSIGNED_SHORT RGBA8UI 32 4x8 RGBA_INTEGER UNSIGNED_BYTE R32UI 32 1x32 RED_INTEGER UNSIGNED_INT RGBA32I 128 4x32 RGBA_INTEGER INT RGBA16I 64 4x16 RGBA_INTEGER SHORT RGBA8I 32 4x8 RGBA_INTEGER BYTE R32I 32 1x32 RED_INTEGER INT RGBA8 32 4x8 RGBA UNSIGNED_BYTE RGBA8_SNORM 32 4x8 RGBA BYTE Table 8.27: Texel sizes, compatibility classes, and pixel format/type combinations for each image format.

Implementations may support a limited combined number of image units, shader storage blocks, and active fragment shader outputs (see section 14). A link error will be generated if the sum of the number of active image uniforms used in all shaders, the number of active shader storage blocks, and the number of active fragment shader outputs exceeds the implementation-dependent value of MAX_COMBINED_SHADER_OUTPUT_RESOURCES.

8.23.1

Image Unit Queries

The state required for each image unit is summarized in table 21.33 and may be queried using the indexed query commands in that table. The initial values of image unit state are described above for BindImageTexture.

OpenGL ES 3.2 (May 5, 2022)

Chapter 9

Framebuffers and Framebuffer Objects As described in chapter 1 and section 2.1, the GL renders into (and reads values from) a framebuffer. Initially, the GL uses the window-system provided default framebuffer. The storage, dimensions, allocation, and format of the images attached to this framebuffer are managed entirely by the window system. Consequently, the state of the default framebuffer, including its images, can not be changed by the GL, nor can the default framebuffer be deleted by the GL. This chapter begins with an overview of the structure and contents of the framebuffer in section 9.1, followed by describing the commands used to create, destroy, and modify the state and attachments of application-created framebuffer objects which may be used instead of the default framebuffer.

9.1

Framebuffer Overview

The framebuffer consists of a set of pixels arranged as a two-dimensional array. For purposes of this discussion, each pixel in the framebuffer is simply a set of some number of bits. The number of bits per pixel may vary depending on the GL implementation, the type of framebuffer selected, and parameters specified when the framebuffer was created. Creation and management of the default framebuffer is outside the scope of this specification, while creation and management of framebuffer objects is described in detail in section 9.2. Corresponding bits from each pixel in the framebuffer are grouped together into a bitplane; each bitplane contains a single bit from each pixel. These bitplanes are grouped into several logical buffers. These are the color, depth, and stencil 222

9.1. FRAMEBUFFER OVERVIEW

223

buffers. The color buffer actually consists of a number of buffers, and these color buffers serve related but slightly different purposes depending on whether the GL is bound to the default framebuffer or a framebuffer object. For the default framebuffer, the color buffers are the front and the back buffers. Typically the contents of the front buffer are displayed on a color monitor while the contents of the back buffers are invisible; the GL draws to and reads from the back buffer. All color buffers must have the same number of bitplanes, although an implementation or context may choose not to provide back buffers. Further, an implementation or context may choose not to provide depth or stencil buffers. If no default framebuffer is associated with the GL context, the framebuffer is incomplete except when a framebuffer object is bound (see sections 9.2 and 9.4). Framebuffer objects are not visible, and do not have any of the color buffers present in the default framebuffer. Instead, the buffers of an framebuffer object are specified by attaching individual textures or renderbuffers (see section 9) to a set of attachment points. A framebuffer object has an array of color buffer attachment points, numbered zero through n, a depth buffer attachment point, and a stencil buffer attachment point. In order to be used for rendering, a framebuffer object must be complete, as described in section 9.4. Not all attachments of a framebuffer object need to be populated. Each pixel in a color buffer consists of up to four color components. The four color components are named R, G, B, and A, in that order; color buffers are not required to have all four color components. R, G, B, and A components may be represented as signed or unsigned normalized fixed-point, floating-point, or signed or unsigned integer values; all components must have the same representation. Each pixel in a depth buffer consists of a single unsigned integer value in the format described in section 12.6.1 or a floating-point value. Each pixel in a stencil buffer consists of a single unsigned integer value. The number of bitplanes in the color, depth, and stencil buffers is dependent on the currently bound framebuffer. For the default framebuffer, the number of bitplanes is fixed. For framebuffer objects, the number of bitplanes in a given logical buffer may change if the image attached to the corresponding attachment point changes. The GL has two active framebuffers; the draw framebuffer is the destination for rendering operations, and the read framebuffer is the source for readback operations. The same framebuffer may be used for both drawing and reading. Section 9.2 describes the mechanism for controlling framebuffer usage.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

224

The default framebuffer is initially used as the draw and read framebuffer 1 , and the initial state of all provided bitplanes is undefined. The format and encoding of buffers in the draw and read framebuffers can be queried as described in section 9.2.3.

9.2

Binding and Managing Framebuffer Objects

Framebuffer objects encapsulate the state of a framebuffer in a similar manner to the way texture objects encapsulate the state of a texture. In particular, a framebuffer object encapsulates state necessary to describe a collection of color, depth, and stencil logical buffers (other types of buffers are not allowed). For each logical buffer, a framebuffer-attachable image can be attached to the framebuffer to store the rendered output for that logical buffer. Examples of framebuffer-attachable images include texture images and renderbuffer images. Renderbuffers are described further in section 9.2.4 By allowing the images of a renderbuffer to be attached to a framebuffer, the GL provides a mechanism to support off-screen rendering. Further, by allowing the images of a texture to be attached to a framebuffer, the GL provides a mechanism to support render to texture. The default framebuffer for rendering and readback operations is provided by the window system. In addition, named framebuffer objects can be created and operated upon. The name space for framebuffer objects is the unsigned integers, with zero reserved by the GL for the default framebuffer. A framebuffer object is created by binding an unused name (which may be created by GenFramebuffers (see below)) to DRAW_FRAMEBUFFER or READ_FRAMEBUFFER. The binding is effected by calling void BindFramebuffer( enum target, uint framebuffer ); with target set to the desired framebuffer target and framebuffer set to the framebuffer object name. The resulting framebuffer object is a new state vector, comprising all the state and with the same initial values listed in table 21.15, as well as one set of the state values listed in table 21.16 for each attachment point of the framebuffer, with the same initial values. There are the value of MAX_COLOR_ATTACHMENTS color attachment points, plus one each for the depth and stencil attachment points. 1

The window system binding API may allow associating a GL context with two separate “default framebuffers” provided by the window system as the draw and read framebuffers, but if so, both default framebuffers are referred to by the name zero at their respective binding points.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

225

BindFramebuffer may also be used to bind an existing framebuffer object to DRAW_FRAMEBUFFER and/or READ_FRAMEBUFFER. If the bind is successful no change is made to the state of the newly bound framebuffer object, and any previous binding to target is broken. If a framebuffer object is bound to DRAW_FRAMEBUFFER or READ_FRAMEBUFFER, it becomes the target for rendering or readback operations, respectively, until it is deleted or another framebuffer object is bound to the corresponding bind point. Calling BindFramebuffer with target set to FRAMEBUFFER binds framebuffer to both the draw and read targets. Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. While a framebuffer object is bound, GL operations on the target to which it is bound affect the images attached to the bound framebuffer object, and queries of the target to which it is bound return state from the bound object. Queries of the values specified in tables 21.56 and 21.15 are derived from the framebuffer object bound to DRAW_FRAMEBUFFER, with the exception of those marked as properties of the read framebuffer, which are derived from the framebuffer object bound to READ_FRAMEBUFFER. The initial state of DRAW_FRAMEBUFFER and READ_FRAMEBUFFER refers to the default framebuffer. In order that access to the default framebuffer is not lost, it is treated as a framebuffer object with the name of zero. The default framebuffer is therefore rendered to and read from while zero is bound to the corresponding targets. On some implementations, the properties of the default framebuffer can change over time (e.g., in response to window system events such as attaching the context to a new window system drawable.) Framebuffer objects (those with a non-zero name) differ from the default framebuffer in a few important ways. First and foremost, unlike the default framebuffer, framebuffer objects have modifiable attachment points for each logical buffer in the framebuffer. Framebuffer-attachable images can be attached to and detached from these attachment points, which are described further in section 9.2.2. Also, the size and format of the images attached to framebuffer objects are controlled entirely within the GL interface, and are not affected by window system events, such as pixel format selection, window resizes, and display mode changes. Additionally, when rendering to or reading from an application createdframebuffer object, • The pixel ownership test always succeeds. In other words, framebuffer obOpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

226

jects own all of their pixels. • There are no visible color buffer bitplanes. This means there is no color buffer corresponding to the back, front, left, or right color bitplanes. • The only color buffer bitplanes are the ones defined by the framebuffer attachment points named COLOR_ATTACHMENT0 through COLOR_ATTACHMENTn. Each COLOR_ATTACHMENTi adheres to COLOR_ATTACHMENTi = COLOR_ATTACHMENT0 + i2 . • The only depth buffer bitplanes are the ones defined by the framebuffer attachment point DEPTH_ATTACHMENT. • The only stencil buffer bitplanes are the ones defined by the framebuffer attachment point STENCIL_ATTACHMENT. • If the attachment sizes are not all identical, the results of rendering are defined only within the largest area that can fit in all of the attachments. This area is defined as the intersection of rectangles having a lower left of (0, 0) and an upper right of (width, height) for each attachment. Contents of attachments outside this area are undefined after execution of a rendering command (as defined in section 2.4). If there are no attachments, rendering will be limited to a rectangle having a lower left of (0, 0) and an upper right of (width, height), where width and height are the framebuffer object’s default width and height. • If the number of layers of each attachment are not all identical, rendering will be limited to the smallest number of layers of any attachment. If there are no attachments, the number of layers will be taken from the framebuffer object’s default layer count. The command void GenFramebuffers( sizei n, uint *framebuffers ); returns n previously unused framebuffer object names in framebuffers. These names are marked as used, for the purposes of GenFramebuffers only, but they acquire state and type only when they are first bound. 2

The header files define tokens COLOR_ATTACHMENTi for i in the range [0, 31]. Most implementations support fewer than 32 color attachments, and it is an INVALID_OPERATION error to pass an unsupported attachment name to a command accepting color attachment names.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

227

Errors An INVALID_VALUE error is generated if n is negative. Framebuffer objects are deleted by calling void DeleteFramebuffers( sizei n, const uint *framebuffers ); framebuffers contains n names of framebuffer objects to be deleted. After a framebuffer object is deleted, it has no attachments, and its name is again unused. If a framebuffer that is currently bound to one or more of the targets DRAW_FRAMEBUFFER or READ_FRAMEBUFFER is deleted, it is as though BindFramebuffer had been executed with the corresponding target and framebuffer zero. Unused names in framebuffers that have been marked as used for the purposes of GenFramebuffers are marked as unused again. Unused names in framebuffers are silently ignored, as is the value zero. Errors An INVALID_VALUE error is generated if n is negative. The command boolean IsFramebuffer( uint framebuffer ); returns TRUE if framebuffer is the name of an framebuffer object. If framebuffer is zero, or if framebuffer is a non-zero value that is not the name of a framebuffer object, IsFramebuffer returns FALSE. The names bound to the draw and read framebuffer bindings can be queried by calling GetIntegerv with the symbolic constants DRAW_FRAMEBUFFER_BINDING and READ_FRAMEBUFFER_BINDING, respectively. FRAMEBUFFER_BINDING is equivalent to DRAW_FRAMEBUFFER_BINDING.

9.2.1

Framebuffer Object Parameters

Parameters of a framebuffer object are set using the command void FramebufferParameteri( enum target, enum pname, int param );

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

228

The framebuffer object is that which is bound to target. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. pname specifies the parameter of the framebuffer object bound to target to set. When a framebuffer has one or more attachments, the width, height, layer count (see section 9.8), sample count, and sample location pattern of the framebuffer are derived from the properties of the framebuffer attachments. When the framebuffer has no attachments, these properties are taken from framebuffer parameters. When pname is FRAMEBUFFER_DEFAULT_WIDTH, FRAMEBUFFER_DEFAULT_HEIGHT, FRAMEBUFFER_DEFAULT_SAMPLES, or FRAMEBUFFER_DEFAULT_LAYERS, FRAMEBUFFER_DEFAULT_FIXED_SAMPLE_LOCATIONS, param specifies the width, height, layer count, sample count, or sample location pattern, respectively, used when the framebuffer has no attachments. When a framebuffer has no attachments, it is considered layered (see section 9.8) if and only if the value of FRAMEBUFFER_DEFAULT_LAYERS is non-zero. It is considered to have sample buffers if and only if the value of FRAMEBUFFER_DEFAULT_SAMPLES is non-zero. The number of samples in the framebuffer is derived from the value of FRAMEBUFFER_DEFAULT_SAMPLES in an implementationdependent manner similar to that described for the command RenderbufferStorageMultisample (see section 9.2.4). If the framebuffer has sample buffers and the value of FRAMEBUFFER_DEFAULT_FIXED_SAMPLE_LOCATIONS is non-zero, it is considered to have a fixed sample location pattern as described for TexStorage2DMultisample (see section 8.8). Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_ENUM error is generated if pname is not FRAMEBUFFER_DEFAULT_WIDTH, FRAMEBUFFER_DEFAULT_HEIGHT, FRAMEBUFFER_DEFAULT_LAYERS, FRAMEBUFFER_DEFAULT_SAMPLES, or FRAMEBUFFER_DEFAULT_FIXED_SAMPLE_LOCATIONS. An INVALID_VALUE error is generated if pname is FRAMEBUFFER_DEFAULT_WIDTH, FRAMEBUFFER_DEFAULT_HEIGHT, FRAMEBUFFER_DEFAULT_LAYERS, or FRAMEBUFFER_DEFAULT_SAMPLES, and param is either negative or greater than the value of the corresponding implementationdependent limit MAX_FRAMEBUFFER_WIDTH, MAX_FRAMEBUFFER_HEIGHT, MAX_FRAMEBUFFER_LAYERS, or MAX_FRAMEBUFFER_SAMPLES, respectively.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

229

An INVALID_OPERATION error is generated if the default framebuffer is bound to target.

9.2.2

Attaching Images to Framebuffer Objects

Framebuffer-attachable images may be attached to, and detached from, framebuffer objects. In contrast, the image attachments of the default framebuffer may not be changed by the GL. A single framebuffer-attachable image may be attached to multiple framebuffer objects, potentially avoiding some data copies, and possibly decreasing memory consumption. For each logical buffer, a framebuffer object stores a set of state which defines the logical buffer’s attachment point. The attachment point state contains enough information to identify the single image attached to the attachment point, or to indicate that no image is attached. The per-logical buffer attachment point state is listed in table 21.16. There are several types of framebuffer-attachable images: • The image of a renderbuffer object, which is always two-dimensional. • A single level of a two-dimensional or two-dimensional multisample texture. • A single face of a cube map texture level, which is treated as a twodimensional image. • A single layer of a two-dimensional array texture, two-dimensional multisample array texture, or three-dimensional texture, which is treated as a two-dimensional image. • A single layer-face of a cube map array texture, which is treated as a twodimensional image. 9.2.2.1

Layered Images

Additionally, an entire level of a three-dimensional, cube map, cube map array, two-dimensional array, or two-dimensional multisample array texture can be attached to an attachment point. Such attachments are treated as an array of twodimensional images, arranged in layers, and the corresponding attachment point is considered to be layered (also see section 9.8).

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

9.2.3

230

Framebuffer Object Queries

Parameters of a framebuffer object may be queried with the command void GetFramebufferParameteriv( enum target, enum pname, int *params ); The framebuffer object is that which is bound to target. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. pname specifies the parameter of the framebuffer object bound to target to get. pname may be one of FRAMEBUFFER_DEFAULT_WIDTH, FRAMEBUFFER_DEFAULT_HEIGHT, FRAMEBUFFER_DEFAULT_LAYERS, FRAMEBUFFER_DEFAULT_SAMPLES, or FRAMEBUFFER_DEFAULT_FIXED_SAMPLE_LOCATIONS, indicating one of the corresponding parameters set with FramebufferParameteri (see section 9.2.1). These values may only be queried from a framebuffer object, not from a default framebuffer. The value of parameter pname for the framebuffer object is returned in params. Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_ENUM error is generated if pname is not one of the valid values listed above. An INVALID_OPERATION error is generated if the default framebuffer is bound to target. Attachments of a framebuffer object or buffers of a default framebuffer may be queried with the commands void GetFramebufferAttachmentParameteriv( enum target, enum attachment, enum pname, int *params ); The framebuffer object is that which is bound to target. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. If the default framebuffer is bound to target, then attachment must be BACK, identifying the color buffer; DEPTH, identifying the depth buffer; or STENCIL, identifying the stencil buffer. Otherwise, attachment must be one of the attachment points of the framebuffer listed in table 9.1. OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

231

If attachment is DEPTH_STENCIL_ATTACHMENT, the same object must be bound to both the depth and stencil attachment points of the framebuffer object, and information about that object is returned. Upon successful return from GetFramebufferAttachmentParameteriv, if pname is FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE, then params will contain one of NONE, FRAMEBUFFER_DEFAULT, TEXTURE, or RENDERBUFFER, identifying the type of object which contains the attached image. Other values accepted for pname depend on the type of object, as described below. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is NONE, then one of the following conditions is true: • no framebuffer is bound to target, or • the default framebuffer is bound, and – attachment is DEPTH or STENCIL, and the number of depth or stencil bits, respectively, is zero; or, – attachment does not indicate one of the color buffers allocated to the default framebuffer. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is NONE, querying pname FRAMEBUFFER_ATTACHMENT_OBJECT_NAME will return zero, and the only other valid query is FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is not NONE, these queries apply to all other framebuffer types: • If pname is FRAMEBUFFER_ATTACHMENT_RED_SIZE, FRAMEBUFFER_ATTACHMENT_GREEN_SIZE, FRAMEBUFFER_ATTACHMENT_BLUE_SIZE, FRAMEBUFFER_ATTACHMENT_ALPHA_SIZE, FRAMEBUFFER_ATTACHMENT_DEPTH_SIZE, or FRAMEBUFFER_ATTACHMENT_STENCIL_SIZE, then params will contain the number of bits in the corresponding red, green, blue, alpha, depth, or stencil component of the specified attachment. If the requested component is not present in attachment, or if no data storage or texture image has been specified for the attachment, then params will contain zero. • If pname is FRAMEBUFFER_ATTACHMENT_COMPONENT_TYPE, params will contain the format of components of the specified attachment, one of FLOAT, INT, UNSIGNED_INT, SIGNED_NORMALIZED, or UNSIGNED_NORMALIZED for floating-point, signed integer, unsigned integer, signed normalized fixedpoint, or unsigned normalized fixed-point components respectively. If OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

232

no data storage or texture image has been specified for the attachment, params will contain NONE. This query cannot be performed for a combined depth+stencil attachment, since it does not have a single format. • If pname is FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING, params will contain the encoding of components of the specified attachment, one of LINEAR or SRGB for linear or sRGB-encoded components, respectively. Only color buffer components may be sRGB-encoded; such components are treated as described in sections 15.1.5 and 15.1.6. For the default framebuffer, color encoding is determined by the implementation. For framebuffer objects, components are sRGB-encoded if the internal format of a color attachment is one of the color-renderable SRGB formats described in section 8.21. If attachment is not a color attachment, or no data storage or texture image has been specified for the attachment, params will contain the value LINEAR. If

the

value RENDERBUFFER, then

of

FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE

is

• If pname is FRAMEBUFFER_ATTACHMENT_OBJECT_NAME, params will contain the name of the renderbuffer object which contains the attached image. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is TEXTURE, then • If pname is FRAMEBUFFER_ATTACHMENT_OBJECT_NAME, then params will contain the name of the texture object which contains the attached image. • If pname is FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL, then params will contain the mipmap level of the texture object which contains the attached image. • If pname is FRAMEBUFFER_ATTACHMENT_TEXTURE_CUBE_MAP_FACE and the texture object named FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is a cube map texture, then params will contain the cube map face of the cubemap texture object which contains the attached image. Otherwise params will contain the value zero. • If pname is FRAMEBUFFER_ATTACHMENT_LAYERED, then params will contain TRUE if the attachment point is layered (see section 9.2.2.1). Otherwise, params will contain FALSE.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

233

• If pname is FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER and the value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is the name of a threedimensional texture or a two-dimensional array texture, and the value of FRAMEBUFFER_ATTACHMENT_LAYERED is FALSE, then params will contain the texture layer which contains the attached image. Otherwise params will contain zero. Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated if the default framebuffer is bound to target and attachment is not BACK, DEPTH, or STENCIL. An INVALID_OPERATION error is generated if a framebuffer object is bound to target and attachment is COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if a framebuffer object is bound to target, attachment is not one of the attachments in table 9.1, and attachment is not COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated by any combinations of framebuffer type and pname not described above. An INVALID_OPERATION error is generated if the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is NONE and pname is not FRAMEBUFFER_ATTACHMENT_OBJECT_NAME or FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE. An INVALID_OPERATION error is generated if attachment is DEPTH_STENCIL_ATTACHMENT and pname is FRAMEBUFFER_ATTACHMENT_COMPONENT_TYPE. An INVALID_OPERATION error is generated if attachment is DEPTH_STENCIL_ATTACHMENT and different objects are bound to the depth and stencil attachment points of target. 9.2.3.1

Multisample Queries

The values of SAMPLE_BUFFERS and SAMPLES control whether and how multisampling is performed (see section 13.4). They are framebuffer-dependent constants derived from the attachments of a framebuffer object or the buffers of a default framebuffer, but may only be queried directly for the currently bound

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

234

draw framebuffer, by calling GetIntegerv with pname set to SAMPLE_BUFFERS or SAMPLES. While there is no API for querying the values of SAMPLE_BUFFERS and SAMPLES for a framebuffer object or default framebuffer which is not bound as the draw framebuffer, the effective values of these parameters exist, and are determined as defined in this section. These effective values are referred to in a number of places in the Specification. If a framebuffer object or default framebuffer is not framebuffer complete, as defined in section 9.4.2, then the effective values of SAMPLE_BUFFERS and SAMPLES are undefined. Otherwise, the effective value of SAMPLES is equal to the value of RENDERBUFFER_SAMPLES or TEXTURE_SAMPLES (depending on the type of the attached images), which must all have the same value. The effective value of SAMPLE_BUFFERS is one if SAMPLES is non-zero, and zero otherwise.

9.2.4

Renderbuffer Objects

A renderbuffer is a data storage object containing a single image of a renderable internal format. The commands described below allocate and delete a renderbuffer’s image, and attach a renderbuffer’s image to a framebuffer object. The name space for renderbuffer objects is the unsigned integers, with zero reserved by the GL. A renderbuffer object is created by binding a name (which may be created by GenRenderbuffers (see below)) to RENDERBUFFER. The binding is effected by calling void BindRenderbuffer( enum target, uint renderbuffer ); with target set to RENDERBUFFER and renderbuffer set to the renderbuffer object name. If renderbuffer is not zero, then the resulting renderbuffer object is a new state vector, initialized with a zero-sized memory buffer, and comprising all the state and with the same initial values listed in table 21.17. Any previous binding to target is broken. BindRenderbuffer may also be used to bind an existing renderbuffer object. If the bind is successful, no change is made to the state of the newly bound renderbuffer object, and any previous binding to target is broken. While a renderbuffer object is bound, GL operations on the target to which it is bound affect the bound renderbuffer object, and queries of the target to which a renderbuffer object is bound return state from the bound object.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

235

The name zero is reserved. A renderbuffer object cannot be created with the name zero. If renderbuffer is zero, then any previous binding to target is broken and the target binding is restored to the initial state. In the initial state, the reserved name zero is bound to RENDERBUFFER. There is no renderbuffer object corresponding to the name zero, so client attempts to modify or query renderbuffer state for the target RENDERBUFFER while zero is bound will generate GL errors, as described in section 9.2.6. The current RENDERBUFFER binding can be determined by calling GetIntegerv with the symbolic constant RENDERBUFFER_BINDING. Errors An INVALID_ENUM error is generated if target is not RENDERBUFFER. The command void GenRenderbuffers( sizei n, uint *renderbuffers ); returns n previously unused renderbuffer object names in renderbuffers. These names are marked as used, for the purposes of GenRenderbuffers only, but they acquire renderbuffer state only when they are first bound. Errors An INVALID_VALUE error is generated if n is negative. Renderbuffer objects are deleted by calling void DeleteRenderbuffers( sizei n, const uint *renderbuffers ); where renderbuffers contains n names of renderbuffer objects to be deleted. After a renderbuffer object is deleted, it has no contents, and its name is again unused. If a renderbuffer that is currently bound to RENDERBUFFER is deleted, it is as though BindRenderbuffer had been executed with the target RENDERBUFFER and name of zero. Additionally, special care must be taken when deleting a renderbuffer if the image of the renderbuffer is attached to a framebuffer object (see section 9.2.7). Unused names in renderbuffers that have been marked as used for the purposes of GenRenderbuffers are marked as unused again. Unused names in renderbuffers are silently ignored, as is the value zero.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

236

Errors An INVALID_VALUE error is generated if n is negative. The command boolean IsRenderbuffer( uint renderbuffer ); returns TRUE if renderbuffer is the name of a renderbuffer object. If renderbuffer is zero, or if renderbuffer is a non-zero value that is not the name of a renderbuffer object, IsRenderbuffer returns FALSE. The command void RenderbufferStorageMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height ); establishes the data storage, format, dimensions, and number of samples of a renderbuffer object’s image. target must be RENDERBUFFER. internalformat must be a sized internal format that is color-, depth-, or stencil-renderable, as defined in section 9.4. width and height are the dimensions in pixels of the renderbuffer. Upon success, RenderbufferStorageMultisample deletes any existing data store for the renderbuffer image and the contents of the data store after calling RenderbufferStorageMultisample are undefined. RENDERBUFFER_WIDTH is set to width, RENDERBUFFER_HEIGHT is set to height, and RENDERBUFFER_INTERNAL_FORMAT is set to internalformat. If samples is zero, then RENDERBUFFER_SAMPLES is set to zero. Otherwise samples represents a request for a desired minimum number of samples. Since different implementations may support different sample counts for multisampled rendering, the actual number of samples allocated for the renderbuffer image is implementation-dependent. However, the resulting value for RENDERBUFFER_SAMPLES is guaranteed to be greater than or equal to samples and no more than the next larger sample count supported by the implementation. A GL implementation may vary its allocation of internal component resolution based on any RenderbufferStorage parameter (except target), but the allocation and chosen internal format must not be a function of any other state and cannot be changed once they are established. Errors An INVALID_ENUM error is generated if target is not RENDERBUFFER.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS An INVALID_VALUE error is generated if samples, width, or height is negative. An INVALID_OPERATION error is generated if samples is greater than the maximum number of samples supported for internalformat (see GetInternalformativ in section 20.3). An INVALID_ENUM error is generated if internalformat is not a sized internal format that is color-, depth-, or stencil-renderable, as defined in section 9.4. An INVALID_VALUE error is generated if either width or height is greater than the value of MAX_RENDERBUFFER_SIZE. The command void RenderbufferStorage( enum target, enum internalformat, sizei width, sizei height ); is equivalent to calling RenderbufferStorageMultisample with samples equal to zero.

9.2.5

Required Renderbuffer Formats

Implementations are required to support the following sized and compressed internal formats. Requesting one of these sized internal formats for a renderbuffer will allocate at least the internal component sizes, and exactly the component types shown for that format in the corresponding table: • Color formats which are checked in the “Req. rend.” column of table 8.10. • Depth, depth+stencil, and stencil formats which are checked in the “Req. format” column of table 8.11. The required color formats for renderbuffers are a subset of the required formats for textures (see section 8.5.1). Implementations must support creation of renderbuffers in these required formats with the following numbers of multisamples: • For signed and unsigned integer formats, up to the value of MAX_INTEGER_SAMPLES, which must be at least one. • For formats RGBA16F, R32F, RG32F and RGBA32F, one sample. • For all other formats, up to the value of MAX_SAMPLES samples. OpenGL ES 3.2 (May 5, 2022)

237

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

9.2.6

238

Renderbuffer Object Queries

The command void GetRenderbufferParameteriv( enum target, enum pname, int *params ); returns information about a bound renderbuffer object. target must be RENDERBUFFER and pname must be one of the symbolic values in table 21.17. If pname is RENDERBUFFER_WIDTH, RENDERBUFFER_HEIGHT, RENDERBUFFER_INTERNAL_FORMAT, or RENDERBUFFER_SAMPLES, then params will contain the width in pixels, height in pixels, internal format, or number of samples, respectively, of the image of the renderbuffer currently bound to target. If pname is RENDERBUFFER_RED_SIZE, RENDERBUFFER_GREEN_SIZE, RENDERBUFFER_BLUE_SIZE, RENDERBUFFER_ALPHA_SIZE, RENDERBUFFER_DEPTH_SIZE, or RENDERBUFFER_STENCIL_SIZE, then params will contain the actual resolutions (not the resolutions specified when the image was defined) for the red, green, blue, alpha, depth, or stencil components, respectively, of the image of the renderbuffer currently bound to target. Errors An INVALID_ENUM error is generated if target is not RENDERBUFFER. An INVALID_ENUM error is generated if pname is not one of the renderbuffer state names in table 21.17. An INVALID_OPERATION error is generated if the renderbuffer currently bound to target is zero.

9.2.7

Attaching Renderbuffer Images to a Framebuffer

A renderbuffer can be attached as one of the logical buffers of a currently bound framebuffer object by calling void FramebufferRenderbuffer( enum target, enum attachment, enum renderbuffertarget, uint renderbuffer ); target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. attachment must be set to one of the attachment points of the framebuffer listed in table 9.1. OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

239

renderbuffertarget must be RENDERBUFFER and renderbuffer is zero or the name of a renderbuffer object of type renderbuffertarget to be attached to the framebuffer. If renderbuffer is zero, then the value of renderbuffertarget is ignored. If renderbuffer is not zero and if FramebufferRenderbuffer is successful, then the renderbuffer named renderbuffer will be used as the logical buffer identified by attachment of the framebuffer object currently bound to target. The value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for the specified attachment point is set to RENDERBUFFER and the value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is set to renderbuffer. All other state values of the attachment point specified by attachment are set to their default values listed in table 21.16. No change is made to the state of the renderbuffer object and any previous attachment to the attachment logical buffer of the framebuffer object bound to framebuffer target is broken. If the attachment is not successful, then no change is made to the state of either the renderbuffer object or the framebuffer object. Calling FramebufferRenderbuffer with the renderbuffer name zero will detach the image, if any, identified by attachment, in the framebuffer object currently bound to target. All state values of the attachment point specified by attachment in the object bound to target are set to their default values listed in table 21.16. Setting attachment to the value DEPTH_STENCIL_ATTACHMENT is a special case causing both the depth and stencil attachments of the framebuffer object to be set to renderbuffer, which should have base internal format DEPTH_STENCIL. If a renderbuffer object is deleted while its image is attached to one or more attachment points in a currently bound framebuffer object, then it is as if FramebufferRenderbuffer had been called, with a renderbuffer of zero, for each attachment point to which this image was attached in that framebuffer object. In other words, the renderbuffer image is first detached from all attachment points in that framebuffer object. Note that the renderbuffer image is specifically not detached from any non-bound framebuffers. Detaching the image from any non-bound framebuffers is the responsibility of the application. Name of attachment COLOR_ATTACHMENTi (see caption) DEPTH_ATTACHMENT STENCIL_ATTACHMENT DEPTH_STENCIL_ATTACHMENT

Table 9.1: Framebuffer attachment points. i in COLOR_ATTACHMENTi may range from zero to the value of MAX_COLOR_ATTACHMENTS minus one.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

240

Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated if attachment is COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if attachment is not one of the attachments in table 9.1, and attachment is not COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if renderbuffertarget is not RENDERBUFFER. An INVALID_OPERATION error is generated if renderbuffer is not zero or the name of an existing renderbuffer object of type renderbuffertarget. An INVALID_OPERATION error is generated if zero is bound to target.

9.2.8

Attaching Texture Images to a Framebuffer

The GL supports copying the rendered contents of the framebuffer into the images of a texture object through the use of the routines CopyTexImage* and CopyTexSubImage*. Additionally, the GL supports rendering directly into the images of a texture object. To render directly into a texture image, a specified level of a texture object can be attached as one of the logical buffers of a framebuffer object with the command void FramebufferTexture( enum target, enum attachment, uint texture, int level ); The framebuffer object is that which is bound to target. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. attachment must be one of the attachment points of the framebuffer listed in table 9.1. If texture is non-zero, the specified mipmap level of the texture object named texture is attached to the framebuffer attachment point named by attachment. If texture is the name of one of the types of textures described in the definition of layered textures in section 9.2.2.1, the texture level attached to the framebuffer attachment point is an array of images, and the framebuffer attachment is considered layered.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

241

Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_OPERATION error is generated if attachment is COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if attachment is not one of the attachments in table 9.1, and attachment is not COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_VALUE error is generated if texture is not zero and is not the name of a texture object, or if level is not a supported texture level for texture. An INVALID_OPERATION error is generated if texture is the name of a buffer texture. Additionally, a specified image from a texture object can be attached as one of the logical buffers of a framebuffer object with the command void FramebufferTexture2D( enum target, enum attachment, enum textarget, uint texture, int level ); The framebuffer object is that which is bound to target. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. attachment must be one of the attachment points of the framebuffer listed in table 9.1. If texture is not zero, then texture must either name an existing two-dimensional texture object and textarget must be TEXTURE_2D, texture must name an existing cube map texture and textarget must be one of the cube map face targets from table 8.20, or texture must name an existing multisample texture and textarget must be TEXTURE_2D_MULTISAMPLE. level specifies the mipmap level of the texture image to be attached to the framebuffer, and must satisfy the following conditions: • If texture refers to an immutable-format texture, level must be greater than or equal to zero and smaller than the value of TEXTURE_IMMUTABLE_LEVELS for texture. • If textarget is TEXTURE_2D_MULTISAMPLE, then level must be zero.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

242

• If textarget is one of the cube map face targets from table 8.20, then level must be greater than or equal to zero and less than or equal to log2 of the value of MAX_CUBE_MAP_TEXTURE_SIZE. • If textarget is TEXTURE_2D, level must be greater than or equal to zero and no larger than log2 of the value of MAX_TEXTURE_SIZE.

Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated if attachment is COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if attachment is not one of the attachments in table 9.1, and attachment is not COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_OPERATION error is generated if zero is bound to target. An INVALID_VALUE error is generated if texture is not zero and level is not a supported texture level for textarget, as described above. An INVALID_OPERATION error is generated if texture is not zero and textarget is not one of TEXTURE_2D, TEXTURE_2D_MULTISAMPLE, or one of the cube map face targets from table 8.20. An INVALID_OPERATION error is generated if texture is not zero, and does not name an existing texture object of type matching textarget, as described above. An INVALID_OPERATION error is generated if texture is the name of a buffer texture. A single layer of a three-dimensional or array texture object can be attached as one of the logical buffers of a framebuffer object with the command void FramebufferTextureLayer( enum target, enum attachment, uint texture, int level, int layer ); This command operates similarly to FramebufferTexture2D, except for the additional layer argument which selects a layer of the texture object to attach. layer specifies the layer of a two-dimensional image within texture, except for cube map and cube map array textures. For cube map textures, layer is translated into a cube map face as described in table 8.24. For cube map array textures, layer OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

243

is translated into an array layer and a cube map face as described for layer-face numbers in section 8.5.3. level specifies the mipmap level of the texture image to be attached to the framebuffer, and must satisfy the following conditions: • If texture refers to an immutable-format texture, level must be greater than or equal to zero and smaller than the value of TEXTURE_IMMUTABLE_LEVELS for texture. • If texture is a three-dimensional texture, then level must be greater than or equal to zero and less than or equal to log2 of the value of MAX_3D_TEXTURE_SIZE. • If texture is a two-dimensional array texture, then level must be greater than or equal to zero and less than or equal to log2 of the value of MAX_TEXTURE_SIZE. • If texture is a two-dimensional multisample array texture, then level must be zero. Errors In addition to the corresponding errors for FramebufferTexture when called with the same parameters (other than layer): An INVALID_VALUE error is generated if texture is a three-dimensional texture, and layer is larger than the value of MAX_3D_TEXTURE_SIZE minus one. An INVALID_VALUE error is generated if texture is a two-dimensional array, two-dimensional multisample array or cube map array texture, and layer is larger than the value of MAX_ARRAY_TEXTURE_LAYERS minus one (see section 9.8). An INVALID_VALUE error is generated if texture is non-zero and layer is negative. An INVALID_OPERATION error is generated if texture is non-zero and is not the name of a three-dimensional, two-dimensional multisample array, two-dimensional array, cube map, or cube map array texture. An INVALID_VALUE error is generated if texture is not zero and level is not a supported texture level for texture, as described above. Unlike FramebufferTexture2D, no textarget parameter is accepted.

OpenGL ES 3.2 (May 5, 2022)

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

244

If texture is non-zero and the command does not result in an error, the framebuffer attachment state corresponding to attachment is updated as in FramebufferTexture2D commands, except that the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER is set to layer. 9.2.8.1

Effects of Attaching a Texture Image

The remaining comments in this section apply to all forms of FramebufferTexture*. If texture is zero, any image or array of images attached to the attachment point named by attachment is detached. Any additional parameters (level, textarget, and/or layer) are ignored when texture is zero. All state values of the attachment point specified by attachment are set to their default values listed in table 21.16. If texture is not zero, and if FramebufferTexture* is successful, then the specified texture image will be used as the logical buffer identified by attachment of the framebuffer object currently bound to target. State values of the specified attachment point are set as follows: • The value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is set to TEXTURE. • The value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is set to texture. • The value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL is set to level. • If FramebufferTexture2D is called and texture is a cube map texture, then the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_CUBE_MAP_FACE is set to textarget; otherwise it is set to the default value (NONE). • If FramebufferTextureLayer is called, then the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER is set to layer; otherwise it is set to zero. • If FramebufferTexture is called and texture is the name of one of the types of textures described in the definition of layered textures in section 9.2.2.1, the value of FRAMEBUFFER_ATTACHMENT_LAYERED is set to TRUE; otherwise it is set to FALSE. All other state values of the attachment point specified by attachment are set to their default values listed in table 21.16. No change is made to the state of the texture object, and any previous attachment to the attachment logical buffer of the framebuffer object bound to framebuffer target is broken. If the attachment is not

OpenGL ES 3.2 (May 5, 2022)

9.3. FEEDBACK LOOPS BETWEEN TEXTURES AND THE FRAMEBUFFER245 successful, then no change is made to the state of either the texture object or the framebuffer object. Setting attachment to the value DEPTH_STENCIL_ATTACHMENT is a special case causing both the depth and stencil attachments of the framebuffer object to be set to texture. texture must have base internal format DEPTH_STENCIL, or the depth and stencil framebuffer attachments will be incomplete (see section 9.4.1). If a texture object is deleted while its image is attached to one or more attachment points in a currently bound framebuffer object, then it is as if FramebufferTexture* had been called, with a texture of zero, for each attachment point to which this image was attached in that framebuffer object. In other words, the texture image is first detached from all attachment points in that framebuffer object. Note that the texture image is specifically not detached from any non-bound framebuffer objects. Detaching the texture image from any non-bound framebuffer objects is the responsibility of the application.

9.3

Feedback Loops Between Textures and the Framebuffer

A feedback loop may exist when the data store of a texture object is used as both the source and destination of a GL operation. When a feedback loop exists, undefined behavior results. This section describes rendering feedback loops (see section 8.14.2.1) and texture copying feedback loops (see section 8.6.1) in more detail.

9.3.1

Rendering Feedback Loops

The mechanisms for attaching textures to a framebuffer object do not prevent a twodimensional texture level, a face of a cube map texture level, or a layer of a threedimensional texture from being attached to the draw framebuffer while the same texture is bound to a texture unit. While this condition holds, texturing operations accessing that image will produce undefined results, as described at the end of section 8.14. Conditions resulting in such undefined behavior are defined in more detail below. Such undefined texturing operations are likely to leave the final results of fragment processing operations undefined, and should be avoided. Special precautions need to be taken to avoid attaching a texture image to the currently bound draw framebuffer object while the texture object is currently bound and enabled for texturing. Doing so could lead to the creation of a rendering feedback loop between the writing of pixels by GL rendering operations and the simultaneous reading of those same pixels when used as texels in the currently bound OpenGL ES 3.2 (May 5, 2022)

9.3. FEEDBACK LOOPS BETWEEN TEXTURES AND THE FRAMEBUFFER246 texture. In this scenario, the framebuffer will be considered framebuffer complete (see section 9.4), but the values of fragments rendered while in this state will be undefined. The values of texture samples may be undefined as well, as described under “Rendering Feedback Loops” in section 8.14.2.1 Specifically, the values of rendered fragments are undefined if all of the following conditions are true: • an image from texture object T is attached to the currently bound draw framebuffer object at attachment point A • the texture object T is currently bound to a texture unit U, and • the current programmable vertex and/or fragment processing state makes it possible (see below) to sample from the texture object T bound to texture unit U while either of the following conditions are true: • the value of TEXTURE_MIN_FILTER for texture object T is NEAREST or LINEAR, and the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL for attachment point A is equal to the value of TEXTURE_BASE_LEVEL for the texture object T • the value of TEXTURE_MIN_FILTER for texture object T is one of NEAREST_MIPMAP_NEAREST, NEAREST_MIPMAP_LINEAR, LINEAR_MIPMAP_NEAREST, or LINEAR_MIPMAP_LINEAR, and the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL for attachment point A is within the range specified by the current values of TEXTURE_BASE_LEVEL to q, inclusive, for the texture object T. q is defined in section 8.14.3. For the purpose of this discussion, it is possible to sample from the texture object T bound to texture unit U if the active fragment or vertex shader contains any instructions that might sample from the texture object T bound to U, even if those instructions might only be executed conditionally. Note that if TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL exclude any levels containing image(s) attached to the currently bound draw framebuffer object, then the above conditions will not be met (i.e., the above rule will not cause the values of rendered fragments to be undefined.) Also note that even if writing to a given texture is disabled (e.g. via the color, depth, or stencil writemasks) during rendering, a feedback loop will still occur.

OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

9.3.2

247

Texture Copying Feedback Loops

Similarly to rendering feedback loops, it is possible for a texture image to be attached to the currently bound read framebuffer object while the same texture image is the destination of a CopyTexImage* operation, as described under “Texture Copying Feedback Loops” in section 8.6.1. While this condition holds, a texture copying feedback loop between the writing of texels by the copying operation and the reading of those same texels when used as pixels in the read framebuffer may exist. In this scenario, the values of texels written by the copying operation will be undefined. Specifically, the values of copied texels are undefined if all of the following conditions are true: • an image from texture object T is attached to the currently bound read framebuffer object at attachment point A • the selected read buffer (see section 16.1.1) is attachment point A • T is bound to the texture target of a CopyTexImage* operation • the level argument of the copying operation selects the same image that is attached to A

9.4

Framebuffer Completeness

A framebuffer must be framebuffer complete to effectively be used as the draw or read framebuffer of the GL. The default framebuffer is always complete if it exists; however, if no default framebuffer exists (no window system-provided drawable is associated with the GL context), it is deemed to be incomplete. A framebuffer object is said to be framebuffer complete if all of its attached images, and all framebuffer parameters required to utilize the framebuffer for rendering and reading, are consistently defined and meet the requirements defined below. The rules of framebuffer completeness are dependent on the properties of the attached images, and on certain implementation-dependent restrictions. The internal formats of the attached images can affect the completeness of the framebuffer, so it is useful to first define the relationship between the internal format of an image and the attachment points to which it can be attached. • An internal format is color-renderable if it is one of the sized internal formats from table 8.10 whose “CR” (color-renderable) column is checked in that OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

248

table, or if it is unsized, non-floating-point format RGB or RGBA. No other formats, including compressed internal formats, are color-renderable. • An internal format is depth-renderable if it is one of the formats from table 8.11 whose base internal format is DEPTH_COMPONENT or DEPTH_STENCIL. No other formats are depth-renderable. • An internal format is stencil-renderable if it is one of the formats from table 8.11 whose base internal format is STENCIL_INDEX or DEPTH_STENCIL. No other formats are stencil-renderable.

9.4.1

Framebuffer Attachment Completeness

If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for the framebuffer attachment point attachment is not NONE, then it is said that a framebufferattachable image, named image, is attached to the framebuffer at the attachment point. image is identified by the state in attachment as described in section 9.2.2. The framebuffer attachment point attachment is said to be framebuffer attachment complete if the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE for attachment is NONE (i.e., no image is attached), or if all of the following conditions are true: • image is a component of an existing object with the name specified by the value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME, and of the type specified by the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE. • The width and height of image are greater than zero and less than or equal to the values of the implementation-dependent limits MAX_FRAMEBUFFER_WIDTH and MAX_FRAMEBUFFER_HEIGHT, respectively. • If image is a three-dimensional, two-dimensional array, or cube map array texture and the attachment is not layered, the selected layer is less than the depth (for three-dimensional) or layer count (otherwise) of the texture. • If image is a three-dimensional, two-dimensional array, or cube map array texture and the attachment is layered, the depth or layer count of the texture is less than or equal to the value of the implementation-dependent limit MAX_FRAMEBUFFER_LAYERS. • If image has multiple samples, its sample count is less than or equal to the value of the implementation-dependent limit MAX_FRAMEBUFFER_SAMPLES. OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

249

• If image is not an immutable-format texture, the selected level number is in the range [levelbase , q], where levelbase and q are as defined in section 8.14.3. • If image is not an immutable-format texture and the selected level is not levelbase , the texture must be mipmap complete. • If the attachment is layered and the selected level is levelbase , then the texture must be base complete. • If the attachment is layered and the selected level is not levelbase , then the texture must be mipmap complete. • If attachment is COLOR_ATTACHMENTi, then image must have a colorrenderable internal format. • If attachment is DEPTH_ATTACHMENT, then image must have a depthrenderable internal format. • If attachment is STENCIL_ATTACHMENT, then image must have a stencilrenderable internal format.

9.4.2

Whole Framebuffer Completeness

Each rule below is followed by an error token enclosed in { brackets }. The meaning of these errors is explained below and under “Effects of Framebuffer Completeness on Framebuffer Operations” in section 9.4.4. Note that the error token FRAMEBUFFER_INCOMPLETE_DIMENSIONS is included in the API for OpenGL ES 2.0 compatibility, but cannot be generated by an OpenGL ES 3.0 or later implementation. The framebuffer object bound to target is said to be framebuffer complete if all the following conditions are true: • if target is the default framebuffer, the default framebuffer exists. { FRAMEBUFFER_UNDEFINED } • All framebuffer attachment points are framebuffer attachment complete. { FRAMEBUFFER_INCOMPLETE_ATTACHMENT } • There is at least one image attached to the framebuffer, or the value of the framebuffer’s FRAMEBUFFER_DEFAULT_WIDTH and FRAMEBUFFER_DEFAULT_HEIGHT parameters are both non-zero. OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

250

{ FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT } • The combination of internal formats of the attached images does not violate an implementation-dependent set of restrictions. { FRAMEBUFFER_UNSUPPORTED } • Depth and stencil attachments, if present, are the same image. { FRAMEBUFFER_UNSUPPORTED } • The value of RENDERBUFFER_SAMPLES is the same for all attached renderbuffers; the value of TEXTURE_SAMPLES is the same for all attached textures; and, if the attached images are a mix of renderbuffers and textures, the value of RENDERBUFFER_SAMPLES matches the value of TEXTURE_SAMPLES. FRAMEBUFFER_INCOMPLETE_MULTISAMPLE

• The value of TEXTURE_FIXED_SAMPLE_LOCATIONS is the same for all attached textures; and, if the attached images are a mix of renderbuffers and textures, the value of TEXTURE_FIXED_SAMPLE_LOCATIONS must be TRUE for all attached textures. { FRAMEBUFFER_INCOMPLETE_MULTISAMPLE } • If any framebuffer attachment is layered, all populated attachments must be layered. Additionally, all populated color attachments must be from textures of the same target (three-dimensional, two-dimensional array twodimensional multisample array, cube map, or cube map array textures. { FRAMEBUFFER_INCOMPLETE_LAYER_TARGETS } The token in brackets after each clause of the framebuffer completeness rules specifies the return value of CheckFramebufferStatus (see below) that is generated when that clause is violated. If more than one clause is violated, it is implementation-dependent which value will be returned by CheckFramebufferStatus. Performing any of the following actions may change whether the framebuffer is considered complete or incomplete: • Binding to a different framebuffer with BindFramebuffer. OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

251

• Attaching an image to the framebuffer with FramebufferTexture* or FramebufferRenderbuffer. • Detaching an image from the framebuffer with FramebufferTexture* or FramebufferRenderbuffer. • Changing the internal format of a texture image that is attached to the framebuffer by calling TexImage*, TexStorage*, CopyTexImage*, or CompressedTexImage*. • Changing the internal format of a renderbuffer that is attached to the framebuffer by calling RenderbufferStorage*. • Deleting, with DeleteTextures or DeleteRenderbuffers, an object containing an image that is attached to a currently bound framebuffer object. • Associating a different window system-provided drawable, or no drawable, with the default framebuffer using a window system binding API such as those described in section 1.6.3. Although the GL defines a wide variety of internal formats for framebufferattachable images, such as texture images and renderbuffer images, some implementations may not support rendering to particular combinations of internal formats. If the combination of formats of the images attached to a framebuffer object are not supported by the implementation, then the framebuffer is not complete under the clause labeled FRAMEBUFFER_UNSUPPORTED. Implementations are required to support certain combinations of framebuffer internal formats as described under “Required Framebuffer Formats” in section 9.4.3. Because of the implementation-dependent clause of the framebuffer completeness test in particular, and because framebuffer completeness can change when the set of attached images is modified, it is strongly advised, though not required, that an application check to see if the framebuffer is complete prior to rendering. The status of the framebuffer object currently bound to target can be queried by calling enum CheckFramebufferStatus( enum target ); target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER.

or

A value is returned that identifies whether or not the framebuffer object or default framebuffer bound to target is complete when treated as a read or draw framebuffer (as determined by target). If the framebuffer object is complete, then OpenGL ES 3.2 (May 5, 2022)

9.4. FRAMEBUFFER COMPLETENESS

252

FRAMEBUFFER_COMPLETE is returned. Otherwise, the value returned is one of

the error codes defined as the start of section 9.4.2 identifying one of the rules of framebuffer completeness that is violated. If CheckFramebufferStatus generates an error, zero is returned. Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER.

9.4.3

Required Framebuffer Formats

Implementations must support framebuffer objects with up to MAX_COLOR_ATTACHMENTS color attachments, a depth attachment, and a stencil attachment. Each color attachment may be in any of the color-renderable formats described in section 9.4. The depth attachment may be in any of the required depth or combined depth+stencil formats described in sections 8.5.1 and 9.2.5, and the stencil attachment may be in any of the required stencil or combined depth+stencil formats. However, when both depth and stencil attachments are present, implementations must not support framebuffer objects where depth and stencil attachments refer to separate images.

9.4.4

Effects of Framebuffer Completeness on Framebuffer Operations

Errors An INVALID_FRAMEBUFFER_OPERATION error is generated by attempts to render to or read from a framebuffer which is not framebuffer complete. This error is generated regardless of whether fragments are actually read from or written to the framebuffer. For example, it is generated when a rendering command is called and the framebuffer is incomplete, even if RASTERIZER_DISCARD is enabled. An INVALID_FRAMEBUFFER_OPERATION error is generated by rendering commands (see section 2.4), and commands that read from the framebuffer such as ReadPixels, CopyTexImage*, and CopyTexSubImage*, if called while the framebuffer is not framebuffer complete.

OpenGL ES 3.2 (May 5, 2022)

9.5. MAPPING BETWEEN PIXEL AND ELEMENT IN ATTACHED IMAGE253

9.4.5

Effects of Framebuffer State on Framebuffer Dependent Values

The values of the state variables listed in table 21.56 may change when a change is made to the current framebuffer binding, to the state of the currently bound framebuffer object, or to an image attached to that framebuffer object. Most such state is dependent on the draw framebuffer (DRAW_FRAMEBUFFER_BINDING), but IMPLEMENTATION_COLOR_READ_TYPE and IMPLEMENTATION_COLOR_READ_FORMAT are dependent on the read framebuffer (READ_FRAMEBUFFER_BINDING). When the relevant framebuffer binding is zero, the values of the state variables listed in table 21.56 are implementation defined. When the relevant framebuffer binding is non-zero, if the currently bound framebuffer object is not framebuffer complete, then the values of the state variables listed in table 21.56 are undefined. When the relevant framebuffer binding is non-zero and the currently bound draw framebuffer object is framebuffer complete, then the values of the state variables listed in table 21.56 are completely determined by the relevant framebuffer binding, the state of the currently bound framebuffer object, and the state of the images attached to that framebuffer object. The values of RED_BITS, GREEN_BITS, BLUE_BITS, and ALPHA_BITS are defined only if all color attachments of the draw framebuffer have identical formats, in which case the color component depths of color attachment zero are returned. The values returned for DEPTH_BITS and STENCIL_BITS are the depth or stencil component depth of the corresponding attachment of the draw framebuffer, respectively. The actual sizes of the color, depth, or stencil bit planes can be obtained by querying an attachment point using GetFramebufferAttachmentParameteriv, or querying the object attached to that point. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE at a particular attachment point is RENDERBUFFER, the sizes may be determined by calling GetRenderbufferParameteriv as described in section 9.2.6.

9.5

Mapping between Pixel and Element in Attached Image

When DRAW_FRAMEBUFFER_BINDING is non-zero, an operation that writes to the framebuffer modifies the image attached to the selected logical buffer, and an operation that reads from the framebuffer reads from the image attached to the selected logical buffer. If the attached image is a renderbuffer image, then the window coordinates

OpenGL ES 3.2 (May 5, 2022)

9.6. CONVERSION TO FRAMEBUFFER-ATTACHABLE IMAGE COMPONENTS254 (xw , yw ) corresponds to the value in the renderbuffer image at the same coordinates. If the attached image is a texture image, then the window coordinates (xw , yw ) correspond to the texel (i, j, k) from figure 8.6 as follows: i = xw j = yw k = layer where layer is the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER for the selected logical buffer. For a two-dimensional texture, k and layer are irrelevant.

9.6

Conversion to Framebuffer-Attachable Image Components

When an enabled color value is written to the framebuffer while the draw framebuffer binding is non-zero, for each draw buffer the R, G, B, and A values are converted to internal components as described in table 8.8, according to the table row corresponding to the internal format of the framebuffer-attachable image attached to the selected logical buffer, and the resulting internal components are written to the image attached to logical buffer. The masking operations described in section 15.2.2 are also effective.

9.7

Conversion to RGBA Values

When a color value is read while the read framebuffer binding is non-zero, or is used as the source of blending while the draw framebuffer binding is non-zero, components of that color taken from the framebuffer-attachable image attached to the selected logical buffer are first converted to R, G, B, and A values according to table 14.1 and the internal format of the attached image.

9.8

Layered Framebuffers

A framebuffer is considered to be layered if it is complete and all of its populated attachments are layered, as described in section 9.2.8. When rendering to a layered framebuffer, each fragment generated by the GL is assigned a layer number. The layer number for a fragment is zero if OpenGL ES 3.2 (May 5, 2022)

9.8. LAYERED FRAMEBUFFERS

255

• geometry shaders are disabled, or • the current geometry shader does not statically assign a value to the built-in output variable gl_Layer. Otherwise, the layer for each point, line, or triangle emitted by the geometry shader is taken from the gl_Layer output of one of the vertices of the primitive. The vertex used is implementation-dependent and may be queried as described in section 11.3.4. To get defined results, all vertices of each primitive emitted should set the same value for gl_Layer. Since the EndPrimitive built-in function starts a new output primitive, defined results can be achieved if EndPrimitive is called between two vertices emitted with different layer numbers. A layer number written by a geometry shader has no effect if the framebuffer is not layered. When fragments are written to a layered framebuffer, the fragment’s layer number selects an image from the array of images at each attachment point to use for the stencil test (see section 15.1.2), depth buffer test (see section 15.1.3), and for blending and color buffer writes (see section 15.1.5). If the fragment’s layer number is negative, or greater than or equal to the minimum number of layers of any attachment, the effects of the fragment on the framebuffer contents are undefined. When the Clear or ClearBuffer* commands described in section 15.2.3 are used to clear a layered framebuffer attachment, all layers of the attachment are cleared. When commands such as ReadPixels read from a layered framebuffer, the image at layer zero of the selected attachment is always used to obtain pixel values. When cube map texture levels are attached to a layered framebuffer, there are six layers, numbered zero through five. Each layer number corresponds to a cube map face, as shown in table 8.24. When cube map array texture levels are attached to a layered framebuffer, the layer number corresponds to a layer-face. The layer-face is be translated into an array layer and a cube map face as described in section 8.23 for layer-face numbers passed to BindImageTexture.

OpenGL ES 3.2 (May 5, 2022)

Chapter 10

Vertex Specification and Drawing Commands Most geometric primitives are drawn by specifying a series of generic attribute sets corresponding to the vertices of a primitive using DrawArrays or one of the other drawing commands defined in section 10.5. Points, lines, polygons, and a variety of related geometric primitives (see section 10.1) can be drawn in this way. The process of specifying attributes of a vertex and passing them to a shader is referred to as transferring a vertex to the GL. Vertex Shader Processing and Vertex State Each vertex is specified with one or more generic vertex attributes. Each attribute is specified with one, two, three, or four scalar values. Generic vertex attributes can be accessed from within vertex shaders (see section 11.1) and used to compute values for consumption by later processing stages. Before vertex shader execution, the state required by a vertex is its generic vertex attributes. Vertex shader execution processes vertices producing a homogeneous vertex position and any outputs explicitly written by the vertex shader. Figure 10.1 shows the sequence of operations that builds a primitive (point, line segment, or polygon) from a sequence of vertices. After a primitive is formed, it is clipped to a clip volume. This may modify the primitive by altering vertex coordinates and vertex shader outputs. In the case of line and polygon primitives, clipping may insert new vertices into the primitive. The vertices defining a primitive to be rasterized have output variables associated with them.

256

257

Figure 10.1. Vertex processing and primitive assembly.

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

10.1

258

Primitive Types

A sequence of vertices is passed to the GL using DrawArrays or one of the other drawing commands defined in section 10.5. There is no limit to the number of vertices that may be specified, other than the size of the vertex arrays. The mode parameter of these commands determines the type of primitives to be drawn using the vertices. Primitive types and the corresponding mode parameters are summarized below.

10.1.1

Points

A series of individual points are specified with mode POINTS. Each vertex defines a separate point.

10.1.2

Line Strips

A series of one or more connected line segments are specified with mode LINE_STRIP. In this case, the first vertex specifies the first segment’s start point while the second vertex specifies the first segment’s endpoint and the second segment’s start point. In general, the ith vertex (for i > 1) specifies the beginning of the ith segment and the end of the i − 1st. The last vertex specifies the end of the last segment. If only one vertex is specified, then no primitive is generated. The required state consists of the processed vertex produced from the last vertex that was sent (so that a line segment can be generated from it to the current vertex), and a boolean flag indicating if the current vertex is the first vertex.

10.1.3

Line Loops

A line loop is specified with mode LINE_LOOP. Loops are the same as line strips except that a final segment is added from the final specified vertex to the first vertex. The required state consists of the processed first vertex, in addition to the state required for line strips.

10.1.4

Separate Lines

Individual line segments, each defined by a pair of vertices, are specified with mode LINES. The first two vertices passed define the first segment, with subsequent pairs of vertices each defining one more segment. If the number of vertices passed is odd, then the last vertex is ignored. The state required is the same as for line strips but it is used differently: a processed vertex holding the first vertex of the current

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

4

2

259

2

2 3

6 4 4 5

1

3

5

(a)

1

5 1

(b)

3

(c)

Figure 10.2. (a) A triangle strip. (b) A triangle fan. (c) Independent triangles. The numbers give the sequencing of the vertices in order within the vertex arrays. Note that in (a) and (b) triangle edge ordering is determined by the first triangle, while in (c) the order of each triangle’s edges is independent of the other triangles.

segment, and a boolean flag indicating whether the current vertex is odd or even (a segment start or end).

10.1.5

Triangle Strips

A triangle strip is a series of triangles connected along shared edges, and is specified with mode TRIANGLE_STRIP. In this case, the first three vertices define the first triangle (and their order is significant). Each subsequent vertex defines a new triangle using that point along with two vertices from the previous triangle. If fewer than three vertices are specified, no primitive is produced. See figure 10.2. The required state consists of a flag indicating if the first triangle has been completed, two stored processed vertices (called vertex A and vertex B), and a one bit pointer indicating which stored vertex will be replaced with the next vertex. When a series of vertices are transferred to the GL, the pointer is initialized to point to vertex A. Each successive vertex toggles the pointer. Therefore, the first vertex is stored as vertex A, the second stored as vertex B, the third stored as vertex A, and so on. Any vertex after the second one sent forms a triangle from vertex A, vertex B, and the current vertex (in that order).

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

10.1.6

260

Triangle Fans

A triangle fan is specified with mode TRIANGLE_FAN, and is the same as a triangle strip with one exception: each vertex after the first always replaces vertex B of the two stored vertices.

10.1.7

Separate Triangles

Separate triangles are specified with mode TRIANGLES. In this case, the 3i + 1st, 3i + 2nd, and 3i + 3rd vertices (in that order) determine a triangle for each i = 0, 1, . . . , n − 1, where there are 3n + k vertices drawn. k is either 0, 1, or 2; if k is not zero, the final k vertices are ignored. For each triangle, vertex A is vertex 3i and vertex B is vertex 3i + 1. Otherwise, separate triangles are the same as a triangle strip.

10.1.8

Lines with Adjacency

Lines with adjacency are specified with mode LINES_ADJACENCY, and are independent line segments where each endpoint has a corresponding adjacent vertex that can be accessed by a geometry shader (see section 11.3). If a geometry shader is not active, the adjacent vertices are ignored. A line segment is drawn from the 4i + 2nd vertex to the 4i + 3rd vertex for each i = 0, 1, . . . , n − 1, where there are 4n + k vertices passed. k is either 0, 1, 2, or 3; if k is not zero, the final k vertices are ignored. For line segment i, the 4i + 1st and 4i + 4th vertices are considered adjacent to the 4i + 2nd and 4i + 3rd vertices, respectively (see figure 10.3).

10.1.9

Line Strips with Adjacency

Line strips with adjacency are specified with mode LINE_STRIP_ADJACENCY and are similar to line strips, except that each line segment has a pair of adjacent vertices that can be accessed by a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored. A line segment is drawn from the i + 2nd vertex to the i + 3rd vertex for each i = 0, 1, . . . , n − 1, where there are n + 3 vertices passed. If there are fewer than four vertices, all vertices are ignored. For line segment i, the i + 1st and i + 4th vertex are considered adjacent to the i + 2nd and i + 3rd vertices, respectively (see figure 10.3).

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

261

Figure 10.3. Lines with adjacency (a) and line strips with adjacency (b). The vertices connected with solid lines belong to the main primitives; the vertices connected by dashed lines are the adjacent vertices that may be used in a geometry shader.

Figure 10.4. Triangles with adjacency. The vertices connected with solid lines belong to the main primitive; the vertices connected by dashed lines are the adjacent vertices that may be used in a geometry shader.

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

262

Figure 10.5. Triangle strips with adjacency. The vertices connected with solid lines belong to the main primitives; the vertices connected by dashed lines are the adjacent vertices that may be used in a geometry shader.

10.1.10

Triangles with Adjacency

Triangles with adjacency are specified with mode TRIANGLES_ADJACENCY, and are similar to separate triangles except that each triangle edge has an adjacent vertex that can be accessed by a geometry shader. If a geometry shader is not active, the adjacent vertices are ignored. The 6i + 1st, 6i + 3rd, and 6i + 5th vertices (in that order) determine a triangle for each i = 0, 1, . . . , n − 1, where there are 6n + k vertices passed. k is either 0, 1, 2, 3, 4, or 5; if k is non-zero, the final k vertices are ignored. For triangle i, the i + 2nd, i + 4th, and i + 6th vertices are considered adjacent to edges from the i + 1st to the i + 3rd, from the i + 3rd to the i + 5th, and from the i + 5th to the i + 1st vertices, respectively (see figure 10.4).

OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

Primitive only (i = 0, n = 1) first (i = 0) middle (i odd) middle (i even) last (i = n − 1, i odd) last (i = n − 1, i even)

263 Primitive Vertices 1st 2nd 3rd 1 3 5 1 3 5 2i + 3 2i + 1 2i + 5 2i + 1 2i + 3 2i + 5 2i + 3 2i + 1 2i + 5 2i + 1 2i + 3 2i + 5

Adjacent Vertices 1/2 2/3 3/1 2 6 4 2 7 4 2i − 1 2i + 4 2i + 7 2i − 1 2i + 7 2i + 4 2i − 1 2i + 4 2i + 6 2i − 1 2i + 6 2i + 4

Table 10.1: Triangles generated by triangle strips with adjacency. Each triangle is drawn using the vertices whose numbers are in the 1st, 2nd, and 3rd columns under primitive vertices, in that order. The vertices in the 1/2, 2/3, and 3/1 columns under adjacent vertices are considered adjacent to the edges from the first to the second, from the second to the third, and from the third to the first vertex of the triangle, respectively. The six rows correspond to six cases: the first and only triangle (i = 0, n = 1), the first triangle of several (i = 0, n > 0), “odd” middle triangles (i = 1, 3, 5 . . .), “even” middle triangles (i = 2, 4, 6, . . .), and special cases for the last triangle, when i is either even or odd. For the purposes of this table, the first vertex passed is numbered 1 and the first triangle is numbered 0.

10.1.11

Triangle Strips with Adjacency

Triangle strips with adjacency are specified with mode TRIANGLE_STRIP_ADJACENCY, and are similar to triangle strips except that each triangle edge has an adjacent vertex that can be accessed by a geometry shader (see section 11.3). If a geometry shader is not active, the adjacent vertices are ignored. In triangle strips with adjacency, n triangles are drawn where there are 2(n + 2) + k vertices passed. k is either 0 or 1; if k is 1, the final vertex is ignored. If there are fewer than 6 vertices, the entire primitive is ignored. The order of vertices used to draw each triangle, and which vertices are considered adjacent to each edge of the triangle (see figure 10.5), is implementation defined but must be according to either table 10.1 or 10.2.

10.1.12

Separate Patches

Separate patches are specified with mode PATCHES. A patch is an ordered collection of vertices used for primitive tessellation (section 11.2). The vertices comprising a patch have no implied geometric ordering. The vertices of a patch are used by tessellation shaders and the fixed-function tessellator to generate new point, line, OpenGL ES 3.2 (May 5, 2022)

10.1. PRIMITIVE TYPES

Primitive only (i = 0, n = 1) first (i = 0) middle (i odd) middle (i even) last (i = n − 1, i odd) last (i = n − 1, i even)

264 Primitive Vertices 1st 2nd 3rd 1 3 5 1 3 5 2i + 1 2i + 5 2i + 3 2i + 1 2i + 3 2i + 5 2i + 1 2i + 5 2i + 3 2i + 1 2i + 3 2i + 5

Adjacent Vertices 1/2 2/3 3/1 2 6 4 2 7 4 2i + 4 2i + 7 2i − 1 2i − 1 2i + 7 2i + 4 2i + 4 2i + 6 2i − 1 2i − 1 2i + 6 2i + 4

Table 10.2: Alternative ordering to triangles generated by triangle strips with adjacency (see table 10.1).

or triangle primitives. Each patch in the series has a fixed number of vertices, which is specified by calling void PatchParameteri( enum pname, int value ); with pname set to PATCH_VERTICES. Errors An INVALID_ENUM error is generated if pname is not PATCH_VERTICES. An INVALID_VALUE error is generated if value is less than or equal to zero, or greater than the implementation-dependent maximum patch size (the value of MAX_PATCH_VERTICES). The patch size is initially three vertices. If the number of vertices in a patch is given by v, the vi + 1st through vi + vth vertices (in that order) determine a patch for each i = 0, 1, . . . n − 1, where there are vn + k vertices. k is in the range [0, v − 1]; if k is not zero, the final k vertices are ignored.

10.1.13

General Considerations For Polygon Primitives

A polygon primitive is one generated from a drawing command with mode TRIANGLE_FAN, TRIANGLE_STRIP, TRIANGLES, TRIANGLES_ADJACENCY, or TRIANGLE_STRIP_ADJACENCY. The order of vertices in such a primitive is significant in polygon rasterization (see section 13.7.1) and fragment shading (see section 14.2.2).

OpenGL ES 3.2 (May 5, 2022)

10.2. CURRENT VERTEX ATTRIBUTE VALUES

10.2

265

Current Vertex Attribute Values

The commands in this section are used to specify current attribute values. These values are used by drawing commands to define the attributes transferred for a vertex when a vertex array defining a required attribute is not enabled, as described in section 10.3.

10.2.1

Current Generic Attributes

Vertex shaders (see section 11.1) access an array of 4-component generic vertex attributes. The first slot of this array is numbered zero, and the size of the array is specified by the value of the implementation-dependent constant MAX_VERTEX_ATTRIBS. The current values of a generic shader attribute declared as a floating-point scalar, vector, or matrix may be changed at any time by issuing one of the commands void VertexAttrib{1234}f( uint index,float values ); void VertexAttrib{1234}fv( uint index,const float *values ); void VertexAttribI4{i ui}( uint index, T values ); void VertexAttribI4{i ui}v( uint index, const T values ); The VertexAttribI* commands specify signed or unsigned fixed-point values that are stored as signed or unsigned integers, respectively. Such values are referred to as pure integers. All other VertexAttrib* commands specify values that are converted directly to the internal floating-point representation. The resulting value(s) are loaded into the generic attribute at slot index, whose components are named x, y, z, and w. The VertexAttrib1* family of commands sets the x coordinate to the provided single argument while setting y and z to 0 and w to 1. Similarly, VertexAttrib2* commands set x and y to the specified values, z to 0 and w to 1; VertexAttrib3* commands set x, y, and z, with w set to 1, and VertexAttrib4* commands set all four coordinates. The VertexAttrib* entry points may also be used to load shader attributes declared as a floating-point matrix. Each column of a matrix takes up one generic 4-component attribute slot out of the MAX_VERTEX_ATTRIBS available slots. Matrices are loaded into these slots in column major order. Matrix columns are loaded in increasing slot numbers.

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

266

When values for a vertex shader attribute variable are sourced from a current generic attribute value, the attribute must be specified by a command compatible with the data type of the variable. The values loaded into a shader attribute variable bound to generic attribute index are undefined if the current value for attribute index was not specified by • VertexAttrib[1234]* for single-precision floating-point scalar, vector, and matrix types • VertexAttribI[1234]i or VertexAttribI[1234]iv, for signed integer scalar and vector types • VertexAttribI[1234]ui or VertexAttribI[1234]uiv, for unsigned integer scalar and vector types

Errors An INVALID_VALUE error is generated for all VertexAttrib* commands if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS.

10.2.2

Vertex Attribute Queries

Current generic vertex attribute values can be queried using the GetVertexAttrib* commands as described in section 10.6.

10.2.3

Required State

The state required to support vertex specification consists of the value of MAX_VERTEX_ATTRIBS four-component vectors to store generic vertex attributes. The initial values for all generic vertex attributes are (0.0, 0.0, 0.0, 1.0).

10.3

Vertex Arrays

Vertex data are placed into arrays that are stored in the client’s address space (described here) or in the server’s address space (described in section 10.3.7). Blocks of data in these arrays may then be used to specify multiple geometric primitives through the execution of a single GL command.

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

10.3.1

267

Specifying Arrays for Generic Vertex Attributes

A generic vertex attribute array is described by an index into an array of vertex buffer bindings which contain the vertex data and state describing how that data is organized. The commands void VertexAttribFormat( uint attribindex, int size, enum type, boolean normalized, uint relativeoffset ); void VertexAttribIFormat( uint attribindex, int size, enum type, uint relativeoffset ); describe the organization of vertex arrays. attribindex identifies the generic vertex attribute array. size indicates the number of values per vertex that are stored in the array. type specifies the data type of the values stored in the array. Table 10.3 indicates the allowable values for size and type. For type the values BYTE, UNSIGNED_BYTE, SHORT, UNSIGNED_SHORT, INT, UNSIGNED_INT FLOAT, and HALF_FLOAT indicate the corresponding GL data type shown in table 8.4. A type of FIXED indicates the data type fixed. A type of INT_2_10_10_10_REV or UNSIGNED_INT_2_10_10_10_REV, indicates respectively four signed or unsigned elements packed into a single uint; both correspond to the term packed in table 10.3. The components are packed as shown in figure 8.4. packed is not a GL type, but indicates commands accepting multiple components packed into a single uint. The “Integer Handling” column in table 10.3 indicates how integer and fixedpoint data are handled. “integer” means that they remain as integer values; such data are referred to as pure integers. “flag” means that either normalize or cast behavior applies, as described below, depending on whether the normalized flag to the command is TRUE or FALSE, respectively. normalize means that values are converted to floating-point by normalizing to [0, 1] (for unsigned types) or [−1, 1] (for signed types), as described in equations 2.1 and 2.2, respectively. cast means that values are converted to floating-point directly. The normalized flag is ignored for floating-point data types, including fixed, float, and half. relativeoffset is a byte offset of the first element relative to the start of the vertex buffer binding this attribute fetches from. Errors An INVALID_VALUE error is generated if attribindex is greater than or

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

Command VertexAttribPointer, AttribFormat

268

Vertex-

VertexAttribIPointer, VertexAttribIFormat

Sizes 1, 2, 3, 4

Integer Handling flag

1, 2, 3, 4

integer

Types byte, ubyte, short, ushort, int, uint, fixed, float, half, packed byte, ubyte, short, ushort, int, uint

Table 10.3: Vertex array sizes (values per vertex) and data types for generic vertex attributes. See the body text for a full description of each column.

equal to the value of MAX_VERTEX_ATTRIBS. An INVALID_VALUE error is generated if size is not one of the values shown in table 10.3 for the corresponding command. An INVALID_ENUM error is generated if type is not one of the parameter token names from table 8.4 corresponding to one of the allowed GL data types for that command as shown in table 10.3. An INVALID_OPERATION error is generated under any of the following conditions: • if the default vertex array object is currently bound (see section 10.4); • type is INT_2_10_10_10_REV or UNSIGNED_INT_2_10_10_10_REV, and size is not 4. An INVALID_VALUE error is generated if relativeoffset is larger than the value of MAX_VERTEX_ATTRIB_RELATIVE_OFFSET. A vertex buffer object is created by binding a name returned by GenBuffers to a bind point of the currently bound vertex array object. The binding is effected with the command void BindVertexBuffer( uint bindingindex, uint buffer, intptr offset, sizei stride ); The vertex buffer buffer is bound to the bind point bindingindex1 . 1

In order for buffer to be affected by any of the buffer object manipulation functions, such as BindBuffer or MapBufferRange, it must separately be bound to one of the general binding points.

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

269

Pointers to the ith and (i + 1)st elements of the array differ by stride basic machine units, the pointer to the (i + 1)st element being greater. offset specifies the offset in basic machine units of the first element in the vertex buffer. If buffer has not been previously bound, the GL creates a new state vector, initialized with a zero-sized memory buffer and comprising all the state and with the same initial values listed in table 6.2, just as for BindBuffer. BindVertexBuffer may also be used to bind an existing buffer object. If the bind is successful no change is made to the state of the newly bound buffer object, and any previous binding to bindingindex is broken. If buffer is zero, any buffer object bound to bindingindex is detached. Errors An INVALID_OPERATION error is generated if buffer is not zero, the name of an existing buffer object or a name returned from a previous call to GenBuffers, or if such a name has since been deleted with DeleteBuffers. An INVALID_VALUE error is generated if bindingindex is greater than or equal to the value of MAX_VERTEX_ATTRIB_BINDINGS. An INVALID_VALUE error is generated if stride or offset is negative, or if stride is greater than the value of MAX_VERTEX_ATTRIB_STRIDE. An INVALID_OPERATION error is generated if the default vertex array object is bound. The association between a vertex attribute and the vertex buffer binding used by that attribute is set by the command void VertexAttribBinding( uint attribindex, uint bindingindex ); Errors An INVALID_VALUE error is generated if attribindex is greater than or equal to the value of MAX_VERTEX_ATTRIBS. An INVALID_VALUE error is generated if bindingindex is greater than or equal to the value of MAX_VERTEX_ATTRIB_BINDINGS. An INVALID_OPERATION error is generated if the default vertex array object is bound. The one, two, three, or four values in an array that correspond to a single vertex comprise an array element. The values within each array element are stored sequentially in memory. OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

270

When values for a vertex shader attribute variable are sourced from an enabled generic vertex attribute array, the array must be specified by a command compatible with the data type of the variable. The values loaded into a shader attribute variable bound to generic attribute index are undefined if the array for index was not specified by: • VertexAttribFormat, for floating-point base type attributes; • VertexAttribIFormat with type BYTE, SHORT, or INT for signed integer base type attributes; or • VertexAttribIFormat with type UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT for unsigned integer base type attributes. The commands void VertexAttribPointer( uint index, int size, enum type, boolean normalized, sizei stride, const void *pointer ); void VertexAttribIPointer( uint index, int size, enum type, sizei stride, const void *pointer ); control vertex attribute state, a vertex buffer binding, and the mapping between a vertex attribute and a vertex buffer binding. They are equivalent to (assuming no errors are generated, and with the exception that no errors are generated if the default vertex array object is bound): if (the default vertex array object is bound and no buffer is bound to ARRAY_BUFFER) { vertex_buffer = temporary buffer offset = 0; } else { vertex_buffer = <buffer bound to ARRAY_BUFFER> offset = (char *)pointer - (char *)NULL; } VertexAttrib*Format(index, size, type, {normalized, }, 0); VertexAttribBinding(index, index); if (stride != 0) { effectiveStride = stride; } else { compute effectiveStride based on size and type; OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

271

} VERTEX_ATTRIB_ARRAY_STRIDE[index] = stride; VERTEX_ATTRIB_ARRAY_POINTER[index] = pointer; // This sets VERTEX_BINDING_STRIDE to effectiveStride

BindVertexBuffer(index, vertex_buffer, offset, effectiveStride); If stride is specified as zero, then array elements are stored sequentially. Errors An INVALID_VALUE error is generated if stride is greater than the value of MAX_VERTEX_ATTRIB_STRIDE. An INVALID_OPERATION error is generated if a non-zero vertex array object is bound, no buffer is bound to ARRAY_BUFFER, and pointer is not NULL2 . In addition, any of the errors defined by VertexAttrib*Format and VertexAttribBinding may be generated if the parameters passed to those commands in the equivalent code above would generate those errors. An individual generic vertex attribute array is enabled or disabled by calling one of void EnableVertexAttribArray( uint index ); void DisableVertexAttribArray( uint index ); where index identifies the generic vertex attribute array to enable or disable. Errors An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS.

10.3.2

Vertex Attribute Divisors

Each generic vertex attribute has a corresponding divisor which modifies the rate at which attributes advance, which is useful when rendering multiple instances of primitives in a single draw call. If the divisor is zero, the corresponding attributes advance once per vertex. Otherwise, attributes advance once per divisor instances 2

This error makes it impossible to create a vertex array object containing client array pointers, while still allowing buffer objects to be unbound.

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

272

of the set(s) of vertices being rendered. A generic attribute is referred to as instanced if its corresponding divisor value is non-zero. The command void VertexBindingDivisor( uint bindingindex, uint divisor ); sets the divisor value for attributes taken from the buffer bound to bindingindex. Errors An INVALID_VALUE error is generated if bindingindex is greater than or equal to the value of MAX_VERTEX_ATTRIB_BINDINGS. An INVALID_OPERATION error is generated if the default vertex array object is bound. The command void VertexAttribDivisor( uint index, uint divisor ); is equivalent to (assuming no errors are generated, and with the exception that no errors are generated if the default vertex array object is bound): VertexAttribBinding(index, index); VertexBindingDivisor(index, divisor); Errors An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS. An INVALID_OPERATION error is generated if the default vertex array object is bound.

10.3.3

Transferring Array Elements

When a vertex is transferred to the GL by DrawArrays, DrawElements, or the other Draw* commands described below, each generic attribute is expanded to four components. If size is one then the x component of the attribute is specified by the array; the y, z, and w components are implicitly set to 0, 0, and 1, respectively. If size is two then the x and y components of the attribute are specified by the array; the z and w components are implicitly set to 0 and 1, respectively. If size is three then x, y, and z are specified, and w is implicitly set to 1. If size is four then all components are specified. OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

10.3.4

273

Primitive Restart

Primitive restarting is enabled or disabled by calling one of the commands void Enable( enum target ); and void Disable( enum target ); with target PRIMITIVE_RESTART_FIXED_INDEX. When DrawElements, DrawElementsInstanced, or DrawRangeElements transfers a set of generic attribute array elements to the GL, if the index within the vertex arrays corresponding to that set is equal to 2N − 1, where N is 8, 16 or 32 if the type is UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT, respectively, then the GL does not process those elements as a vertex. Instead, it is as if the drawing command ended with the immediately preceding transfer, and another drawing command is immediately started with the same parameters, but only transferring the immediately following element through the end of the originally specified elements. When one of the *BaseVertex drawing commands specified in section 10.5 is used, the primitive restart comparison occurs before the basevertex offset is added to the array index. Implementations are not required to support primitive restart for separate patch primitives (primitive type PATCHES). Support can be queried by calling GetBooleanv with pname PRIMITIVE_RESTART_FOR_PATCHES_SUPPORTED. A value of FALSE indicates that primitive restart is treated as disabled when drawing patches, no matter the value of the enables. A value of TRUE indicates that primitive restart behaves normally for patches.

10.3.5

Robust Buffer Access

Robust buffer access is enabled by creating a context with robust access enabled through the window system binding APIs. When enabled, indices within the element array (see section 10.3.8) that reference vertex data that lies outside the enabled attribute’s vertex buffer object result in undefined values for the corresponding attributes, but cannot result in application failure. Robust buffer access behavior may be queried by calling GetIntegerv with pname CONTEXT_FLAGS, as described in section 20.2.

OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

10.3.6

274

Packed Vertex Data Formats

Vertex data formats UNSIGNED_INT_2_10_10_10_REV and INT_2_10_10_10_REV describe packed, 4 component formats stored in a single 32-bit word. For UNSIGNED_INT_2_10_10_10_REV, the first (x), second (y), and third (z) components are represented as 10-bit unsigned integer values and the fourth (w) component is represented as a 2-bit unsigned integer value. For INT_2_10_10_10_REV, the x, y and z components are represented as 10bit signed two’s complement integer values and the w component is represented as a 2-bit signed two’s complement integer value. The normalized value is used to indicate whether to normalize the data to [0, 1] (for unsigned types) or [−1, 1] (for signed types). During normalization, the conversion rules specified in equations 2.1 and 2.2 are followed. Figure 10.6 describes how these components are laid out in a 32-bit word. 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

w

z

y

8

7

6

5

4

3

2

1

0

x

Figure 10.6: Packed component layout. Bit numbers are indicated for each component.

10.3.7

Vertex Arrays in Buffer Objects

Blocks of vertex array data may be stored in buffer objects with the same format and layout options described in section 10.3. A buffer object binding point is added to the client state associated with each vertex array index. The commands that specify the locations and organizations of vertex arrays copy the buffer object name that is bound to ARRAY_BUFFER to the binding point corresponding to the vertex array index being specified. For example, the VertexAttribPointer command copies the value of ARRAY_BUFFER_BINDING (the queriable name of the buffer binding corresponding to the target ARRAY_BUFFER) to the client state variable VERTEX_ATTRIB_ARRAY_BUFFER_BINDING for the specified index. The drawing commands using vertex arrays described in section 10.5 operate as previously defined, except that data for enabled generic attribute arrays are sourced from buffers if the array’s buffer binding is non-zero. When an array is sourced from a buffer object for a vertex attribute, the bindingindex set with VertexAttribBinding for that attribute indicates which vertex buffer binding is used. The sum of the relativeoffset set for the attribute OpenGL ES 3.2 (May 5, 2022)

10.3. VERTEX ARRAYS

275

with VertexAttrib*Format and the offset set for the vertex buffer with BindVertexBuffer is used as the offset in basic machine units of the first element in that buffer’s data store. When a generic attribute array is sourced from client memory, the vertex attribute binding state is ignored. Instead, the parameters set with VertexAttrib*Pointer for that attribute indicate the location in client memory of attribute values and their size, type, and stride.

10.3.8

Array Indices in Buffer Objects

Blocks of array indices may be stored in buffer objects with the same format options that are supported for client-side index arrays. Initially zero is bound to ELEMENT_ARRAY_BUFFER, indicating that DrawElements, DrawRangeElements, and DrawElementsInstanced are to source their indices from arrays passed as their indices parameters. A buffer object is bound to ELEMENT_ARRAY_BUFFER by calling BindBuffer with target set to ELEMENT_ARRAY_BUFFER, and buffer set to the name of the buffer object. If no corresponding buffer object exists, one is initialized as defined in section 6. While a non-zero buffer object name is bound to ELEMENT_ARRAY_BUFFER, DrawElements, DrawRangeElements, and DrawElementsInstanced source their indices from that buffer object, using their indices parameters as offsets into the buffer object in the same fashion as described in section 10.3.7. DrawElementsBaseVertex, DrawRangeElementsBaseVertex, and DrawElementsInstancedBaseVertex also source their indices from that buffer object, adding the basevertex offset to the appropriate vertex index as a final step before indexing into the vertex buffer; this does not affect the calculation of the base pointer for the index array. In some cases performance will be optimized by storing indices and array data in separate buffer objects, and by creating those buffer objects with the corresponding binding points.

10.3.9

Indirect Commands in Buffer Objects

Arguments to the indirect commands DrawArraysIndirect and DrawElementsIndirect (see section 10.5), and to DispatchComputeIndirect (see section 17) are sourced from the buffer object currently bound to the corresponding indirect buffer target (see table 10.4), using the command’s indirect parameter as an offset into the buffer object in the same fashion as described in section 10.3.7.

OpenGL ES 3.2 (May 5, 2022)

10.4. VERTEX ARRAY OBJECTS Indirect Command Name DrawArraysIndirect DrawElementsIndirect DispatchComputeIndirect

276 Indirect Buffer target DRAW_INDIRECT_BUFFER DRAW_INDIRECT_BUFFER DISPATCH_INDIRECT_BUFFER

Table 10.4: Indirect commands and corresponding indirect buffer targets.

Buffer objects are created and/or bound to a target as described in section 6.1. Initially zero is bound to each target. Arguments are stored in buffer objects as structures (for Draw*Indirect) or arrays (for DispatchComputeIndirect) of tightly packed 32-bit integers.

10.4

Vertex Array Objects

The buffer objects that are to be used by the vertex stage of the GL are collected together to form a vertex array object. All state related to the definition of data used by the vertex processor is encapsulated in a vertex array object. The name space for vertex array objects is the unsigned integers, with zero reserved by the GL to represent the default vertex array object. The command void GenVertexArrays( sizei n, uint *arrays ); returns n previous unused vertex array object names in arrays. These names are marked as used, for the purposes of GenVertexArrays only, but they acquire array state only when they are first bound. Errors An INVALID_VALUE error is generated if n is negative. Vertex array objects are deleted by calling void DeleteVertexArrays( sizei n, const uint *arrays ); arrays contains n names of vertex array objects to be deleted. Once a vertex array object is deleted it has no contents and its name is again unused. If a vertex array object that is currently bound is deleted, the binding for that object reverts to zero and the default vertex array becomes current. Unused names in arrays that have been marked as used for the purposes of GenVertexArrays are marked as unused again. Unused names in arrays are silently ignored, as is the value zero. OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

277

Errors An INVALID_VALUE error is generated if n is negative. A vertex array object is created by binding a name returned by GenVertexArrays with the command void BindVertexArray( uint array ); array is the vertex array object name. The resulting vertex array object is a new state vector, comprising all the state and with the same initial values listed in table 21.3. BindVertexArray may also be used to bind an existing vertex array object. If the bind is successful no change is made to the state of the bound vertex array object, and any previous binding is broken. The currently bound vertex array object is used for all commands which modify vertex array state, such as VertexAttribPointer and EnableVertexAttribArray; all commands which draw from vertex arrays, such as DrawArrays and DrawElements; and all queries of vertex array state (see chapter 20). Errors An INVALID_OPERATION error is generated if array is not zero or a name returned from a previous call to GenVertexArrays, or if such a name has since been deleted with DeleteVertexArrays. The command boolean IsVertexArray( uint array ); returns TRUE if array is the name of a vertex array object. If array is zero, or a non-zero value that is not the name of a vertex array object, IsVertexArray returns FALSE. No error is generated if array is not a valid vertex array object name.

10.5

Drawing Commands Using Vertex Arrays

The command void DrawArraysOneInstance( enum mode, int first, sizei count, int instance, uint baseinstance );

OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

278

does not exist in the GL, but is used to describe functionality in the rest of this section. This command constructs a sequence of geometric primitives by successively transferring elements for count vertices. Elements first through first + count − 1 of each enabled non-instanced array are transferred to the GL. If count is zero, no elements are transferred. mode specifies what kind of primitives are constructed, and must be one of the primitive types defined in section 10.1. If an enabled vertex attribute array is instanced (it has a non-zero divisor as specified by VertexAttribDivisor), the element index that is transferred to the GL, for all vertices, is given by 3   instance + baseinstance divisor If an array corresponding to an attribute required by a vertex shader is not enabled, then the corresponding element is taken from the current attribute state (see section 10.2). If an array is enabled, the corresponding current vertex attribute value is unaffected by the execution of DrawArraysOneInstance. The index of any element transferred to the GL by DrawArraysOneInstance is referred to as its vertex ID, and may be read by a vertex shader as gl_VertexID. The vertex ID of the ith element transferred is first + i. The value of instance may be read by a vertex shader as gl_InstanceID, as described in section 11.1.3.9. Errors An INVALID_ENUM error is generated if mode is not one of the primitive types defined in section 10.1. Specifying f irst < 0 results in undefined behavior. Generating an INVALID_VALUE error is recommended in this case. An INVALID_VALUE error is generated if count is negative. The command void DrawArrays( enum mode, int first, sizei count ); is equivalent to DrawArraysOneInstance(mode, f irst, count, 0, 0); 3

baseinstance is included for commonality with OpenGL, but its value will always be zero in unextended OpenGL ES.

OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

279

The command void DrawArraysInstanced( enum mode, int first, sizei count, sizei instancecount ); behaves identically to DrawArrays except that instancecount instances of the range of elements are executed and the value of instance advances for each iteration. Those attributes that have non-zero values for divisor, as specified by VertexAttribDivisor, advance once every divisor instances. DrawArraysInstanced is equivalent to if (mode, count, or instancecount is invalid) generate appropriate error else { for (i = 0; i < instancecount; i++) { DrawArraysOneInstance(mode, f irst, count, i, 0); } } The command void DrawArraysIndirect( enum mode, const void *indirect ); is equivalent to typedef struct { uint count; uint instanceCount; uint first; uint reservedMustBeZero; } DrawArraysIndirectCommand; DrawArraysIndirectCommand *cmd = (DrawArraysIndirectCommand *)indirect; DrawArraysInstanced(mode, cmd->first, cmd->count, cmd->instanceCount); Unlike DrawArraysInstanced, the first argument is unsigned and cannot cause an error.

OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

280

DrawArraysIndirect requires that all data sourced for the command, including the DrawArraysIndirectCommand structure, be in buffer objects, and may not be called when the default vertex array object is bound. All elements of DrawArraysIndirectCommand are tightly-packed 32-bit values. Errors An INVALID_OPERATION error is generated if zero is bound to VERTEX_ARRAY_BINDING, DRAW_INDIRECT_BUFFER or to any enabled vertex array. An INVALID_OPERATION error is generated if the command would source data beyond the end of the buffer object. An INVALID_VALUE error is generated if indirect is not a multiple of the size, in basic machine units, of uint. Results are undefined if reservedMustBeZero is non-zero, but may not result in program termination. The command void DrawElementsOneInstance( enum mode, sizei count, enum type, const void *indices, int instance, int basevertex, uint baseinstance ); does not exist in the GL, but is used to describe functionality in the rest of this section. This command constructs a sequence of geometric primitives by successively transferring elements for count vertices to the GL. The index of any element transferred to the GL by DrawElementsOneInstance is referred to as its vertex ID, and may be read by a vertex shader as gl_VertexID. If no element array buffer is bound, the vertex ID of the ith element transferred is indices[i] + basevertex. Otherwise, the vertex ID of the ith element transferred is the sum of basevertex and the value stored in the currently bound element array buffer at offset indices + i × sizeof (type). If the vertex ID is larger than the maximum value representable by type, it should behave as if the calculation were upconverted to 32-bit unsigned integers (with wrapping on overflow conditions). Behavior of DrawElementsOneInstance is undefined if the vertex ID is negative for any element, and should be handled as described in section 6.4. type must be one of UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT, indicating that the index values are of GL type ubyte, ushort, or uint respectively. mode specifies what kind of primitives are constructed, and must be one of the primitive types defined in section 10.1. OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

281

If an enabled vertex attribute array is instanced (it has a non-zero divisor as specified by VertexAttribDivisor), the element index that is transferred to the GL, for all vertices, is given by 4   instance + baseinstance divisor If type is UNSIGNED_INT, an implementation may restrict the maximum value that can be used as an index to less than the maximum value that can be represented by the uint type. The maximum value supported by an implementation may be queried by calling GetInteger64v with pname MAX_ELEMENT_INDEX. If an array corresponding to a generic attribute is not enabled, then the corresponding element is taken from the current attribute state (see section 10.2). If an array is enabled, the corresponding current vertex attribute value is unaffected by the execution of DrawElementsOneInstance. The value of instance may be read by a vertex shader as gl_InstanceID, as described in section 11.1.3.9. Errors An INVALID_ENUM error is generated if mode is not one of the primitive types defined in section 10.1. An INVALID_ENUM error is generated if type is not UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT. Using an index value greater than MAX_ELEMENT_INDEX will result in undefined implementation-dependent behavior, unless primitive restart is enabled (see section 10.3.4) and the index value is 232 − 1. The command void DrawElements( enum mode, sizei count, enum type, const void *indices ); behaves identically to DrawElementsOneInstance with instance, basevertex, and baseinstance set to zero; the effect of calling DrawElements(mode, count, type, indices); is equivalent to 4 As described for DrawArraysOneInstance above, the value of baseinstance will always be zero.

OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

282

if (mode, count or type is invalid) generate appropriate error else DrawElementsOneInstance(mode, count, type, indices, 0, 0, 0); The command void DrawElementsInstanced( enum mode, sizei count, enum type, const void *indices, sizei instancecount ); behaves identically to DrawElements except that instancecount instances of the set of elements are executed and the value of instance advances between each set. Instanced attributes are advanced as they do during execution of DrawArraysInstanced. It has the same effect as: if (mode, count, instancecount, or type is invalid) generate appropriate error else { for (int i = 0; i < instancecount; i++) { DrawElementsOneInstance(mode, count, type, indices, i, 0, 0); } } The command void DrawRangeElements( enum mode, uint start, uint end, sizei count, enum type, const void *indices ); is a restricted form of DrawElements. mode, count, type, and indices match the corresponding arguments to DrawElements, with the additional constraint that all index values identified by indices must lie between start and end inclusive. Implementations denote recommended maximum amounts of vertex and index data, which may be queried by calling GetIntegerv with the symbolic constants MAX_ELEMENTS_VERTICES and MAX_ELEMENTS_INDICES. If end − start + 1 is greater than the value of MAX_ELEMENTS_VERTICES, or if count is greater than the value of MAX_ELEMENTS_INDICES, then the call may operate at reduced performance. There is no requirement that all vertices in the range [start, end] be referenced. However, the implementation may partially process unused vertices, reducing performance from what could be achieved with an optimal index set. OpenGL ES 3.2 (May 5, 2022)

10.5. DRAWING COMMANDS USING VERTEX ARRAYS

283

Errors An INVALID_VALUE error is generated if end < start. Invalid mode, count, or type parameters generate the same errors as would the corresponding call to DrawElements. It is an error for index values (other than the primitive restart index, when primitive restart is enabled) to lie outside the range [start, end], but implementations are not required to check for this. Such indices will cause implementation-dependent behavior. The commands void DrawElementsBaseVertex( enum mode, sizei count, enum type, const void *indices, int basevertex ); void DrawRangeElementsBaseVertex( enum mode, uint start, uint end, sizei count, enum type, const void *indices, int basevertex ); void DrawElementsInstancedBaseVertex( enum mode, sizei count, enum type, const void *indices, sizei instancecount, int basevertex ); are equivalent to the commands with the same base name (without the BaseVertex suffix), except that the basevertex value passed to DrawElementsOneInstance is the basevertex value of these commands, instead of zero. For DrawRangeElementsBaseVertex, the values taken from indices for each element transferred must be in the range [start, end] prior to adding the basevertex offset. Index values lying outside this range are treated in the same way as DrawRangeElements. The command void DrawElementsIndirect( enum mode, enum type, const void *indirect ); is equivalent to typedef struct { uint count; uint instanceCount; uint firstIndex; int baseVertex; uint reservedMustBeZero; OpenGL ES 3.2 (May 5, 2022)

10.6. VERTEX ARRAY AND VERTEX ARRAY OBJECT QUERIES } DrawElementsIndirectCommand; if (no element array buffer is bound) { generate appropriate error } else { DrawElementsIndirectCommand *cmd = (DrawElementsIndirectCommand *)indirect; DrawElementsInstancedBaseVertex(mode, cmd->count, type, cmd->firstIndex * size-of-type, cmd->instanceCount, cmd->baseVertex); } DrawElementsIndirect requires that all data sourced for the command, including the DrawElementsIndirectCommand structure, be in buffer objects, and may not be called when the default vertex array object is bound. All elements of DrawElementsIndirectCommand are tightly-packed 32-bit values. Errors An INVALID_OPERATION error is generated if zero is bound to VERTEX_ARRAY_BINDING, DRAW_INDIRECT_BUFFER, ELEMENT_ARRAY_BUFFER, or to any enabled vertex array. An INVALID_OPERATION error is generated if the command would source data beyond the end of the buffer object. An INVALID_VALUE error is generated if indirect is not a multiple of the size, in basic machine units, of uint. Results are undefined if reservedMustBeZero is non-zero, but may not result in program termination.

10.6

Vertex Array and Vertex Array Object Queries

Queries of vertex array state variables are qualified by the value of VERTEX_ARRAY_BINDING to determine which vertex array object is queried. Table 21.3 defines the set of state stored in a vertex array object. The commands void GetVertexAttribfv( uint index, enum pname, float *params ); OpenGL ES 3.2 (May 5, 2022)

284

10.6. VERTEX ARRAY AND VERTEX ARRAY OBJECT QUERIES

285

void GetVertexAttribiv( uint index, enum pname, int *params ); void GetVertexAttribIiv( uint index, enum pname, int *params ); void GetVertexAttribIuiv( uint index, enum pname, uint *params ); obtain the vertex attribute state named by pname for the generic vertex attribute numbered index and places the information in the array params. pname must be one of VERTEX_ATTRIB_ARRAY_BUFFER_BINDING, VERTEX_ATTRIB_ARRAY_ENABLED, VERTEX_ATTRIB_ARRAY_SIZE, VERTEX_ATTRIB_ARRAY_STRIDE, VERTEX_ATTRIB_ARRAY_TYPE, VERTEX_ATTRIB_ARRAY_NORMALIZED, VERTEX_ATTRIB_ARRAY_INTEGER, VERTEX_ATTRIB_ARRAY_DIVISOR, VERTEX_ATTRIB_BINDING, VERTEX_ATTRIB_RELATIVE_OFFSET, or CURRENT_VERTEX_ATTRIB. Note that all the queries except CURRENT_VERTEX_ATTRIB return values stored in the currently bound vertex array object (the value of VERTEX_ARRAY_BINDING). If the zero object is bound, these values are client state. Queries of VERTEX_ATTRIB_ARRAY_BUFFER_BINDING and VERTEX_ATTRIB_ARRAY_DIVISOR map the requested attribute index to a binding index via the VERTEX_ATTRIB_BINDING state, and then return the value of VERTEX_BINDING_BUFFER or VERTEX_BINDING_DIVISOR, respectively. All but CURRENT_VERTEX_ATTRIB return information about generic vertex attribute arrays. The enable state of a generic vertex attribute array is set by the command EnableVertexAttribArray and cleared by DisableVertexAttribArray. The size, stride, type, relative offset, normalized flag, and unconverted integer flag are set by the commands VertexAttribFormat and VertexAttribIFormat. The normalized flag is always set to FALSE by VertexAttribIFormat. The unconverted integer flag is always set to FALSE by VertexAttribFormat and TRUE VertexAttribIFormat. The query CURRENT_VERTEX_ATTRIB returns the current value for the generic attribute index. GetVertexAttribfv reads and returns the current attribute values as floating-point values; GetVertexAttribiv reads them as floating-point values and converts them to integer values; GetVertexAttribIiv reads and returns them as integers; GetVertexAttribIuiv reads and returns them as unsigned integers. The results of the query are undefined if the current attribute values are read using one data type but were specified using a different one.

OpenGL ES 3.2 (May 5, 2022)

10.7. REQUIRED STATE

286

Errors An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS. An INVALID_ENUM error is generated if pname is not one of the values listed above. The command void GetVertexAttribPointerv( uint index, enum pname, const void **pointer ); obtains the pointer named pname for the vertex attribute numbered index and places the information in the array pointer. pname must be VERTEX_ATTRIB_ARRAY_POINTER. The value returned is queried from the currently bound vertex array object. If the zero object is bound, the value is queried from client state. Errors An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS. Finally, the buffer bound to ELEMENT_ARRAY_BUFFER may be queried by calling GetIntegerv with the symbolic constant ELEMENT_ARRAY_BUFFER_BINDING.

10.7

Required State

Let the number of supported generic vertex attributes (the value of MAX_VERTEX_ATTRIBS) be n. Let the number of supported generic vertex attribute bindings (the value of MAX_VERTEX_ATTRIB_BINDINGS) be k. Then the state required to implement vertex arrays consists of n boolean values, n memory pointers, n integer stride values, n symbolic constants representing array types, n integers representing values per element, n boolean values indicating normalization, n boolean values indicating whether the attribute values are pure integers, k integers representing vertex attribute divisors, n integer vertex attribute binding indices, n integer relative offsets, k 64-bit integer vertex binding offsets, and k integer vertex binding strides, In the initial state, the boolean values are each false, the memory pointers are each NULL, the strides are each zero, the array types are each FLOAT, the integers representing values per element are each four, the normalized and pure integer flags OpenGL ES 3.2 (May 5, 2022)

10.7. REQUIRED STATE

287

are each false, the divisors are each zero, the binding indices are i for each attribute i, the relative offsets are each zero, the vertex binding offsets are each zero, and the vertex binding strides are each 16.

OpenGL ES 3.2 (May 5, 2022)

Chapter 11

Programmable Vertex Processing When the program object currently in use for the vertex stage (see section 7.3) includes a vertex shader, its shader is considered active and is used to process vertices transferred to the GL (see section 11.1). The resulting transformed vertices are then processed as described in chapter 12. If the current vertex stage program object has no vertex shader, or no program object is current for the vertex stage, the results of programmable vertex processing are undefined.

11.1

Vertex Shaders

Vertex shaders describe the operations that occur on vertex values and their associated data. When the program object currently in use for the vertex stage includes a vertex shader, its vertex shader is considered active and is used to process vertices. Vertex attributes are per-vertex values available to vertex shaders, and are specified as described in section 10.2.

11.1.1

Vertex Attributes

Vertex shaders can define named attribute variables, which are bound to generic vertex attributes transferred by drawing commands. This binding can be specified by the application before the program is linked, or automatically assigned by the GL when the program is linked. When an attribute variable declared using one of the scalar or vector data types enumerated in table 11.3 is bound to a generic attribute index i, its value(s) are taken from the components of generic attribute i. The generic attribute components

288

11.1. VERTEX SHADERS

289

Data type

Component

scalar scalar scalar scalar two-component vector two-component vector two-component vector three-component vector three-component vector four-component vector

0 1 2 3 0 1 2 0 1 0

Components used x y z w (x, y) (y, z) (z, w) (x, y, z) (y, z, w) (x, y, z, w)

Table 11.1: Generic attribute components accessed by attribute variables.

used depend on the type of the variable specified in the variable declaration, as identified in table 11.1. When an attribute variable declared using a matrix type is bound to a generic attribute index i, its values are taken from consecutive generic attributes beginning with generic attribute i. Such matrices are treated as an array of column vectors with values taken from the generic attributes identified in table 11.2. Individual column vectors are taken from generic attribute components according to table 11.1, using the vector type from table 11.2. The command void BindAttribLocation( uint program, uint index, const char *name ); specifies that the attribute variable named name in program program should be bound to generic vertex attribute index when the program is next linked. If name was bound previously, its assigned binding is replaced with index, but the new binding becomes effective only when the program is next linked. name must be a null-terminated string. BindAttribLocation has no effect until the program is linked. In particular, it doesn’t modify the bindings of active attribute variables in a program that has already been linked. When a program is linked, any active attributes without a binding specified either through BindAttribLocation or explicitly set within the shader text will automatically be bound to vertex attributes by the GL. Such bindings can be queried using the command GetAttribLocation. LinkProgram will fail if the OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

290

Data type

Column vector type

mat2 mat2x3 mat2x4 mat3x2 mat3 mat3x4 mat4x2 mat4x3 mat4

two-component vector three-component vector four-component vector two-component vector three-component vector four-component vector two-component vector three-component vector four-component vector

Generic attributes used i, i + 1 i, i + 1 i, i + 1 i, i + 1, i + 2 i, i + 1, i + 2 i, i + 1, i + 2 i, i + 1, i + 2, i + 3 i, i + 1, i + 2, i + 3 i, i + 1, i + 2, i + 3

Table 11.2: Generic attributes and vector types used by column vectors of matrix variables bound to generic attribute index i.

Data type float vec2 vec3 vec4

Command VertexAttrib1* VertexAttrib2* VertexAttrib3* VertexAttrib4*

Table 11.3: Scalar and vector vertex attribute types and VertexAttrib* commands used to set the values of the corresponding generic attribute.

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

291

assigned binding of an active attribute variable would cause the GL to reference a non-existent generic attribute (one greater than or equal to the value of MAX_VERTEX_ATTRIBS). LinkProgram will fail if the attribute bindings specified either by BindAttribLocation or explicitly set within the shader text do not leave not enough space to assign a location for an active matrix attribute which requires multiple contiguous generic attributes. If an active attribute has a binding explicitly set within the shader text and a different binding assigned by BindAttribLocation, the assignment in the shader text is used. BindAttribLocation may be issued before any vertex shader objects are attached to a program object. Hence it is allowed to bind any name (except a name starting with "gl_") to an index, including a name that is never used as an attribute in any vertex shader object. Assigned bindings for attribute variables that do not exist or are not active are ignored. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS. An INVALID_OPERATION error is generated if name starts with the reserved "gl_" prefix). To determine the set of active vertex attribute variables used by a program, applications can query the properties and active resources of the PROGRAM_INPUT interface of a program including a vertex shader. Additionally, the command void GetActiveAttrib( uint program, uint index, sizei bufSize, sizei *length, int *size, enum *type, char *name ); can be used to determine properties of the active input variable assigned the index index in program object program. If no error occurs, the command is equivalent to const enum props[] = { ARRAY_SIZE, TYPE }; GetProgramResourceName(program, PROGRAM_INPUT, index, bufSize, length, name); GetProgramResourceiv(program, PROGRAM_INPUT, OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

292

index, 1, &props[0], 1, NULL, size); GetProgramResourceiv(program, PROGRAM_INPUT, index, 1, &props[1], 1, NULL, (int *)type); For GetActiveAttrib, all active vertex shader input variables are enumerated, including the special built-in inputs gl_VertexID and gl_InstanceID. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if index is not the index of an active input variable in program. An INVALID_VALUE error is generated for all values of index if program does not include a vertex shader, as it has no active vertex attributes. An INVALID_VALUE error is generated if bufSize is negative. The command int GetAttribLocation( uint program, const char *name ); can be used to determine the location assigned to the active input variable named name in program object program. Errors If program has been linked successfully but contains no vertex shader, no error is generated but -1 will be returned. An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_OPERATION error is generated and -1 is returned if program has not been linked successfully. Otherwise, the command is equivalent to GetProgramResourceLocation(program, PROGRAM_INPUT, name);

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

293

There is an implementation-dependent limit on the number of active attribute variables in a vertex shader. A program with more than the value of MAX_VERTEX_ATTRIBS active attribute variables may fail to link, unless devicedependent optimizations are able to make the program fit within available hardware resources. The values of generic attributes sent to generic attribute index i are part of current state. If a new program object has been made active, then these values will be tracked by the GL in such a way that the same values will be observed by attributes in the new program object that are also bound to index i. Binding more than one attribute name to the same location is referred to as aliasing, and is not permitted in OpenGL ES Shading Language 3.00 or later vertex shaders. LinkProgram will fail when this condition exists. However, aliasing is possible in OpenGL ES Shading Language 1.00 vertex shaders. This will only work if only one of the aliased attributes is active in the executable program, or if no path through the shader consumes more than one attribute of a set of attributes aliased to the same location. A link error can occur if the linker determines that every path through the shader consumes multiple aliased attributes, but implementations are not required to generate an error in this case. The compiler and linker are allowed to assume that no aliasing is done, and may employ optimizations that work only in the absence of aliasing.

11.1.2

Vertex Shader Variables

Vertex shaders can access uniforms belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. Vertex shaders also have access to samplers to perform texturing operations, as described in section 7.9. 11.1.2.1

Output Variables

A vertex shader may define one or more output variables or outputs (see the OpenGL ES Shading Language Specification). The OpenGL ES Shading Language Specification also defines a set of built-in outputs that vertex shaders can write to (see section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification). These output variables are used as the mechanism to communicate values to the next active stage in the vertex processing pipeline: either the tessellation control shader, the tessellation evaluation shader, the geometry shader, or the fixed-function vertex processing stages leading to rasterization.

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

294

If the output variables are passed directly to the vertex processing stages leading to rasterization, the values of all outputs are expected to be interpolated across the primitive being rendered, unless flatshaded. Otherwise the values of all outputs are collected by the primitive assembly stage and passed on to the subsequent pipeline stage once enough data for one primitive has been collected. The number of components (individual scalar numeric values) of output variables that can be written by the vertex shader, whether or not a tessellation control, tessellation evaluation, or geometry shader is active, is given by the value of the implementation-dependent constant MAX_VERTEX_OUTPUT_COMPONENTS. Outputs declared as vectors, matrices, and arrays will all consume multiple components. When a program is linked, all components of any outputs written by a vertex shader will count against this limit. A program whose vertex shader writes more than the value of MAX_VERTEX_OUTPUT_COMPONENTS components worth of outputs may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Additionally, when linking a program containing only a vertex and fragment shader, there is a limit on the total number of components used as vertex shader outputs or fragment shader inputs. This limit is given by the value of the implementation-dependent constant MAX_VARYING_COMPONENTS. The implementation-dependent constant MAX_VARYING_VECTORS has a value equal to the value of MAX_VARYING_COMPONENTS divided by four. Each output variable component used as either a vertex shader output or fragment shader input count against this limit, except for the components of gl_Position. A program that accesses more than this limit’s worth of components of outputs may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Each program object can specify a set of output variables from one shader to be recorded in transform feedback mode (see section 12.2). The variables that can be recorded are those emitted by the first active shader, in order, from the following list: • geometry shader • tessellation evaluation shader • vertex shader The set of variables to record is specified with the command

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

295

void TransformFeedbackVaryings( uint program, sizei count, const char * const *varyings, enum bufferMode ); program specifies the program object. count specifies the number of output variables used for transform feedback. varyings is an array of count zero-terminated strings specifying the names of the outputs to use for transform feedback. The variables specified in varyings can be either built-in (beginning with "gl_") or user-defined variables. Each variable can either be a basic type or an array of basic types. Structure, array of array and array of structure types cannot be captured directly. Base-level members of aggregates can be captured by specifying the fully qualified path identifying the member, using the same rules with which active resource lists are enumerated for program interfaces as described in section 7.3.1.1, with one exception. To allow capturing whole arrays or individual elements of an array, there are additional rules for array variables. To capture a single element, the name of the output array is specified with a constant-integer index "name[x]" where name is the name of the array variable and x is the constant-integer index of the array element. To capture the whole of the output array, name is specified without the array index or square brackets. Output variables are written out in the order they appear in the array varyings. bufferMode is either INTERLEAVED_ATTRIBS or SEPARATE_ATTRIBS, and identifies the mode used to capture the outputs when transform feedback is active. Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. An INVALID_VALUE error is generated if count is negative. An INVALID_ENUM error is generated if bufferMode is not SEPARATE_ATTRIBS or INTERLEAVED_ATTRIBS. An INVALID_VALUE error is generated if bufferMode is SEPARATE_ATTRIBS and count is greater than the value of the implementation-dependent limit MAX_TRANSFORM_FEEDBACK_SEPARATE_ATTRIBS. The state set by TransformFeedbackVaryings has no effect on the execution of the program until program is subsequently linked. When LinkProgram is called, the program is linked so that the values of the specified outputs for the vertices of each primitive generated by the GL are written to a single buffer object (if the buffer mode is INTERLEAVED_ATTRIBS) or multiple buffer objects (if the OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

296

buffer mode is SEPARATE_ATTRIBS). A program will fail to link if: • the count specified by TransformFeedbackVaryings is non-zero, but the program object has no vertex, tessellation evaluation, or geometry shader; • any variable name specified in the varyings array is not declared as a builtin or user-defined output variable in the shader stage whose outputs can be recorded; • any two entries in the varyings array specify the same output variable or include the same elements from an array variable (different elements from the same array are permitted); • the total number of components to capture in any output in varyings is greater than the value of MAX_TRANSFORM_FEEDBACK_SEPARATE_COMPONENTS and the buffer mode is SEPARATE_ATTRIBS; or • the total number of components to capture is greater than the value of MAX_TRANSFORM_FEEDBACK_INTERLEAVED_COMPONENTS and the buffer mode is INTERLEAVED_ATTRIBS. When a program is linked, a list of output variables that will be captured in transform feedback mode is built as described in section 7.3. The variables in this list are assigned consecutive indices, beginning with zero. The total number of variables in the list may be queried by calling GetProgramiv (section 7.12) with a pname of TRANSFORM_FEEDBACK_VARYINGS. To determine the set of output variables in a linked program object that will be captured in transform feedback mode, applications can query the properties and active resources of the TRANSFORM_FEEDBACK_VARYING interface. Additionally, the dedicated command void GetTransformFeedbackVarying( uint program, uint index, sizei bufSize, sizei *length, sizei *size, enum *type, char *name ); can be used to enumerate properties of a single output variable captured in transform feedback mode, and is equivalent to const enum props[] = { ARRAY_SIZE, TYPE }; GetProgramResourceName(program, TRANSFORM_FEEDBACK_VARYING, index, bufSize, length, name); GetProgramResourceiv(program, TRANSFORM_FEEDBACK_VARYING, OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

297

index, 1, &props[0], 1, NULL, size); GetProgramResourceiv(program, TRANSFORM_FEEDBACK_VARYING, index, 1, &props[1], 1, NULL, (int *)type);

11.1.3

Shader Execution

If there is an active program object present for the vertex, tessellation control, tessellation evaluation, or geometry shader stages, the executable code for these active programs is used to process incoming vertex values. The following sequence of operations is performed: • Vertices are processed by the vertex shader (see section 11.1) and assembled into primitives as described in sections 10.1 through 10.3. • If the current program contains a tessellation control shader, each individual patch primitive is processed by the tessellation control shader (section 11.2.1). Otherwise, primitives are passed through unmodified. If active, the tessellation control shader consumes its input patch and produces a new patch primitive, which is passed to subsequent pipeline stages. • If the current program contains a tessellation evaluation shader, each individual patch primitive is processed by the tessellation primitive generator (section 11.2.2) and tessellation evaluation shader (see section 11.2.3). Otherwise, primitives are passed through unmodified. When a tessellation evaluation shader is active, the tessellation primitive generator produces a new collection of point, line, or triangle primitives to be passed to subsequent pipeline stages. The vertices of these primitives are processed by the tessellation evaluation shader. The patch primitive passed to the tessellation primitive generator is consumed by this process. • If the current program contains a geometry shader, each individual primitive is processed by the geometry shader (section 11.3). Otherwise, primitives are passed through unmodified. If active, the geometry shader consumes its input patch. However, each geometry shader invocation may emit new vertices, which are arranged into primitives and passed to subsequent pipeline stages. Following shader execution, the fixed-function operations described in chapter 12 are applied. Special considerations for vertex shader execution are described in the following sections. OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS 11.1.3.1

298

Shader Only Texturing

This section describes texture functionality that is accessible through shaders (of all types). Also refer to chapter 8 and to section 8.7 (“Texture Functions”) of the OpenGL ES Shading Language Specification. 11.1.3.2

Texel Fetches

The texelFetch built-ins provide the ability to extract a single texel from a specified texture image. Texel fetches cannot access cube map textures. The integer coordinates (i, j, k) passed to texelFetch are used to pointsample the texture image. The level of detail accessed is computed by adding the specified level-of-detail parameter lod to the base level of the texture, levelbase . Texel fetch proceeds similarly to the steps described for texture access in section 11.1.3.5, with the exception that none of the operations controlled by sampler object state are performed, including: • level of detail clamping; • texture wrap mode application; • filtering (however, a mipmapped minification filter is required to access any level of detail other than the base level); • depth comparison. The steps that are performed are: • validation of texel coordinates as described below, including the computed level-of-detail, (i, j, k), the specified level for array textures, and texture completeness; • sRGB conversion of fetched values as described in section 8.21; • conversion to base color Cb ; • component swizzling. The results of texelFetch built-ins are undefined if any of the following conditions hold: • the computed level of detail is less than the texture’s base level (levelbase ) or greater than the maximum defined level, q (see section 8.14.3) OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

299

• the computed level of detail is not the texture’s base level and the texture’s minification filter is NEAREST or LINEAR • the layer specified for array textures is negative or greater than or equal to the number of layers in the array texture • the texel coordinates (i, j, k) refer to a texel outside the extents of the computed level of detail, where any of i<0

i ≥ ws

j<0

j ≥ hs

k<0

k ≥ ds

and the size parameters ws , hs , and ds refer to the width, height, and depth of the image • the texture being accessed is not complete, as defined in section 8.17. In all the above cases, the result of the texture fetch is undefined in each case. 11.1.3.3

Multisample Texel Fetches

Multisample buffers do not have mipmaps, and there is no level of detail parameter for multisample texel fetches. Instead, an integer parameter selects the sample number to be fetched from the buffer. The number identifying the sample is the same as the value used to query the sample location using GetMultisamplefv. Multisample textures are not filtered when samples are fetched, and filter state is ignored. The results of a multisample texel fetch are undefined if any of the following conditions hold: • the texel coordinates (i, j, k) refer to a texel outside the extents of the multisample texture image, where any of i<0

i ≥ ws

j<0

j ≥ hs

k<0

k ≥ ds

and the size parameters ws , hs , and ds refer to the width, height, and depth of the image • the specified sample number does not exist (is negative, or greater than or equal to the number of samples in the texture). OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

300

Additionally, these fetches may only be performed on a multisample texture sampler. No other sample or fetch commands may be performed on a multisample texture sampler. 11.1.3.4

Texture Queries

The textureSize built-ins provide the ability to query the size of a texture image. The level-of-detail value lod passed in as an argument to the texture size functions is added to the levelbase of the texture to determine a texture image level. The dimensions of that image level are then returned. The value returned is undefined if: • the texture is not complete; • the texture is not mipmap complete, and the computed texture level is not levelbase ; or • the computed texture image level is outside the range [levelbase , q], where q is defined in section 8.14.3. When querying the size of an array texture, both the dimensions and the layer index are returned. 11.1.3.5

Texture Access

Shaders have the ability to do a lookup into a texture map. The maximum number of texture image units available to shaders are the values of the implementationdependant constants • MAX_VERTEX_TEXTURE_IMAGE_UNITS (for vertex shaders), • MAX_TESS_CONTROL_TEXTURE_IMAGE_UNITS (for tessellation control shaders), • MAX_TESS_EVALUATION_TEXTURE_IMAGE_UNITS (for tessellation evaluation shaders), • MAX_GEOMETRY_TEXTURE_IMAGE_UNITS (for geometry shaders), • MAX_TEXTURE_IMAGE_UNITS (for fragment shaders), and • MAX_COMPUTE_TEXTURE_IMAGE_UNITS (for compute shaders),

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

301

All active shaders combined cannot use more than the value of MAX_If more than one pipeline stage accesses the same texture image unit, each such access counts separately against the MAX_COMBINED_TEXTURE_IMAGE_UNITS limit. When a texture lookup is performed in a shader, the filtered texture value τ is computed in the manner described in sections 8.14 and 8.15, and converted to a texture base color Cb as shown in table 14.1, followed by application of the texture swizzle as described in section 14.2.1 to compute the texture source color Cs and As . The resulting four-component vector (Rs , Gs , Bs , As ) is returned to the shader. Texture lookup functions (see section 8.7 (“Texture Functions”) of the OpenGL ES Shading Language Specification) may return floating-point, signed, or unsigned integer values depending on the function and the internal format of the texture. In shaders other than fragment shaders, it is not possible to perform automatic level-of-detail calculations using partial derivatives of the texture coordinates with respect to window coordinates as described in section 8.14. Hence, there is no automatic selection of an image array level. Minification or magnification of a texture map is controlled by a level-of-detail value optionally passed as an argument in the texture lookup functions. If the texture lookup function supplies an explicit levelof-detail value l, then the pre-bias level-of-detail value λbase (x, y) = l (replacing equation 8.6). If the texture lookup function does not supply an explicit level-ofdetail value, then λbase (x, y) = 0. The scale factor ρ(x, y) and its approximation function f (x, y) (see discussion in section 8.14.1) are ignored. Texture lookups involving textures with depth component data generate a texture base color Cb either using depth data directly or by performing a comparison with the Dref value used to perform the lookup, as described in section 8.20.1, and expanding the resulting value Rt to a color Cb = (Rt , 0, 0, 1). In either case, swizzling of Cb is then performed as described above, but only the first component Cs [0] is returned to the shader. The comparison operation is requested in the shader by using any of the shadow sampler types (sampler*Shadow), and in the texture using the TEXTURE_COMPARE_MODE parameter. These requests must be consistent; the results of a texture lookup are undefined if any of the following conditions are true: COMBINED_TEXTURE_IMAGE_UNITS texture image units.

• The sampler used in a texture lookup function is not one of the shadow sampler types, the texture object’s base internal format is DEPTH_COMPONENT or DEPTH_STENCIL, and the TEXTURE_COMPARE_MODE is not NONE. • The sampler used in a texture lookup function is one of the shadow sampler types, the texture object’s base internal format is DEPTH_COMPONENT or DEPTH_STENCIL, and the TEXTURE_COMPARE_MODE is NONE. OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

302

• The sampler used in a texture lookup function is one of the shadow sampler types, and the texture object’s base internal format is not DEPTH_COMPONENT or DEPTH_STENCIL. • The sampler used in a texture lookup function is one of the shadow sampler types, the texture object’s internal format is DEPTH_STENCIL, and the DEPTH_STENCIL_TEXTURE_MODE is not DEPTH_COMPONENT. The stencil index texture internal component is ignored if the base internal format is DEPTH_STENCIL and the value of DEPTH_STENCIL_TEXTURE_MODE is not STENCIL_INDEX. Texture lookups involving texture objects with an internal format of DEPTH_STENCIL can read the stencil value as described in section 8.20 by setting the DEPTH_STENCIL_TEXTURE_MODE to STENCIL_INDEX. Textures with a STENCIL_INDEX base internal format may also be used to read stencil data. The stencil value is read as an integer and assigned to Rt . An unsigned integer sampler should be used to lookup the stencil component, otherwise the results are undefined. If a sampler is used in a shader and the sampler’s associated texture is not complete, as defined in section 8.17, (0.0, 0.0, 0.0, 1.0), in floating-point, will be returned for a non-shadow sampler and 0 for a shadow sampler. In this case, if the sampler is declared in the shader as a signed or unsigned integer sampler type, undefined values are returned as specified in section 9.9 (“Texture Functions”) of the OpenGL ES Shading Language Specification when the texture format and sampler type are unsupported combinations. 11.1.3.6

Atomic Counter Access

Shaders have the ability to set and get atomic counters. The maximum number of atomic counters available to shaders are the values of the implementation dependent constants • MAX_VERTEX_ATOMIC_COUNTERS (for vertex shaders) • MAX_TESS_CONTROL_ATOMIC_COUNTERS shaders),

(for

tessellation

control

• MAX_TESS_EVALUATION_ATOMIC_COUNTERS (for tessellation evaluation shaders), • MAX_GEOMETRY_ATOMIC_COUNTERS (for geometry shaders),

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

303

• MAX_FRAGMENT_ATOMIC_COUNTERS (for fragment shaders), and • MAX_COMPUTE_ATOMIC_COUNTERS (for compute shaders) All active shaders combined cannot use more than the value of MAX_COMBINED_ATOMIC_COUNTERS atomic counters. If more than one pipeline stage accesses the same atomic counter, each such access counts separately against the MAX_COMBINED_ATOMIC_COUNTERS limit. 11.1.3.7

Image Access

Shaders have the ability to read and write to textures using image uniforms. The maximum number of image uniforms available to individual shader stages are the values of the implementation dependent constants • MAX_VERTEX_IMAGE_UNIFORMS (for vertex shaders), • MAX_TESS_CONTROL_IMAGE_UNIFORMS (for tessellation control shaders), • MAX_TESS_EVALUATION_IMAGE_UNIFORMS (for tessellation evaluation shaders), • MAX_GEOMETRY_IMAGE_UNIFORMS (for geometry shaders), • MAX_FRAGMENT_IMAGE_UNIFORMS (for fragment shaders), and • MAX_COMPUTE_IMAGE_UNIFORMS (for compute shaders) All active shaders combined cannot use more than the value of MAX_COMBINED_IMAGE_UNIFORMS image units. If more than one shader stage accesses the same image uniform, each such access counts separately against the MAX_COMBINED_IMAGE_UNIFORMS limit. 11.1.3.8

Shader Storage Buffer Access

Shaders have the ability to read and write to buffer memory via buffer variables in shader storage blocks. The maximum number of shader storage blocks available to shaders are the values of the implementation dependent constants • MAX_VERTEX_SHADER_STORAGE_BLOCKS (for vertex shaders), • MAX_TESS_CONTROL_SHADER_STORAGE_BLOCKS (for tessellation control shaders), OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

304

• MAX_TESS_EVALUATION_SHADER_STORAGE_BLOCKS (for tessellation evaluation shaders), • MAX_GEOMETRY_SHADER_STORAGE_BLOCKS (for geometry shaders), • MAX_FRAGMENT_SHADER_STORAGE_BLOCKS (for fragment shaders), and • MAX_COMPUTE_SHADER_STORAGE_BLOCKS (for compute shaders) All active shaders combined cannot use more than the value of MAX_COMBINED_SHADER_STORAGE_BLOCKS shader storage blocks. If more than one pipeline stage accesses the same shader storage block, each such access separately against this combined limit. 11.1.3.9

Shader Inputs

Besides having access to vertex attributes and uniform variables, vertex shaders can access the read-only built-in variables gl_VertexID and gl_InstanceID. gl_VertexID holds the integer index i implicitly passed by DrawArrays or one of the other drawing commands defined in section 10.5. The value of gl_VertexID is defined if and only if all enabled vertex arrays have non-zero buffer object bindings. gl_InstanceID holds the integer instance number of the current primitive in an instanced draw call (see section 10.5). Section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification also describes these variables. 11.1.3.10

Shader Outputs

A vertex shader can write to user-defined output variables. These values are expected to be interpolated across the primitive it outputs, unless they are specified to be flat shaded. Refer to sections 4.3.6 (“Output Variables”), 7.1 (“Interpolation Qualifiers”), and 7.6 (“Built-In Variables”) of the OpenGL ES Shading Language Specification for more detail. The built-in output gl_Position is intended to hold the homogeneous vertex position. Writing gl_Position is optional. The built-in output gl_PointSize, if written, holds the size of the point to be rasterized, measured in pixels.

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS 11.1.3.11

305

Validation

It is not always possible to determine at link time if a program object can execute successfully, given that LinkProgram can not know the state of the remainder of the pipeline. Therefore validation is done when the first rendering command which triggers shader invocations is issued, to determine if the set of active program objects can be executed. If there is no current program object and no current program pipeline object, the results of rendering commands are undefined. However, this is not an error. Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL or launches compute work if the current set of active program objects cannot be executed, for reasons including: • The current program pipeline object contains a shader interface that doesn’t have an exact match (see section 7.4.1) • A program object is active for at least one, but not all of the shader stages that were present when the program was linked. • One program object is active for at least two shader stages and a second program is active for a shader stage between two stages for which the first program was active. The active compute shader is ignored for the purposes of this test. • There is an active program for tessellation control, tessellation evaluation, or geometry stages with corresponding executable shader, but there is no active program with executable vertex shader. • One but not both of the tessellation control and tessellation evaluation stages have an active program with corresponding executable shader. • There is no current program object specified by UseProgram, there is a current program pipeline object, and the current program for any shader stage has been relinked since being applied to the pipeline object via UseProgramStages with the PROGRAM_SEPARABLE parameter set to FALSE. • There is no current program object specified by UseProgram, there is a

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

306

current program pipeline object, and that object is empty (no executable code is installed for any stage). • Any two active samplers in the set of active program objects are of different types, but refer to the same texture image unit, • The sum of the number of active samplers for each active program exceeds the maximum number of texture image units allowed. • The sum of the number of active shader storage blocks used by the current program objects exceeds the combined limit on the number of active shader storage blocks (the value of MAX_COMBINED_SHADER_STORAGE_BLOCKS). The INVALID_OPERATION error generated by these rendering commands may not provide enough information to find out why the currently active program object would not execute. No information at all is available about a program object that would still execute, but is inefficient or suboptimal given the current GL state. As a development aid, use the command void ValidateProgram( uint program ); to validate the program object program against the current GL state. Each program object has a boolean status, VALIDATE_STATUS, that is modified as a result of validation. This status can be queried with GetProgramiv (see section 7.12). If validation succeeded this status will be set to TRUE, otherwise it will be set to FALSE. If validation succeeded, no INVALID_OPERATION validation error will be generated if program is made current via UseProgram, given the current state. If validation failed, such errors are generated under the current state. ValidateProgram will check for all the conditions described in this section, and may check for other conditions as well. For example, it could give a hint on how to optimize some piece of shader code. The information log of program is overwritten with information on the results of the validation, which could be an empty string. The results written to the information log are typically only useful during application development; an application should not expect different GL implementations to produce identical information. A shader should not fail to compile, and a program object should not fail to link due to lack of instruction space or lack of temporary variables. Implementations should ensure that all valid shaders and program objects may be successfully compiled, linked and executed.

OpenGL ES 3.2 (May 5, 2022)

11.1. VERTEX SHADERS

307

Errors An INVALID_VALUE error is generated if program is not the name of either a program or shader object. An INVALID_OPERATION error is generated if program is the name of a shader object. Separable program objects may have validation failures that cannot be detected without the complete program pipeline. Mismatched interfaces, improper usage of program objects together, and the same state-dependent failures can result in validation errors for such program objects. As a development aid, use the command void ValidateProgramPipeline( uint pipeline ); to validate the program pipeline object pipeline against the current GL state. Each program pipeline object has a boolean status, VALIDATE_STATUS, that is modified as a result of validation. This status can be queried with GetProgramPipelineiv (see section 7.12). If validation succeeded, no INVALID_OPERATION validation error will be generated if pipeline is bound and no program were made current via UseProgram, given the current state. If validation failed, such errors are generated under the current state. If pipeline is a name that has been generated (without subsequent deletion) by GenProgramPipelines, but refers to a program pipeline object that has not been previously bound, the GL first creates a new state vector in the same manner as when BindProgramPipeline creates a new program pipeline object. Errors An INVALID_OPERATION error is generated if pipeline is not a name returned from a previous call to GenProgramPipelines or if such a name has since been deleted by DeleteProgramPipelines, 11.1.3.12

Undefined Behavior

When using array or matrix variables in a shader, it is possible to access a variable with an index computed at run time that is outside the declared extent of the variable. Such out-of-bounds accesses have undefined behavior, and system errors (possibly including program termination) may occur. The level of protection provided against such errors in the shader is implementation-dependent. Applications that require defined behavior for out-of-bounds accesses should range check all computed indices before dereferencing the array, vector or matrix. OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

11.2

308

Tessellation

Tessellation is a process that reads a patch primitive and generates new primitives used by subsequent pipeline stages. The generated primitives are formed by subdividing a single triangle or quad primitive according to fixed or shader-computed levels of detail and transforming each of the vertices produced during this subdivision. Tessellation functionality is controlled by two types of tessellation shaders: tessellation control shaders and tessellation evaluation shaders. Tessellation is considered active if and only if the active program object or program pipeline object includes both a tessellation control shader and a tessellation evaluation shader. The tessellation control shader is used to read an input patch provided by the application, and emit an output patch. The tessellation control shader is run once for each vertex in the output patch and computes the attributes of that vertex. Additionally, the tessellation control shader may compute additional per-patch attributes of the output patch. The most important per-patch outputs are the tessellation levels, which are used to control the number of subdivisions performed by the tessellation primitive generator. The tessellation control shader may also write additional per-patch attributes for use by the tessellation evaluation shader. If no tessellation control shader is active, patch primitives may not be provided by the application. If a tessellation evaluation shader is active, the tessellation primitive generator subdivides a triangle or quad primitive into a collection of points, lines, or triangles according to the tessellation levels of the patch and the set of layout declarations specified in the tessellation evaluation shader text. When a tessellation evaluation shader is active, it is run on each vertex generated by the tessellation primitive generator to compute the final position and other attributes of the vertex. The tessellation evaluation shader can read the relative location of the vertex in the subdivided output primitive, given by an (u, v) or (u, v, w) coordinate, as well as the position and attributes of any or all of the vertices in the input patch. Tessellation operates only on patch primitives. Patch primitives are not supported by pipeline stages below the tessellation evaluation shader. A non-separable program object or program pipeline object that includes a tessellation shader of any kind must also include a vertex shader. Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if the current program state has one but not both of a

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

309

tessellation control shader and tessellation evaluation shader. An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if tessellation is active and the primitive mode is not PATCHES. An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if tessellation is not active and the primitive mode is PATCHES. An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if the current program state has a tessellation shader but no vertex shader.

11.2.1

Tessellation Control Shaders

The tessellation control shader consumes an input patch provided by the application and emits a new output patch. The input patch is an array of vertices with attributes corresponding to output variables written by the vertex shader. The output patch consists of an array of vertices with attributes corresponding to per-vertex output variables written by the tessellation control shader and a set of per-patch attributes corresponding to per-patch output variables written by the tessellation control shader. Tessellation control output variables are per-vertex by default, but may be declared as per-patch using the patch qualifier. The number of vertices in the output patch is fixed when the program is linked, and is specified in tessellation control shader source code using the output layout qualifier vertices, as described in the OpenGL ES Shading Language Specification. A program will fail to link if the output patch vertex count is not specified by the tessellation control shader object attached to the program, if it is less than or equal to zero, or if it is greater than the implementation-dependent maximum patch size. The output patch vertex count may be queried by calling GetProgramiv with pname TESS_CONTROL_OUTPUT_VERTICES. Tessellation control shaders are created as described in section 7.1, using a type of TESS_CONTROL_SHADER. When a new input patch is received, the tessellation control shader is run once for each vertex in the output patch. The tessellation control shader invocations collectively specify the per-vertex and per-patch attributes of the output patch. The per-vertex attributes are obtained from the per-vertex output variables written by each invocation. Each tessellation control shader invocation may only write to per-vertex output variables corresponding to its own output patch vertex. The output patch vertex number corresponding to a given tessellation control shader invocation is given by the built-in variable gl_InvocationID. Perpatch attributes are taken from the per-patch output variables, which may be written by any tessellation control shader invocation. While tessellation control shader OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

310

invocations may read any per-vertex and per-patch output variable and write any per-patch output variable, reading or writing output variables also written by other invocations has ordering hazards discussed below. 11.2.1.1

Tessellation Control Shader Variables

Tessellation control shaders can access uniforms belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. Tessellation control shaders also have access to samplers to perform texturing operations, as described in section 7.9. Tessellation control shaders can access the transformed attributes of all vertices for their input primitive using input variables. A vertex shader writing to output variables generates the values of these input variables. Values for any inputs that are not written by a vertex shader are undefined. Additionally, tessellation control shaders can write to one or more output variables, including per-vertex attributes for the vertices of the output patch and perpatch attributes of the patch. Tessellation control shaders can also write to a set of built-in per-vertex and per-patch outputs defined in the OpenGL ES Shading Language. The per-vertex and per-patch attributes of the output patch are used by the tessellation primitive generator (section 11.2.2) and may be read by tessellation evaluation shader (section 11.2.3). 11.2.1.2

Tessellation Control Shader Execution Environment

If there is an active program for the tessellation control stage, the executable version of the program’s tessellation control shader is used to process patches resulting from the primitive assembly stage. When tessellation control shader execution completes, the input patch is consumed. A new patch is assembled from the per-vertex and per-patch output variables written by the shader and is passed to subsequent pipeline stages. There are several special considerations for tessellation control shader execution described in the following sections.

11.2.1.2.1 Texture Access Section 11.1.3.1 describes texture lookup functionality accessible to a vertex shader. The texel fetch and texture size query functionality described there also applies to tessellation control shaders.

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

311

11.2.1.2.2 Tessellation Control Shader Inputs Section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification describes the builtin variable array gl_in available as input to a tessellation control shader. gl_in receives values from equivalent built-in output variables written by the vertex shader (section 11.1.3). Each array element of gl_in is a structure holding a value for a specific vertex of the input patch. The length of gl_in is equal to the implementation-dependent maximum patch size (gl_MaxPatchVertices). Behavior is undefined if gl_in is indexed with a vertex index greater than or equal to the current patch size. The sole member of each element of the gl_in array is gl_Position. Tessellation control shaders have available several other built-in input variables not replicated per-vertex and not contained in gl_in, including: • The variable gl_PatchVerticesIn holds the number of vertices in the input patch being processed by the tessellation control shader. • The variable gl_PrimitiveID is filled with the number of primitives processed by the drawing command which generated the input vertices. The first primitive generated by a drawing command is numbered zero, and the primitive ID counter is incremented after every individual point, line, or triangle primitive is processed. The counter is reset to zero between each instance drawn. Restarting a primitive topology using the primitive restart index has no effect on the primitive ID counter. • The variable gl_InvocationID holds an invocation number for the current tessellation control shader invocation. Tessellation control shaders are invoked once per output patch vertex, and invocations are numbered beginning with zero. Similarly to the built-in inputs, each user-defined input variable has a value for each vertex and thus needs to be declared as arrays or inside input blocks declared as arrays. Declaring an array size is optional. If no size is specified, it will be taken from the implementation-dependent maximum patch size (gl_MaxPatchVertices). If a size is specified, it must match the maximum patch size; otherwise, a compile or link error will occur. Since the array size may be larger than the number of vertices found in the input patch, behavior is undefined if a per-vertex input variable is accessed using an index greater than or equal to the number of vertices in the input patch. Similarly to the limit on vertex shader output components (see section 11.1.2.1), there is a limit on the number of components of input variables

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

312

that can be read by the tessellation control shader, given by the value of the implementation-dependent constant MAX_TESS_CONTROL_INPUT_COMPONENTS. When a program is linked, all components of any input read by a tessellation control shader will count against this limit. A program whose tessellation control shader exceeds this limit may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS (see section 11.1.2.1).

11.2.1.2.3 Tessellation Control Shader Outputs Section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification describes the built-in variable array gl_out available as an output for a tessellation control shader. gl_out passes values to equivalent built-in input variables read by subsequent shader stages or to subsequent fixed functionality vertex processing pipeline stages. Each array element of gl_out is a structure holding values for a specific vertex of the output patch. The length of gl_out is equal to the output patch size specified in the tessellation control shader output layout declaration. The sole member of each element of the gl_out array is gl_Position. It behaves identically to the equivalently named vertex shader output (see section 11.1.3). Tessellation shaders additionally have three built-in per-patch output arrays, gl_TessLevelOuter, gl_TessLevelInner and gl_BoundingBox. These arrays are not replicated for each output patch vertex and are not members of gl_out. gl_TessLevelOuter is an array of four floating-point values specifying the approximate number of segments that the tessellation primitive generator should use when subdividing each outer edge of the primitive it subdivides. gl_TessLevelInner is an array of two floating-point values specifying the approximate number of segments used to produce a regularly-subdivided primitive interior. The values written to gl_TessLevelOuter and gl_TessLevelInner need not be integers, and their interpretation depends on the type of primitive the tessellation primitive generator will subdivide and other tessellation parameters, as discussed in the following section. gl_BoundingBox is an array of two vec4 values that should be used instead of the value of PRIMITIVE_BOUNDING_BOX as the primitive bounding box (see section 13.2) for primitives generated from the output patch. A tessellation control shader may also declare user-defined per-vertex output variables. User-defined per-vertex output variables are declared with the qualifier out and have a value for each vertex in the output patch. Such variables must be declared as arrays or inside output blocks declared as arrays. Declaring an array size is optional. If no size is specified, it will be taken from the output patch size OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

313

declared in the shader. If a size is specified, it must match the maximum patch size; otherwise, a compile or link error will occur. While per-vertex output variables are declared as arrays indexed by vertex number, each tessellation control shader invocation may write only to those outputs corresponding to its output patch vertex. Tessellation control shaders must use the special variable gl_InvocationID as the vertex number index when writing to per-vertex output variables. Additionally, a tessellation control shader may declare per-patch output variables using the qualifier patch out. Unlike per-vertex outputs, per-patch outputs do not correspond to any specific vertex in the patch, and are not indexed by vertex number. Per-patch outputs declared as arrays have multiple values for the output patch; similarly declared per-vertex outputs would indicate a single value for each vertex in the output patch. User-defined per-patch outputs are not used by the tessellation primitive generator, but may be read by tessellation evaluation shaders. There are several limits on the number of components of output variables that can be written by the tessellation control shader. The number of components of active per-vertex output variables may not exceed the value of MAX_TESS_CONTROL_OUTPUT_COMPONENTS. The number of components of active per-patch output variables may not exceed the value of MAX_TESS_PATCH_COMPONENTS. The built-in outputs gl_TessLevelOuter and gl_TessLevelInner are not counted against the per-patch limit. The built-in output gl_BoundingBox, if statically assigned by the shader, is counted against the per-patch limit. The total number of components of active per-vertex and per-patch outputs is derived by multiplying the per-vertex output component count by the output patch size and then adding the per-patch output component count. The total component count may not exceed MAX_TESS_CONTROL_TOTAL_OUTPUT_COMPONENTS. When a program is linked, all components of any output variable written by a tessellation control shader will count against this limit. A program exceeding any of these limits may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS (see section 11.1.2.1).

11.2.1.2.4 Tessellation Control Shader Execution Order For tessellation control shaders with a declared output patch size greater than one, the shader is invoked more than once for each input patch. The order of execution of one tessellation control shader invocation relative to the other invocations for the same input patch is largely undefined. The built-in function barrier provides some control over relative execution order. When a tessellation control shader calls the barrier OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

314

function, its execution pauses until all other invocations have also called the same function. Output variable assignments performed by any invocation executed prior to calling barrier will be visible to any other invocation after the call to barrier returns. Shader output values read in one invocation but written by another may be undefined without proper use of barrier; full rules are found in the OpenGL ES Shading Language Specification. The barrier function may only be called inside the main entry point of the tessellation control shader and may not be called in code containing potentially divergent flow of control. In particular, barrier may not be called inside a switch statement, in either sub-statement of an if statement, inside a do, for, or while loop, or at any point after a return statement in the function main.

11.2.2

Tessellation Primitive Generation

The tessellation primitive generator consumes the input patch and produces a new set of basic primitives (points, lines, or triangles). These primitives are produced by subdividing a geometric primitive (rectangle or triangle) according to the perpatch tessellation levels written by the tessellation control shader. This subdivision is performed in an implementation- dependent manner. The type of subdivision performed by the tessellation primitive generator is specified by an input layout declaration in the tessellation evaluation shader using one of the identifiers triangles, quads, and isolines. For triangles, the primitive generator subdivides a triangle primitive into smaller triangles. For quads, the primitive generator subdivides a rectangle primitive into smaller triangles. For isolines, the primitive generator subdivides a rectangle primitive into a collection of line segments arranged in strips stretching horizontally across the rectangle. Each vertex produced by the primitive generator has an associated (u, v, w) or (u, v) position in a normalized parameter space, with parameter values in the range [0, 1], as illustrated in figure 11.1. For triangles, the vertex position is a barycentric coordinate (u, v, w), where u + v + w = 1, and indicates the relative influence of the three vertices of the triangle on the position of the vertex. For quads and isolines, the position is a (u, v) coordinate indicating the relative horizontal and vertical position of the vertex relative to the subdivided rectangle. The subdivision process is explained in more detail in subsequent sections. A patch is discarded by the tessellation primitive generator if any relevant outer tessellation level is less than or equal to zero. Patches will also be discarded if any relevant outer tessellation level corresponds to a floating-point NaN (not a number) in implementations supporting NaN. When patches are discarded, no new primitives will be generated and the tessellation evaluation program will not be run. For quads, all four outer levels are relevant. For triangles and isolines, only OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

315

Figure 11.1. Domain parameterization for tessellation generator primitive modes (triangles, quads, or isolines). The coordinates illustrate the value of gl_TessCoord at the corners of the domain. The labels on the edges indicate the inner (IL0 and IL1) and outer (OL0 through OL3) tessellation level values used to control the number of subdivisions along each edge of the domain.

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

316

the first three or two outer levels, respectively, are relevant. Negative inner levels will not cause a patch to be discarded; they will be clamped as described below. Each of the tessellation levels is used to determine the number and spacing of segments used to subdivide a corresponding edge. The method used to derive the number and spacing of segments is specified by an input layout declaration in the tessellation evaluation shader using one of the identifiers equal_spacing, fractional_even_spacing, or fractional_odd_spacing. If no spacing is specified in the tessellation evaluation shader, equal_spacing will be used. If equal_spacing is used, the floating-point tessellation level is first clamped to the range [1, max], where max is the implementation-dependent maximum tessellation level (the value of MAX_TESS_GEN_LEVEL). The result is rounded up to the nearest integer n, and the corresponding edge is divided into n segments of equal length in (u, v) space. If fractional_even_spacing is used, the tessellation level is first clamped to the range [2, max] and then rounded up to the nearest even integer n. If fractional_odd_spacing is used, the tessellation level is clamped to the range [1, max − 1] and then rounded up to the nearest odd integer n. If n is one, the edge will not be subdivided. Otherwise, the corresponding edge will be divided into n − 2 segments of equal length, and two additional segments of equal length that are typically shorter than the other segments. The length of the two additional segments relative to the others will decrease monotonically with the value of n − f , where f is the clamped floating-point tessellation level. When n − f is zero, the additional segments will have equal length to the other segments. As n − f approaches 2.0, the relative length of the additional segments approaches zero. The two additional segments should be placed symmetrically on opposite sides of the subdivided edge. The relative location of these two segments is undefined, but must be identical for any pair of subdivided edges with identical values of f . When the tessellation primitive generator produces triangles (in the triangles or quads modes), the orientation of all triangles can be specified by an input layout declaration in the tessellation evaluation shader using the identifiers cw and ccw. If the order is cw, the vertices of all generated triangles will have a clockwise ordering in (u, v) or (u, v, w) space, as illustrated in figure 11.1. If the order is ccw, the vertices will be specified in counter-clockwise order. If no layout is specified, ccw will be used. For all primitive modes, the tessellation primitive generator is capable of generating points instead of lines or triangles. If an input layout declaration in the tessellation evaluation shader specifies the identifier point_mode, the primitive generator will generate one point for each distinct vertex produced by tessellation. Otherwise, the primitive generator will produce a collection of line segments or triangles according to the primitive mode. When tessellating triangles or quads in OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

317

point mode with fractional odd spacing, the tessellation primitive generator may produce “interior” vertices that are positioned on the edge of the patch if an inner tessellation level is less than or equal to one. Such vertices are considered distinct from vertices produced by subdividing the outer edge of the patch, even if there are pairs of vertices with identical coordinates. The points, lines, or triangles produced by the tessellation primitive generator are passed to subsequent pipeline stages in an implementation-dependent order. 11.2.2.1

Triangle Tessellation

If the tessellation primitive mode is triangles, an equilateral triangle is subdivided into a collection of triangles covering the area of the original triangle. First, the original triangle is subdivided into a collection of concentric equilateral triangles. The edges of each of these triangles are subdivided, and the area between each triangle pair is filled by triangles produced by joining the vertices on the subdivided edges. The number of concentric triangles and the number of subdivisions along each triangle except the outermost is derived from the first inner tessellation level. The edges of the outermost triangle are subdivided independently, using the first, second, and third outer tessellation levels to control the number of subdivisions of the u = 0 (left), v = 0 (bottom), and w = 0 (right) edges, respectively. The second inner tessellation level and the fourth outer tessellation level have no effect in this mode. If the first inner tessellation level and all three outer tessellation levels are exactly one after clamping and rounding, only a single triangle with (u, v, w) coordinates of (0, 0, 1), (1, 0, 0), and (0, 1, 0) is generated. If the inner tessellation level is one and any of the outer tessellation levels is greater than one, the inner tessellation level is treated as though it were originally specified as 1 +  and will be rounded up to result in a two- or three-segment subdivision according to the tessellation spacing. When used with fractional odd spacing, the three-segment subdivision may produce “inner” vertices positioned on the edge of the triangle. If any tessellation level is greater than one, tessellation begins by producing a set of concentric inner triangles and subdividing their edges. First, the three outer edges are temporarily subdivided using the clamped and rounded first inner tessellation level and the specified tessellation spacing, generating n segments. For the outermost inner triangle, the inner triangle is degenerate – a single point at the center of the triangle – if n is two. Otherwise, for each corner of the outer triangle, an inner triangle corner is produced at the intersection of two lines extended perpendicular to the corner’s two adjacent edges running through the vertex of the subdivided outer edge nearest that corner. If n is three, the edges of the inner triangle are not subdivided and it is the final triangle in the set of concentric triangles. OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

318

Figure 11.2. Inner triangle tessellation with inner tessellation levels of (a) five and (b) four, respectively (not to scale). Solid black circles depict vertices along the edges of the concentric triangles. The edges of inner triangles are subdivided by intersecting the edge with segments perpendicular to the edge passing through each inner vertex of the subdivided outer edge. Dotted lines depict edges connecting corresponding vertices on the inner and outer triangle edges.

Otherwise, each edge of the inner triangle is divided into n − 2 segments, with the n − 1 vertices of this subdivision produced by intersecting the inner edge with lines perpendicular to the edge running through the n − 1 innermost vertices of the subdivision of the outer edge. Once the outermost inner triangle is subdivided, the previous subdivision process repeats itself, using the generated triangle as an outer triangle. This subdivision process is illustrated in figure 11.2. Once all the concentric triangles are produced and their edges are subdivided, the area between each pair of adjacent inner triangles is filled completely with a set of non-overlapping triangles. In this subdivision, two of the three vertices of each triangle are taken from adjacent vertices on a subdivided edge of one triangle; the third is one of the vertices on the corresponding edge of the other triangle. If the innermost triangle is degenerate (i.e., a point), the triangle containing it is subdivided into six triangles by connecting each of the six vertices on that triangle with the center point. If the innermost triangle is not degenerate, that triangle is added to the set of generated triangles as-is. OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

319

After the area corresponding to any inner triangles is filled, the primitive generator generates triangles to cover area between the outermost triangle and the outermost inner triangle. To do this, the temporary subdivision of the outer triangle edge above is discarded. Instead, the u = 0, v = 0, and w = 0 edges are subdivided according to the first, second, and third outer tessellation levels, respectively, and the tessellation spacing. The original subdivision of the first inner triangle is retained. The area between the outer and first inner triangles is completely filled by non-overlapping triangles as described above. If the first (and only) inner triangle is degenerate, a set of triangles is produced by connecting each vertex on the outer triangle edges with the center point. After all triangles are generated, each vertex in the subdivided triangle is assigned a barycentric (u, v, w) coordinate based on its location relative to the three vertices of the outer triangle. The algorithm used to subdivide the triangular domain in (u, v, w) space into individual triangles is implementation-dependent. However, the set of triangles produced will completely cover the domain, and no portion of the domain will be covered by multiple triangles. The order in which the generated triangles passed to subsequent pipeline stages and the order of the vertices in those triangles are both implementation-dependent. However, when depicted in a manner similar to figure 11.2, the order of the vertices in the generated triangles will be either all clockwise or all counter-clockwise, according to the vertex order layout declaration. 11.2.2.2

Quad Tessellation

If the tessellation primitive mode is quads, a rectangle is subdivided into a collection of triangles covering the area of the original rectangle. First, the original rectangle is subdivided into a regular mesh of rectangles, where the number of rectangles along the u = 0 and u = 1 (vertical) and v = 0 and v = 1 (horizontal) edges are derived from the first and second inner tessellation levels, respectively. All rectangles, except those adjacent to one of the outer rectangle edges, are decomposed into triangle pairs. The outermost rectangle edges are subdivided independently, using the first, second, third, and fourth outer tessellation levels to control the number of subdivisions of the u = 0 (left), v = 0 (bottom), u = 1 (right), and v = 1 (top) edges, respectively. The area between the inner rectangles of the mesh and the outer rectangle edges is filled by triangles produced by joining the vertices on the subdivided outer edges to the vertices on the edge of the inner rectangle mesh. If both clamped inner tessellation levels and all four clamped outer tessellation levels are exactly one, only a single triangle pair covering the outer rectangle OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

320

is generated. Otherwise, if either clamped inner tessellation level is one, that tessellation level is treated as though it were originally specified as 1 +  and will result in a two- or three-segment subdivision depending on the tessellation spacing. When used with fractional odd spacing, the three-segment subdivision may produce “inner” vertices positioned on the edge of the rectangle. If any tessellation level is greater than one, tessellation begins by subdividing the u = 0 and u = 1 edges of the outer rectangle into m segments using the clamped and rounded first inner tessellation level and the tessellation spacing. The v = 0 and v = 1 edges are subdivided into n segments using the second inner tessellation level. Each vertex on the u = 0 and v = 0 edges is joined with the corresponding vertex on the u = 1 and v = 1 edges to produce a set of vertical and horizontal lines that divide the rectangle into a grid of smaller rectangles. The primitive generator emits a pair of non-overlapping triangles covering each such rectangle not adjacent to an edge of the outer rectangle. The boundary of the region covered by these triangles forms an inner rectangle, the edges of which are subdivided by the grid vertices that lie on the edge. If either m or n is two, the inner rectangle is degenerate, and one or both of the rectangle’s “edges” consist of a single point. This subdivision is illustrated in figure 11.3. After the area corresponding to the inner rectangle is filled, the primitive generator must produce triangles to cover area between the inner and outer rectangles. To do this, the subdivision of the outer rectangle edge above is discarded. Instead, the u = 0, v = 0, u = 1, and v = 1 edges are subdivided according to the first, second, third, and fourth outer tessellation levels, respectively, and the tessellation spacing. The original subdivision of the inner rectangle is retained. The area between the outer and inner rectangles is completely filled by non-overlapping triangles. Two of the three vertices of each triangle are adjacent vertices on a subdivided edge of one rectangle; the third is one of the vertices on the corresponding edge of the other rectangle. If either edge of the innermost rectangle is degenerate, the area near the corresponding outer edges is filled by connecting each vertex on the outer edge with the single vertex making up the inner “edge”. The algorithm used to subdivide the rectangular domain in (u, v) space into individual triangles is implementation-dependent. However, the set of triangles produced will completely cover the domain, and no portion of the domain will be covered by multiple triangles. The order in which the generated triangles are passed to subsequent pipeline stages and the order of the vertices in those triangles are both implementation-dependent. However, when depicted in a manner similar to figure 11.3, the order of the vertices in the generated triangles will be either all clockwise or all counter-clockwise, according to the vertex order layout declaration.

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

321

Figure 11.3. Inner quad tessellation with inner tessellation levels of (a) (4, 2) and (b) (7, 4), respectively. Gray regions on the bottom figure depict the 10 inner rectangles, each of which will be subdivided into two triangles. Solid black circles depict vertices on the boundary of the outer and inner rectangles, where the inner rectangle on the top figure is degenerate (a single line segment). Dotted lines depict the horizontal and vertical edges connecting corresponding vertices on the inner and outer rectangle edges.

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION 11.2.2.3

322

Isoline Tessellation

If the tessellation primitive mode is isolines, a set of independent horizontal line segments is drawn. The segments are arranged into connected strips called isolines, where the vertices of each isoline have a constant v coordinate and u coordinates covering the full range [0, 1]. The number of isolines generated is derived from the first outer tessellation level; the number of segments in each isoline is derived from the second outer tessellation level. Both inner tessellation levels and the third and fourth outer tessellation levels have no effect in this mode. As with quad tessellation above, isoline tessellation begins with a rectangle. The u = 0 and u = 1 edges of the rectangle are subdivided according to the first outer tessellation level. For the purposes of this subdivision, the tessellation spacing is ignored and treated as equal_spacing. An isoline is drawn connecting each vertex on the u = 0 rectangle edge with the corresponding vertex on the u = 1 rectangle edge, except that no line is drawn between (0, 1) and (1, 1). If the number of isolines on the subdivided u = 0 and u = 1 edges is n, this process will result in n equally spaced lines with constant v coordinates of 0, n1 , n2 , . . . , n−1 n . Each of the n isolines is then subdivided according to the second outer tessellation level and the tessellation spacing, resulting in m line segments. Each segment of each line is emitted by the tessellation primitive generator, as illustrated in figure 11.4. The order in which the generated line segments are passed to subsequent pipeline stages and the order of the vertices in each generated line segment are both implementation-dependent.

11.2.3

Tessellation Evaluation Shaders

If active, the tessellation evaluation shader takes the (u, v) or (u, v, w) location of each vertex in the primitive subdivided by the tessellation primitive generator, and generates a vertex with a position and associated attributes. The tessellation evaluation shader can read any of the vertices of its input patch, which is the output patch produced by the tessellation control shader. Tessellation evaluation shaders are created as described in section 7.1, using a type of TESS_EVALUATION_SHADER. Each invocation of the tessellation evaluation shader writes the attributes of exactly one vertex. The number of vertices evaluated per patch depends on the tessellation level values computed by the tessellation control shaders. Tessellation evaluation shader invocations run independently, and no invocation can access the variables belonging to another invocation. All invocations are capable of accessing all the vertices of their corresponding input patch. The number of the vertices in the input patch is fixed and is equal to the tessel-

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

323

Figure 11.4. Isoline tessellation with the first two outer tessellation levels of (a) (1, 3) and (b) (4, 6), respectively. Line segments connecting the vertices marked with solid black circles are emitted by the primitive generator. Vertices marked with empty circles correspond to (u, v) coordinates of (0, 1) and (1, 1), where no line segments are generated.

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

324

lation control shader output patch size parameter in effect when the program was last linked. 11.2.3.1

Tessellation Evaluation Shader Variables

Tessellation evaluation shaders can access uniforms belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. Tessellation evaluation shaders also have access to samplers to perform texturing operations, as described in section 7.9. Tessellation evaluation shaders can access the transformed attributes of all vertices for their input primitive using input variables. A tessellation control shader writing to output variables generates the values of these input variables. Values for any input variables that are not written by a tessellation control shader are undefined. Additionally, tessellation evaluation shaders can write to one or more output variables that will be passed to subsequent programmable shader stages or fixed functionality vertex pipeline stages. 11.2.3.2

Tessellation Evaluation Shader Execution Environment

If there is an active program for the tessellation evaluation stage, the executable version of the program’s tessellation evaluation shader is used to process vertices produced by the tessellation primitive generator. During this processing, the shader may access the input patch processed by the primitive generator. When tessellation evaluation shader execution completes, a new vertex is assembled from the output variables written by the shader and is passed to subsequent pipeline stages. There are several special considerations for tessellation evaluation shader execution described in the following sections.

11.2.3.2.1 Texture Access Section 11.1.3.1 describes texture lookup functionality accessible to a vertex shader. The texel fetch and texture size query functionality described there also applies to tessellation evaluation shaders. 11.2.3.3

Tessellation Evaluation Shader Inputs

Section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification describes the built-in variable array gl_in available as input to a tessellation evaluation shader. gl_in receives values from equivalent built-in output variables

OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

325

written by a previous shader (section 11.1.3). Each array element of gl_in is a structure holding values for a specific vertex of the input patch. The length of gl_in is equal to the implementation-dependent maximum patch size (gl_MaxPatchVertices). Behavior is undefined if gl_in is indexed with a vertex index greater than or equal to the current patch size. The sole member of each element of the gl_in array is gl_Position. Tessellation evaluation shaders have available several other built-in input variables not replicated per-vertex and not contained in gl_in, including: • The variables gl_PatchVerticesIn and gl_PrimitiveID are filled with the number of the vertices in the input patch and a primitive number, respectively. They behave exactly as the identically named inputs for tessellation control shaders. • The variable gl_TessCoord is a three-component floating-point vector consisting of the (u, v, w) coordinate of the vertex being processed by the tessellation evaluation shader. The values of u, v, and w are in the range [0, 1], and vary linearly across the primitive being subdivided. For tessellation primitive modes of quads or isolines, the w value is always zero. The (u, v, w) coordinates are generated by the tessellation primitive generator in a manner dependent on the primitive mode, as described in section 11.2.2. gl_TessCoord is not an array; it specifies the location of the vertex being processed by the tessellation evaluation shader, not of any vertex in the input patch. • The variables gl_TessLevelOuter and gl_TessLevelInner are arrays holding outer and inner tessellation levels of the patch, as used by the tessellation primitive generator. Tessellation level values loaded in these variables will be prior to the clamping and rounding operations performed by the primitive generator as described in section 11.2.2. For triangular tessellation, gl_TessLevelOuter[3] and gl_TessLevelInner[1] will be undefined. For isoline tessellation, gl_TessLevelOuter[2], gl_TessLevelOuter[3], and both values in gl_TessLevelInner are undefined. The special tessellation control shader output gl_BoundingBox is consumed by the tessellation primitive generator, and is not available as an input to the tessellation evaluation shader. A tessellation evaluation shader may also declare user-defined per-vertex input variables. User-defined per-vertex input variables are declared with the qualifier in and have a value for each vertex in the input patch. User-defined per-vertex OpenGL ES 3.2 (May 5, 2022)

11.2. TESSELLATION

326

input variables have a value for each vertex and thus need to be declared as arrays or inside input blocks declared as arrays. Declaring an array size is optional. If no size is specified, it will be taken from the implementation-dependent maximum patch size (gl_MaxPatchVertices). If a size is specified, it must match the maximum patch size; otherwise, a compile or link error will occur. Since the array size may be larger than the number of vertices found in the input patch, behavior is undefined if a per-vertex input variable is accessed using an index greater than or equal to the number of vertices in the input patch. Additionally, a tessellation evaluation shader may declare per-patch input variables using the qualifier patch in. Unlike per-vertex inputs, per-patch inputs do not correspond to any specific vertex in the patch, and are not indexed by vertex number. Per-patch inputs declared as arrays have multiple values for the input patch; similarly declared per-vertex inputs would indicate a single value for each vertex in the output patch. User-defined per-patch input variables are filled with corresponding per-patch output values written by the tessellation control shader. Similarly to the limit on vertex shader output components (see section 11.1.2.1), there is a limit on the number of components of per-vertex and per-patch input variables that can be read by the tessellation evaluation shader, given by the values of the implementation-dependent constants MAX_TESS_EVALUATION_INPUT_COMPONENTS and MAX_TESS_PATCH_COMPONENTS, respectively. The built-in inputs gl_TessLevelOuter and gl_TessLevelInner are not counted against the per-patch limit. When a program is linked, all components of any input variable read by a tessellation evaluation shader will count against this limit. A program whose tessellation evaluation shader exceeds this limit may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS (see section 11.1.2.1). 11.2.3.4

Tessellation Evaluation Shader Outputs

Tessellation evaluation shaders have a built-in output variable used to pass values to an equivalent built-in input variable read by subsequent shader stages or to subsequent fixed functionality vertex processing pipeline stages. This variable is gl_Position, and behaves identically to the equivalently named vertex shader output (see section 11.1.3). A tessellation evaluation shader may also declare userdefined per-vertex output variables. Similarly to the limit on vertex shader output components (see section 11.1.2.1), there is a limit on the number of components of output variables that can be written by the tessellation evaluation shader, given by the values OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

327

of the implementation-dependent constant MAX_TESS_EVALUATION_OUTPUT_COMPONENTS. When a program is linked, all components of any output variable written by a tessellation evaluation shader will count against this limit. A program whose tessellation evaluation shader exceeds this limit may fail to link, unless devicedependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS. (see section 11.1.2.1).

11.3

Geometry Shaders

After vertices are processed, they are arranged into primitives, as described in section 10.1. This section describes geometry shaders, an additional pipeline stage defining operations to further process those primitives. Geometry shaders are defined by source code in the OpenGL ES Shading Language, in the same manner as vertex shaders. They operate on a single primitive at a time and emit one or more output primitives, all of the same type, which are then processed like an equivalent GL primitive specified by the application. The original primitive is discarded after geometry shader execution. The inputs available to a geometry shader are the transformed attributes of all the vertices that belong to the primitive. Additional adjacency primitives are available which also make the transformed attributes of neighboring vertices available to the shader. The results of the shader are a new set of transformed vertices, arranged into primitives by the shader. The geometry shader pipeline stage is inserted after primitive assembly, prior to transform feedback (see section 12.2). Geometry shaders are created as described in section 7.1 using a type of GEOMETRY_SHADER. They are attached to and used in program objects as described in section 7.3. When the program object currently in use includes a geometry shader, its geometry shader is considered active, and is used to process primitives. If the program object has no geometry shader, this stage is bypassed. A non-separable program object or program pipeline object that includes a geometry shader must also include a vertex shader. Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if the current program state has a geometry shader but no vertex shader.

OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

11.3.1

328

Geometry Shader Input Primitives

A geometry shader can operate on one of five input primitive types. Depending on the input primitive type, one to six input vertices are available when the shader is executed. Each input primitive type supports a subset of the primitives provided by the GL. Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if a geometry shader is active and the primitive mode parameter is incompatible with the input primitive type of the geometry shader of the active geometry program object, as discussed below. If a tessellation evaluation shader is not active, the mode parameter passed to drawing commands is used for purposes of this error check. Otherwise, the type of primitive emitted by that shader is used. A geometry shader that accesses more input vertices than are available for a given input primitive type can be successfully compiled, because the input primitive type is not part of the shader object. However, a program object containing a shader object that accesses more input vertices than are available for the input primitive type of the program object will not link. The input primitive type is specified in the geometry shader source code using an input layout qualifier, as described in the OpenGL ES Shading Language Specification. A program will fail to link if the input primitive type is not specified by the geometry shader object attached to the program. The input primitive type may be queried by calling GetProgramiv with pname GEOMETRY_INPUT_TYPE. The supported types and the corresponding OpenGL ES Shading Language input layout qualifier keywords are: Points (points) Geometry shaders that operate on points are valid only for the POINTS primitive type. There is only a single vertex available for each geometry shader invocation. Lines (lines) Geometry shaders that operate on line segments are valid only for the LINES, LINE_STRIP, and LINE_LOOP primitive types. There are two vertices available for each geometry shader invocation. The first vertex refers to the vertex at the beginning of the line segment and the second vertex refers to the vertex at the end of the line segment. See also section 11.3.4. OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

329

Lines with Adjacency (lines_adjacency) Geometry shaders that operate on line segments with adjacent vertices are valid only for the LINES_ADJACENCY and LINE_STRIP_ADJACENCY primitive types. There are four vertices available for each program invocation. The second vertex refers to attributes of the vertex at the beginning of the line segment and the third vertex refers to the vertex at the end of the line segment. The first and fourth vertices refer to the vertices adjacent to the beginning and end of the line segment, respectively. Triangles (triangles) Geometry shaders that operate on triangles are valid for the TRIANGLES, TRIANGLE_STRIP and TRIANGLE_FAN primitive types. There are three vertices available for each program invocation. The first, second and third vertices refer to attributes of the first, second and third vertex of the triangle, respectively. Triangles with Adjacency (triangles_adjacency) Geometry shaders that operate on triangles with adjacent vertices are valid for the TRIANGLES_ADJACENCY and TRIANGLE_STRIP_ADJACENCY primitive types. There are six vertices available for each program invocation. The first, third and fifth vertices refer to attributes of the first, second and third vertex of the triangle, respectively. The second, fourth and sixth vertices refer to attributes of the vertices adjacent to the edges from the first to the second vertex, from the second to the third vertex, and from the third to the first vertex, respectively.

11.3.2

Geometry Shader Output Primitives

A geometry shader can generate primitives of one of three types. The supported output primitive types are points (POINTS), line strips (LINE_STRIP), and triangle strips (TRIANGLE_STRIP). The vertices output by the geometry shader are assembled into points, lines, or triangles based on the output primitive type in the manner described in section 10.5. The resulting primitives are then further processed as described in section 11.3.4. If the number of vertices emitted by the geometry shader is not sufficient to produce a single primitive, nothing is drawn. The number of vertices output by the geometry shader is limited to a maximum count specified in the shader. The output primitive type and maximum output vertex count are specified in the geometry shader source code using an output layout qualifier, as described in section 4.4.2.2 (“Geometry Outputs”) of the OpenGL ES Shading Language Specification. A program will fail to link if either the output primitive type or maximum output vertex count are not specified by the geometry shader object attached to the

OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

330

program. The output primitive type and maximum output vertex count of a linked program may be queried by calling GetProgramiv with the symbolic constants GEOMETRY_OUTPUT_TYPE and GEOMETRY_VERTICES_OUT, respectively.

11.3.3

Geometry Shader Variables

Geometry shaders can access uniforms belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. Geometry shaders also have access to samplers to perform texturing operations, as described in section 7.9. Geometry shaders can access the transformed attributes of all vertices for their input primitive type using input variables. A vertex or tessellation shader (the upstream shader for the geometry shader) writing to output variables generates the values of these input variables. Values for any inputs that are not written by a vertex shader are undefined. Additionally, a geometry shader has access to a builtin variable that holds the ID of the current primitive. This ID is generated by the primitive assembly stage that sits in between the vertex and geometry shader. Additionally, geometry shaders can write to one or more output variables for each vertex they output. These values are optionally flatshaded (using the OpenGL ES Shading Language qualifier flat) and clipped, then the clipped values interpolated across the primitive (if not flatshaded). The results of these interpolations are available to the fragment shader.

11.3.4

Geometry Shader Execution Environment

If there is an active program for the geometry stage, the executable version of the program’s geometry shader is used to process primitives resulting from the primitive assembly stage. There are several special considerations for geometry shader execution described in the following sections. 11.3.4.1

Texture Access

Section 11.1.3.1 describes texture lookup functionality accessible to a vertex shader. The texel fetch and texture size query functionality described there also applies to geometry shaders.

OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS 11.3.4.2

331

Instanced Geometry Shaders

For each input primitive received by the geometry shader pipeline stage, the geometry shader may be run once or multiple times. The number of times a geometry shader should be executed for each input primitive may be specified using a layout qualifier in a geometry shader of a linked program. If the invocation count is not specified in any layout qualifier, the invocation count will be one. Each separate geometry shader invocation is assigned a unique invocation number. For a geometry shader with N invocations, each input primitive spawns N invocations, numbered 0 through N − 1. The built-in input variable gl_InvocationID may be used by a geometry shader invocation to determine its invocation number. When executing instanced geometry shaders, the output primitives generated from each input primitive are passed to subsequent pipeline stages using the shader invocation number to order the output. The first primitives received by the subsequent pipeline stages are those emitted by the shader invocation numbered zero, followed by those from the shader invocation numbered one, and so forth. Additionally, all output primitives generated from a given input primitive are passed to subsequent pipeline stages before any output primitives generated from subsequent input primitives. 11.3.4.3

Geometry Shader Inputs

Section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification describes the built-in variable array gl_in[] available as input to a geometry shader. gl_in[] receives values from the equivalent built-in output variables written by the upstream shader, and each array element of gl_in[] is a structure holding values for a specific vertex of the input primitive. The length of gl_in[] is determined by the geometry shader input type (see section 11.3.1). The members of each element of the gl_in[] array are: • Structure member gl_Position holds the per-vertex position written by the upstream shader to its built-in output variable gl_Position. Note that writing to gl_Position from either the upstream or geometry shader is optional (also see section 7.1 (“Built-In Variables”) of the OpenGL ES Shading Language Specification). Geometry shaders also have available the built-in input variable gl_PrimitiveIDIn, which is not an array and has no vertex shader equivalent. It is filled with the number of primitives processed by the drawing command which generated the input vertices. The first primitive generated by a drawing command OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

332

is numbered zero, and the primitive ID counter is incremented after every individual point, line, or triangle primitive is processed. For triangles drawn in point or line mode, the primitive ID counter is incremented only once, even though multiple points or lines may eventually be drawn. The counter is reset to zero between each instance drawn. Restarting a primitive topology using the primitive restart index has no effect on the primitive ID counter. Similarly to the built-in inputs, each user-defined input has a value for each vertex and thus needs to be declared as arrays or inside input blocks declared as arrays. Declaring an array size is optional. If no size is specified, it will be inferred by the linker from the input primitive type. If a size is specified, it must match the number of vertices for the input primitive type; otherwise a link error will occur. Similarly to the limit on vertex shader output components (see section 11.1.2.1), there is a limit on the number of components of input variables that can be read by the geometry shader, given by the value of the implementationdependent constant MAX_GEOMETRY_INPUT_COMPONENTS. When a program is linked, all components of any input read by a geometry shader will count against this limit. A program whose geometry shader exceeds this limit may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS (see section 11.1.2.1). 11.3.4.4

Geometry Shader Outputs

A geometry shader is limited in the number of vertices it may emit per invocation. The maximum number of vertices a geometry shader can possibly emit is specified in the geometry shader source and may be queried after linking by calling GetProgramiv with pname GEOMETRY_VERTICES_OUT. If a single invocation of a geometry shader emits more vertices than this value, the emitted vertices may have no effect. There are two implementation-dependent limits on the value of GEOMETRY_VERTICES_OUT; it may not exceed the value of MAX_GEOMETRY_OUTPUT_VERTICES, and the product of the total number of vertices and the sum of all components of all active output variables may not exceed the value of MAX_GEOMETRY_TOTAL_OUTPUT_COMPONENTS. LinkProgram will fail if it determines that the total component limit would be violated. A geometry shader can write to built-in as well as user-defined output variables. These values are expected to be interpolated across the primitive it outputs, unless they are specified to be flat shaded. To enable seamlessly inserting or removing a geometry shader from a program object, the rules, names and types of the builtOpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

333

in and user-defined output variables are the same as for the vertex shader. Refer to section 11.1.2.1, and to sections 4.3.6 (“Output Variables”) and 7.1 (“Built-In Language Variables”) of the OpenGL ES Shading Language Specification for more detail. After a geometry shader emits a vertex, all output variables are undefined, as described in section 8.15 (“Geometry Shader Functions”) of the OpenGL ES Shading Language Specification. The built-in output gl_Position is intended to hold the homogeneous vertex position. Writing gl_Position is optional. The built-in output gl_PrimitiveID holds the primitive ID counter read by the fragment shader, replacing the value of gl_PrimitiveID generated by drawing commands when no geometry shader is active. The geometry shader must write to gl_PrimitiveID for the provoking vertex (see section 12.4) of a primitive being generated, or the primitive ID counter read by the fragment shader for that primitive is undefined. The built-in output gl_Layer is used in layered rendering, and discussed further in the next section. Similarly to the limit on vertex shader output components (see section 11.1.2.1), there is a limit on the number of components of output variables that can be written by the geometry shader, given by the value of the implementationdependent constant MAX_GEOMETRY_OUTPUT_COMPONENTS. When a program is linked, all components of any output variable written by a geometry shader will count against this limit. A program whose geometry shader exceeds this limit may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS (see section 11.1.2.1). 11.3.4.5

Layer Selection

Geometry shaders can be used to render to one of several different layers of cube map, three-dimensional, cube map array, or two-dimensional array textures. This functionality allows an application to bind an entire complex texture to a framebuffer object, and render primitives to arbitrary layers computed at run time. For example, it can be used to project and render a scene onto all six faces of a cubemap texture in one pass. The layer to render to is specified by writing to the built-in output variable gl_Layer. Layered rendering requires the use of framebuffer objects (see section 9.8). The specific vertex of a primitive that is used to select the rendering layer is implementation-dependent and thus portable applications will assign the same OpenGL ES 3.2 (May 5, 2022)

11.3. GEOMETRY SHADERS

334

layer for all vertices in a primitive. The vertex convention followed for gl_Layer may be determined by calling GetIntegerv with pname LAYER_PROVOKING_VERTEX. If the value returned is FIRST_VERTEX_CONVENTION, selection is always taken from the first vertex of a primitive. If the value returned is LAST_VERTEX_CONVENTION, the selection is always taken from the last vertex of a primitive. If the value returned is UNDEFINED_VERTEX, the selection is not guaranteed to be taken from any specific vertex in the primitive. The vertex considered the provoking vertex for particular primitive types is given in table 12.3. 11.3.4.6

Primitive Type Mismatches and Drawing Commands

Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL, and no fragments will be rendered, if a mismatch exists between the type of primitive being drawn and the input primitive type of a geometry shader. A mismatch exists under any of the following conditions: • the input primitive type of the current geometry shader is POINTS and mode is not POINTS, • the input primitive type of the current geometry shader is LINES and mode is not LINES, LINE_STRIP, or LINE_LOOP, • the input primitive type of the current geometry shader is TRIANGLES and mode is not TRIANGLES, TRIANGLE_STRIP or TRIANGLE_FAN, • the input primitive type of the current geometry shader is LINES_ADJACENCY and mode is not LINES_ADJACENCY or LINE_STRIP_ADJACENCY, or, • the input primitive type of the current geometry shader is TRIANGLES_ADJACENCY and mode is not TRIANGLES_ADJACENCY or TRIANGLE_STRIP_ADJACENCY.

OpenGL ES 3.2 (May 5, 2022)

Chapter 12

Fixed-Function Vertex Post-Processing After programmable vertex processing, the following fixed-function operations are applied to vertices of the resulting primitives: • Transform feedback (see section 12.2). • Primitive queries (see section 12.3). • Flatshading (see section 12.4). • Clipping (see section 12.5). • Shader output clipping (see section 12.5.1). • Perspective division on clip coordinates (see section 12.6). • Viewport mapping, including depth range scaling (see section 12.6.1). • Front face determination (see section 13.7.1). • Generic attribute clipping (see section 12.5.1). Next, rasterization is performed on primitives as described in chapter 13).

12.1

The Last Vertex Processing Stage

In the remainder of this chapter, the last vertex processing stage is the shader stage which passes outputs on to the operations in this chapter. The last vertex processing stage is: 335

12.2. TRANSFORM FEEDBACK

336

• The geometry shader stage, if a geometry shader is active; • Otherwise, the tessellation evaluation stage, if a tessellation evaluation shader is active; • Otherwise, the vertex shader stage, if a vertex shader is active.

12.2

Transform Feedback

In transform feedback mode, attributes of the vertices of transformed primitives passed to the transform feedback stage are written out to one or more buffer objects. The vertices are fed back before flatshading and clipping. The transformed vertices may be optionally discarded after being stored into one or more buffer objects, or they can be passed on down to the clipping stage for further processing. The set of attributes captured is determined when a program is linked. The data captured in transform feedback mode depends on the active programs on each of the shader stages. If a program is active for the geometry shader stage, transform feedback captures the vertices of each primitive emitted by the geometry shader. Otherwise, if a program is active for the tessellation evaluation shader stage, transform feedback captures each primitive produced by the tessellation primitive generator, whose vertices are processed by the tessellation evaluation shader. Otherwise, transform feedback captures each primitive processed by the vertex shader. Transform feedback captures the vertices of each primitive emitted by the last vertex processing stage. If separable program objects are in use, the set of attributes captured is taken from the program object active on the last vertex processing stage. The set of attributes to capture in transform feedback mode for any other program active on a previous shader stage is ignored.

12.2.1

Transform Feedback Objects

The set of buffer objects used to capture vertex output variables and related state are stored in a transform feedback object. The set of attributes captured in transform feedback mode is determined using the state of the active program object. The name space for transform feedback objects is the unsigned integers. The name zero designates the default transform feedback object. The command void GenTransformFeedbacks( sizei n, uint *ids );

OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK

337

returns n previously unused transform feedback object names in ids. These names are marked as used, for the purposes of GenTransformFeedbacks only, but they acquire transform feedback state only when they are first bound. Errors An INVALID_VALUE error is generated if n is negative. Transform feedback objects are deleted by calling void DeleteTransformFeedbacks( sizei n, const uint *ids ); ids contains n names of transform feedback objects to be deleted. After a transform feedback object is deleted it has no contents, and its name is again unused. Unused names in ids that have been marked as used for the purposes of GenTransformFeedbacks are marked as unused again. Unused names in ids are silently ignored, as is the value zero. The default transform feedback object cannot be deleted. Active transform feedback objects cannot be deleted. If an inactive transform feedback object that is bound to the context is deleted it is first unbound, as if BindTransformFeedback(TRANSFORM_FEEDBACK, 0); were called. In the initial state, a default transform feedback object is bound and treated as a transform feedback object with a name of zero. That object is bound any time BindTransformFeedback is called with id of zero. Errors An INVALID_VALUE error is generated if n is negative. An INVALID_OPERATION error is generated if the transform feedback operation for any object named by ids is currently active. The command boolean IsTransformFeedback( uint id ); returns TRUE if id is the name of a transform feedback object. If id is zero, or a non-zero value that is not the name of a transform feedback object, IsTransformFeedback returns FALSE. No error is generated if id is not a valid transform feedback object name. OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK A transform feedback object is created by binding a name returned by GenTransformFeedbacks with the command void BindTransformFeedback( enum target, uint id ); target must be TRANSFORM_FEEDBACK and id is the transform feedback object name. The resulting transform feedback object is a new state vector, comprising all the state and with the same initial values listed in table 21.35. Additionally, the new object is bound to the GL state vector and is used for subsequent transform feedback operations1 . BindTransformFeedback can also be used to bind an existing transform feedback object to the GL state for subsequent use. If the bind is successful, no change is made to the state of the newly bound transform feedback object and any previous binding to target is broken. While a transform feedback buffer is bound, GL operations on the target to which it is bound affect the bound transform feedback object, and queries of the target to which a transform feedback object is bound return state from the bound object. When buffer objects are bound for transform feedback, they are attached to the currently bound transform feedback object. Buffer objects are used for transform feedback only if they are attached to the currently bound transform feedback object. In the initial state, a default transform feedback object is bound and treated as a transform feedback object with a name of zero. That object is bound any time BindTransformFeedback is called with id of zero. Errors An INVALID_ENUM error is generated if target is not TRANSFORM_FEEDBACK. An INVALID_OPERATION error is generated if the transform feedback operation is active on the currently bound transform feedback object, and that operation is not paused (as described below). An INVALID_OPERATION error is generated if id is not zero or a name returned from a previous call to GenTransformFeedbacks, or if such a name has since been deleted with DeleteTransformFeedbacks. 1

The TRANSFORM_FEEDBACK_BUFFER_BINDING generic buffer binding point has now been moved to context state, rather than per-object state, and should be unaffected by this command. It is possible that some drivers do not yet implement this change.

OpenGL ES 3.2 (May 5, 2022)

338

12.2. TRANSFORM FEEDBACK

12.2.2

339

Transform Feedback Primitive Capture

Transform feedback for the currently bound transform feedback object is is started (made active) and finished (made inactive) with the commands void BeginTransformFeedback( enum primitiveMode ); and void EndTransformFeedback( void ); respectively. primitiveMode must be TRIANGLES, LINES, or POINTS, and specifies the output type of primitives that will be recorded into the buffer objects bound for transform feedback (see below). primitiveMode restricts the primitive types that may be rendered while transform feedback is active and not paused. EndTransformFeedback first performs an implicit ResumeTransformFeedback (see below) if transform feedback is paused. BeginTransformFeedback and EndTransformFeedback calls must be paired. Transform feedback is initially inactive. Transform feedback mode captures the values of output variables written by the last vertex processing stage. Errors An INVALID_ENUM error is generated by BeginTransformFeedback if primitiveMode is not TRIANGLES, LINES, or POINTS. An INVALID_OPERATION error is generated by BeginTransformFeedback if transform feedback is active for the current transform feedback object. An INVALID_OPERATION error is generated by EndTransformFeedback if transform feedback is inactive. Transform feedback operations for the currently bound transform feedback object may be paused and resumed by calling void PauseTransformFeedback( void ); and void ResumeTransformFeedback( void ); respectively. When transform feedback operations are paused, transform feedback is still considered active and changing most transform feedback state related to the OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK Transform Feedback primitiveMode

Allowed render primitive modes

POINTS LINES TRIANGLES

POINTS LINES, LINE_LOOP, LINE_STRIP TRIANGLES, TRIANGLE_STRIP, TRIANGLE_FAN

340

Table 12.1: Legal combinations of the transform feedback primitive mode, as passed to BeginTransformFeedback, and the current primitive mode.

object results in an error. However, a new transform feedback object may be bound while transform feedback is paused. When transform feedback is active and not paused, all geometric primitives generated must be compatible with the value of primitiveMode passed to BeginTransformFeedback. Errors An INVALID_OPERATION error is generated by PauseTransformFeedback if the currently bound transform feedback object is not active or is paused. An INVALID_OPERATION error is generated by ResumeTransformFeedback if the currently bound transform feedback object is not active or is not paused. An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if mode is not one of the allowed modes in table 12.1. If a tessellation evaluation or geometry shader is active, the type of primitive emitted by that shader is used instead of the mode parameter passed to drawing commands for the purposes of this error check. If tessellation evaluation and geometry shaders are both active, the output primitive type of the geometry shader will be used for the purposes of this error. Any primitive type may be used while transform feedback is paused. Regions of buffer objects are bound as the targets of transform feedback by calling one of the BindBuffer* commands (see section 6) with target set to TRANSFORM_FEEDBACK_BUFFER. When an individual point, line, or triangle primitive reaches the transform feedback stage while transform feedback is active and not paused, the values of the specified output variables of the vertex are appended to the buffer objects bound to the transform feedback binding points. The output variables of the first vertex OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK

341

received after BeginTransformFeedback are written at the starting offsets of the bound buffer objects set by BindBuffer*, and subsequent output variables are appended to the buffer object. When capturing line and triangle primitives, all output variables of the first vertex are written first, followed by output variables of the subsequent vertices. When writing output variables that are arrays, individual array elements are written in order. For multi-component output variables or elements of output arrays, the individual components are written in order. Variables declared with lowp or mediump precision are promoted to highp before being written. See Table 12.2 showing the output buffer type for each OpenGL ES Shading Language variable type. The value for any output variable specified to be streamed to a buffer object but not actually written by the last vertex processing stage is undefined. The results of appending an output variable to a transform feedback buffer are undefined if any component of that variable would be written at an offset not aligned to the size of the component. When transform feedback is paused, no vertices are recorded. When transform feedback is resumed, subsequent vertices are appended to the bound buffer objects immediately following the last vertex written before transform feedback was paused. Individual lines or triangles of a strip or fan primitive will be extracted and recorded separately. Incomplete primitives are not recorded. Transform feedback can operate in either INTERLEAVED_ATTRIBS or SEPARATE_ATTRIBS mode. In INTERLEAVED_ATTRIBS mode, the values of one or more output variables written by the last vertex processing stage are written, interleaved, into the buffer object bound to the first transform feedback binding point (index = 0). If more than one output variable is written to a buffer object, they will be recorded in the order specified by TransformFeedbackVaryings (see section 11.1.2.1). In SEPARATE_ATTRIBS mode, the first output variable specified by TransformFeedbackVaryings is written to the first transform feedback binding point; subsequent output variables are written to the subsequent transform feedback binding points. The total number of variables that may be captured in separate mode is given by MAX_TRANSFORM_FEEDBACK_SEPARATE_ATTRIBS. In either separate or interleaved modes, all transform feedback binding points that will be written to must have buffer objects bound when BeginTransformFeedback is called. Errors

OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK

342

Keyword

Output Type

float vec2 vec3 vec4 mat2 mat3 mat4 mat2x3 mat2x4 mat3x2 mat3x4 mat4x2 mat4x3 int ivec2 ivec3 ivec4 uint uvec2 uvec3 uvec4

float

int

uint

Table 12.2: OpenGL ES Shading Language keywords declaring each type and corresponding output buffer type.

OpenGL ES 3.2 (May 5, 2022)

12.2. TRANSFORM FEEDBACK

343

An INVALID_OPERATION error is generated by BeginTransformFeedback if any binding point used in transform feedback mode does not have a buffer object bound. In interleaved mode, only the first buffer object binding point is ever written to. An INVALID_OPERATION error is generated by BeginTransformFeedback if no binding points would be used, either because no program object is active or because the active program object has specified no output variables to record. When BeginTransformFeedback is called with an active program containing a vertex, tessellation or geometry shader, the set of output variables captured during transform feedback is taken from the active program object and may not be changed while transform feedback is active. The program object must be active until EndTransformFeedback is called, except while the transform feedback object is paused. Errors An INVALID_OPERATION error is generated : • by UseProgram if the current transform feedback object is active and not paused; • by UseProgramStages if the program pipeline object it refers to is current and the current transform feedback object is active and not paused; • by BindProgramPipeline if the current transform feedback object is active and not paused; • by LinkProgram or ProgramBinary if program is the name of a program being used by one or more transform feedback objects, even if the objects are not currently bound or are paused; • by ResumeTransformFeedback if the program object being used by the current transform feedback object is not active. • by ResumeTransformFeedback if the program pipeline object being used by the current transform feedback object is not bound, if any of its shader stage bindings has changed, or if a single program object is active and overriding it; and

OpenGL ES 3.2 (May 5, 2022)

12.3. PRIMITIVE QUERIES

344

• by BindBufferRange or BindBufferBase if target is TRANSFORM_FEEDBACK_BUFFER and transform feedback is currently active. Buffers should not be simultaneously used for both transform feedback and other purposes in the GL. Specifically, if a buffer object is bound to any of the binding points in an active and not paused transform feedback object, only writes to that binding point are supported, and any other writes to or reads from the buffer generate undefined values. This includes writes to another transform feedback binding point when the same buffer is bound to more than one binding point in the same transform feedback object. Other examples of reads or writes that could generate undefined values include ReadPixels to a pixel buffer object binding point and client access to a buffer mapped with MapBuffer. Commands that attempt to read or write to an active and unpaused transform feedback buffer will have undefined results. Generating an INVALID_OPERATION error is recommended in this case. However, if a buffer object is written and read sequentially by transform feedback and other mechanisms, it is the responsibility of the GL to ensure that data are accessed consistently, even if the implementation performs the operations in a pipelined manner. For example, MapBufferRange may need to block pending the completion of a previous transform feedback operation.

12.3

Primitive Queries

Primitive queries use query objects to track the number of primitives that are generated by the GL and the number of primitives that are written to buffer objects in transform feedback mode. When BeginQuery is called with a target of PRIMITIVES_GENERATED, the primitives generated count maintained by the GL is set to zero. When a generated primitive query is active, the primitives-generated count is incremented every time an emitted primitive reaches the transform feedback stage (see section 12.2), whether or not transform feedback is active. This counter counts the number of primitives emitted by a geometry shader, if active, possibly further tessellated into separate primitives during the transform feedback stage, if active. When BeginQuery is called with a target of TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, the transform feedback primitives written count maintained by the GL is set to zero. When the transform feedback primitive written query is active, the transform feedback primitives written count is incremented every time the vertices of a primitive are recorded into a buffer object. If transform feedback is not active or if a primitive to be recorded does not fit in a buffer object, this counter is not incremented. OpenGL ES 3.2 (May 5, 2022)

12.4. FLATSHADING

345

Type of primitive i point independent line line loop line strip independent triangle triangle strip triangle fan line adjacency line strip adjacency triangle adjacency triangle strip adjacency

Provoking vertex i 2i i + 1, if i < n 1, if i = n i+1 3i i+2 i+2 4i − 1 i+2 6i − 1 2i + 3

Table 12.3: Provoking vertex selection. The output values used for flatshading the ith primitive generated by drawing commands with the indicated primitive type are derived from the corresponding values of the vertex whose index is shown in the table. Vertices are numbered 1 through n, where n is the number of vertices drawn. These two types of queries can be used together to determine if all primitives in a given vertex stream have been written to the bound feedback buffers; if both queries are run simultaneously and the query results are equal, all primitives have been written to the buffer(s). If the number of primitives written is less than the number of primitives generated, one or more buffers overflowed.

12.4

Flatshading

Flatshading an output of the last vertex processing stage means to assign all vertices of the primitive the same value for that output. The output values assigned are those of the provoking vertex of the primitive, as shown in table 12.3. User-defined output variables may be flatshaded by using the flat qualifier when declaring the output, as described in section 4.3.6 (“Interpolation Qualifiers”) of the OpenGL ES Shading Language Specification.

OpenGL ES 3.2 (May 5, 2022)

12.5. PRIMITIVE CLIPPING

12.5

346

Primitive Clipping

Primitives are clipped to the clip volume. In clip coordinates, the clip volume is defined by −wc ≤ xc ≤ wc −wc ≤ yc ≤ wc −wc ≤ zc ≤ wc . If the primitive under consideration is a point, then clipping passes it unchanged if it lies within the near and far clip planes; otherwise, it is discarded. If the primitive is a line segment, then clipping does nothing to it if it lies entirely within the near and far clip planes, and discards it if it lies entirely outside these planes. If part of the line segment lies between the near and far clip planes and part lies outside, then the line segment is clipped and new vertex coordinates are computed for one or both vertices. A clipped line segment endpoint lies on both the original line segment and the near and/or far clip planes. This clipping produces a value, 0 ≤ t ≤ 1, for each clipped vertex. If the coordinates of a clipped vertex are P and the original vertices’ coordinates are P1 and P2 , then t is given by P = tP1 + (1 − t)P2 . The value of t is used to clip outputs of the last vertex processing stage as described in section 12.5.1. If the primitive is a polygon, then it is passed if every one of its edges lies entirely inside the clip volume and either clipped or discarded otherwise. Polygon clipping may cause polygon edges to be clipped, but because polygon connectivity must be maintained, these clipped edges are connected by new edges that lie along the clip volume’s boundary. Thus, clipping may require the introduction of new vertices into a polygon. If it happens that a polygon intersects an edge of the clip volume’s boundary, then the clipped polygon must include a point on this boundary edge.

12.5.1

Clipping Shader Outputs

Next, outputs of the last vertex processing stage are clipped. The output values associated with a vertex that lies within the clip volume are unaffected by clipping. If a primitive is clipped, however, the output values assigned to vertices produced by clipping are clipped.

OpenGL ES 3.2 (May 5, 2022)

12.6. COORDINATE TRANSFORMATIONS

347

Let the output values assigned to the two vertices P1 and P2 of an unclipped edge be c1 and c2 . The value of t (section 12.5) for a clipped point P is used to obtain the output value associated with P as2 c = tc1 + (1 − t)c2 . (Multiplying an output value by a scalar means multiplying each of x, y, z, and w by the scalar.) Polygon clipping may create a clipped vertex along an edge of the clip volume’s boundary. This situation is handled by noting that polygon clipping proceeds by clipping against one half-space at a time. Output value clipping is done in the same way, so that clipped points always occur at the intersection of polygon edges (possibly already clipped) with the clip volume’s boundary. Outputs of integer or unsigned integer type must always be declared with the flat qualifier. Since such outputs are constant over the primitive being rasterized (see sections 13.6.1 and 13.7.1), no interpolation is performed.

12.6

Coordinate Transformations

Clip coordinates for a vertex result from shader execution, which yields a vertex coordinate gl_Position. Perspective division on clip coordinates yields normalized device coordinates, followed by a viewport transformation (see section 12.6.1) to convert these coordinates into window coordinates.   xc  yc   If a vertex in clip coordinates is given by   zc  wc then the vertex’s normalized device coordinates are    xc  xd wc  yd  =  yc  . wc zc zd wc

12.6.1

Controlling the Viewport

The viewport transformation is determined by the viewport’s width and height in pixels, px and py , respectively, and its center (ox , oy ) (also in pixels). The vertex’s 2

Since this computation is performed in clip space before division by wc , clipped output values are perspective-correct.

OpenGL ES 3.2 (May 5, 2022)

12.6. COORDINATE TRANSFORMATIONS

348

 xw window coordinates,  yw  , are given by zw 

  px  xw 2 xd + ox p  yw  =  y yd + oy  . 2 f −n n+f zw 2 zd + 2 

The factor and offset applied to zd encoded by n and f are set using void DepthRangef( float n, float f ); zw may be represented using either a fixed-point or floating-point representation. However, a floating-point representation must be used if the draw framebuffer has a floating-point depth buffer. If an m-bit fixed-point representation is used, we assume that it represents each value 2mk−1 , where k ∈ {0, 1, . . . , 2m − 1}, as k (e.g. 1.0 is represented in binary as a string of all ones). The parameters n and f are clamped to the range [0, 1] when specified. Viewport transformation parameters are specified using void Viewport( int x, int y, sizei w, sizei h ); where x and y give the x and y window coordinates of the viewport’s lower left corner and w and h give the viewport’s width and height, respectively. The viewport parameters shown in the above equations are found from these values as ox = x + w2 oy = y + h2 px = w py = h. Viewport width and height are clamped to implementation-dependent maximums when specified. The maximum width and height may be found by calling GetFloatv with the symbolic constant MAX_VIEWPORT_DIMS. The maximum viewport dimensions must be greater than or equal to the larger of the visible dimensions of the display being rendered to (if a display exists), and the largest renderbuffer image which can be successfully created and attached to a framebuffer object (see chapter 9). Errors

OpenGL ES 3.2 (May 5, 2022)

12.6. COORDINATE TRANSFORMATIONS

349

An INVALID_VALUE error is generated if either w or h is negative. The state required to implement the viewport transformation is four integers and two clamped floating-point values. In the initial state, w and h are set to the width and height, respectively, of the window into which the GL is to do its rendering. If the default framebuffer is bound but no default framebuffer is associated with the GL context (see chapter 9), then w and h are initially set to zero. ox , oy , n, and f are set to w2 , h2 , 0.0, and 1.0, respectively.

OpenGL ES 3.2 (May 5, 2022)

Chapter 13

Fixed-Function Primitive Assembly and Rasterization Rasterization is the process by which a primitive is converted to a two-dimensional image. Each point of this image contains such information as color and depth. Rasterizing a primitive begins by determining which squares of an integer grid in window coordinates are occupied by the primitive, and assigning a depth value to each such square. This process is described in sections 13.1-13.7 for point, line, and triangle primitives. A grid square, including its (x, y) window coordinates, z (depth), and associated data which may be added by fragment shaders, is called a fragment. A fragment is located by its lower left corner, which lies on integer grid coordinates. Rasterization operations also refer to a fragment’s center, which is offset by ( 21 , 21 ) from its lower left corner (and so lies on half-integer coordinates). Fragments need not actually be square, and rasterization rules are not affected by the aspect ratio of fragments. Display of non-square grids, however, will cause rasterized points and line segments to appear fatter in one direction than the other. We assume that fragments are square, since it simplifies antialiasing and texturing. After rasterization, fragments are processed by the early per-fragment tests described in section 13.8, which may modify or discard fragments. Surviving fragments are passed to fragment shaders (see chapter 14). Fragment shaders determine color values for fragments, and may also modify or replace their assigned depth values. Figure 13.1 diagrams the rasterization process. Several factors affect rasterization. Primitives may be discarded before rasterization. Points may be given differing diameters and line segments differing widths. Rasterization only produces fragments corresponding to pixels in the frame-

350

351

From Primitive Assembly

Point Rasterization

Line Rasterization

Triangle Rasterization

Pixel Ownership Test

Scissor Test

Multisample Fragment Operations

Fragment Shading

Other Early Tests

Processed Fragments

Figure 13.1. Rasterization, early per-fragment tests, and fragment shading. Optional early tests described in section 13.8 are included in the “Other Early Tests” box.

OpenGL ES 3.2 (May 5, 2022)

13.1. DISCARDING PRIMITIVES BEFORE RASTERIZATION

352

buffer. Fragments which would be produced by application of any of the primitive rasterization rules described below but which lie outside the framebuffer are not produced, nor are they processed by any later stage of the GL, including any of the early per-fragment tests described in section 13.8.

13.1

Discarding Primitives Before Rasterization

Primitives can be optionally discarded before rasterization by calling Enable and Disable with RASTERIZER_DISCARD. When enabled, primitives are discarded immediately before the rasterization stage, but after the optional transform feedback stage (see section 12.2). When disabled, primitives are passed through to the rasterization stage to be processed normally. When enabled, RASTERIZER_DISCARD also causes the Clear and ClearBuffer* commands to be ignored. The state required to control primitive discard is a bit indicating whether discard is enabled or disabled. The initial value of primitive discard is FALSE.

13.2

Primitive Bounding Box

Implementations may be able to optimize performance if the application provides bounds of primitives that will be generated by the tessellation primitive generator or the geometry shader prior to executing those stages. If the provided bounds are incorrect and primitives extend beyond them, the rasterizer may or may not generate fragments for the portions of primitives outside the bounds. The primitive bounding box is specified with the command void PrimitiveBoundingBox( float minX, float minY, float minZ, float minW, float maxX, float maxY, float maxZ, float maxW ); where minX, minY, minZ, and minW specify the minimum clip space coordinate of the bounding box and maxX, maxY, maxZ, and maxW specify the maximum coordinate. If tessellation is active, each invocation of the tessellation control shader may re-specify the bounding box by writing to the built-in variable gl_BoundingBox. If the shader statically assigns a value to any part of this variable, then gl_BoundingBox[0] is used instead of minX, minY, minZ, minW, and gl_BoundingBox[1] is used instead of maxX, maxY, maxZ, maxW. If the shader contains a static assignment to gl_BoundingBox and there is an execution path through the shader that does not write all components of gl_BoundingBox, the OpenGL ES 3.2 (May 5, 2022)

13.3. INVARIANCE

353

value of unwritten components and corresponding bounding box coordinates is undefined for executions of the shader that take that path. If the tessellation control shader re-specifies the bounding box, the re-specified value is used for primitives generated from the output patch by the primitive generator, any primitives emitted by the geometry shader invocations for those generated primitives, and any primitives further introduced during clipping. The bounding box in clip space is composed of 16 vertices formed by all combinations of the minimum and maximum values for each dimension. This bounding box is clipped against wc > 0, and projected to three dimensions by dividing xc , yc , and zc by wc for each vertex. The viewport transform is then applied to each vertex to produce a three-dimensional bounding volume in window coordinates. The window space bounding volume is expanded in the X and Y dimensions to accomodate the rasterization rules for the primitive type, and to fall on fragment boundaries:   size 0 minwc = minwc − 2.0   size 0 . maxwc = maxwc + 2.0 where the minwc rule is used for x and y window coordinates of bounding volume vertices formed from minX and min respectively, and the maxwc rule is used for x and y window coordinates of bounding volume vertices formed from maxX and maxY respectively. For point primitives, size is the per-primitive point size after clamping to the implementation-defined maximum point size as described in section 13.5. For line primitives, size is the line width, after rounding and clamping as described in section 13.6.2.1. For triangle primitives, size is zero. During rasterization, the rasterizer will generate fragments with window coordinates inside the windows space bounding volume, but may or may not generate fragments with window coordinates outside the bounding volume.

13.3

Invariance

Consider a primitive p0 obtained by translating a primitive p through an offset (x, y) in window coordinates, where x and y are integers. As long as neither p0 nor p is clipped, it must be the case that each fragment f 0 produced from p0 is identical to a corresponding fragment f from p except that the center of f 0 is offset by (x, y) from the center of f .

OpenGL ES 3.2 (May 5, 2022)

13.4. MULTISAMPLING

13.4

354

Multisampling

Multisampling is a mechanism to antialias all GL primitives: points, lines, and polygons. The technique is to sample all primitives multiple times at each pixel. The color sample values are resolved to a single, displayable color. For window system-provided framebuffers, this occurs each time a pixel is updated, so the antialiasing appears to be automatic at the application level. For application-created framebuffers, this must be requested by calling the BlitFramebuffer command (see section 16.2). Because each sample includes color, depth, and stencil information, the color (including texture operation), depth, and stencil functions perform equivalently to the single-sample mode. An additional buffer, called the multisample buffer, is added to the window system-provided framebuffer. Pixel sample values, including color, depth, and stencil values, are stored in this buffer. Samples contain separate color values for each fragment color. When the window system-provided framebuffer includes a multisample buffer, it does not include depth or stencil buffers, even if the multisample buffer does not store depth or stencil values. Color buffers do coexist with the multisample buffer, however. Multisample antialiasing is most valuable for rendering polygons, because it requires no sorting for hidden surface elimination, and it correctly handles adjacent polygons, object silhouettes, and even intersecting polygons. If the value of SAMPLE_BUFFERS (see section 9.2.3.1) is one, the rasterization of all primitives is changed, and is referred to as multisample rasterization. Otherwise, primitive rasterization is referred to as single-sample rasterization. During multisample rendering the contents of a pixel fragment are changed in two ways. First, each fragment includes a coverage value with SAMPLES bits (see section 9.2.3.1). The location at which shading is performed for a given sample (the shading sample location) is queried with the command void GetMultisamplefv( enum pname, uint index, float *val ); pname must be SAMPLE_POSITION, and index corresponds to the sample for which the location should be returned. The shading sample location (x, y) is returned as two floating-point values in (val[0], val[1]) respectively. x and y each lie in the range [0, 1] and represent a location in pixel space at which depth and associated data for that sample are evaulated for a fragment (e.g. where sample shading is performed). (0.5, 0.5) thus corresponds to the pixel center. If the multisample mode does not have fixed shading sample locations, the returned values may only reflect the locations of samples within some pixels. OpenGL ES 3.2 (May 5, 2022)

13.4. MULTISAMPLING

355

Errors An INVALID_ENUM error is generated if pname is not SAMPLE_POSITION. An INVALID_VALUE error is generated if index is greater than or equal to the value of SAMPLES. Second, each fragment includes SAMPLES depth values and sets of associated data, instead of the single depth value and set of associated data that is maintained in single-sample rendering mode. An implementation may choose to assign the same associated data to more than one sample. The location for evaluating such associated data can be anywhere within the pixel including the fragment center or any of the sample locations. The different associated data values need not all be evaluated at the same location. Each pixel fragment thus consists of integer x and y grid coordinates, SAMPLES depth values and sets of associated data, and a coverage value with a maximum of SAMPLES bits. Multisample rasterization is only in effect when the value of SAMPLE_BUFFERS is one. Multisample rasterization of all primitives differs substantially from singlesample rasterization. It is understood that each pixel in the framebuffer has sample locations associated with it. These locations are exact positions, rather than regions or areas, and each is referred to as a sample point. These sample points do not necessarily correspond to the shading sample locations returned by GetMultisamplefv. Their locations cannot be queried, and may lie inside or outside of the unit square that is considered to bound the pixel. The number of these samples may be different than the value of SAMPLES. Furthermore, the relative locations of sample points may be identical for each pixel in the framebuffer, or they may differ. If the value of SAMPLE_BUFFERS is one and the current program object includes a fragment shader with one or more input variables qualified with sample in, the data associated with those variables will be assigned independently. The values for each sample must be evaluated at the location of the sample. The data associated with any other variables not qualified with sample in need not be evaluated independently for each sample. If the sample locations differ per pixel, they should be aligned to window, not screen, boundaries. Otherwise rendering results will be window-position specific. The invariance requirement described in section 13.3 is relaxed for all multisample rasterization, because the sample locations may be a function of pixel location.

OpenGL ES 3.2 (May 5, 2022)

13.5. POINTS

13.4.1

356

Sample Shading

Sample shading can be used to specify a minimum number of unique samples to process for each fragment. Sample shading is controlled by calling Enable or Disable with target SAMPLE_SHADING. If the value of SAMPLE_BUFFERS is zero or SAMPLE_SHADING is disabled, sample shading has no effect. Otherwise, an implementation must provide a minimum of max(dmss × samplese, 1) unique sets of fragment shader inputs for each fragment, where mss is the value of MIN_SAMPLE_SHADING_VALUE and samples is the number of samples (the values of SAMPLES). These are associated with the samples in an implementationdependent manner. The value of MIN_SAMPLE_SHADING_VALUE is specified by calling void MinSampleShading( float value ); with value set to the desired minimum sample shading fraction. value is clamped to [0, 1] when specified. The sample shading fraction may be queried by calling GetFloatv with pname MIN_SAMPLE_SHADING_VALUE. When the sample shading fraction is 1.0, a separate set of fragment shader input values are evaluated for each sample, and each set of values is evaluated at the sample location.

13.5

Points

A point is drawn by generating a set of fragments in the shape of a square centered around the vertex of the point. Each vertex has an associated point size that controls the size of that square or circle. The point size is determined by the last vertex processing stage. If the last vertex processing stage. is not a vertex shader, the point size is 1.0. If the last vertex processing stage is a vertex shader, the point size is taken from the shader built-in gl_PointSize written by the vertex shader, and is clamped to the implementation-dependent point size range. If the value written to gl_PointSize is less than or equal to zero, or if no value is written to gl_PointSize, the point size is undefined. The supported [min, max] range of point sizes may be queried as ALIASED_POINT_SIZE_RANGE, as described in table 21.40. The maximum point size supported must be at least one.

OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

13.5.1

357

Basic Point Rasterization

Point rasterization produces a fragment for each framebuffer pixel whose center lies inside a square centered at the point’s (xw , yw ), with side length equal to the current point size. All fragments produced in rasterizing a point sprite are assigned the same associated data, which are those of the vertex corresponding to the point. However, the fragment shader built-in gl_PointCoord defines a per-fragment coordinate space (s, t) where s varies from 0 to 1 across the point horizontally left-to-right, and t varies from 0 to 1 across the point vertically top-to-bottom. The following formula is used to evaluate (s, t) values:
xf + 21 − xw 1 s= + (13.1) 2 size
yf + 21 − yw 1 t= − (13.2) 2 size where size is the point’s size, xf and yf are the (integral) window coordinates of the fragment, and xw and yw are the exact, unrounded window coordinates of the vertex for the point.

13.5.2

Point Multisample Rasterization

If the value of SAMPLE_BUFFERS is one, then points are rasterized using the following algorithm. Point rasterization produces a fragment for each framebuffer pixel with one or more sample points that intersect a region centered at the point’s (xw , yw ). This region is a square with sides equal to the current point size. Coverage bits that correspond to sample points that intersect the region are 1, other coverage bits are 0. All data associated with each sample for the fragment are the data associated with the point being rasterized.

13.6

Line Segments

A line segment results from a line strip, a line loop, or a series of separate line segments. Line segment rasterization is controlled by several variables. Line width, which may be set by calling void LineWidth( float width ); with an appropriate positive floating-point width, controls the width of rasterized line segments. The default width is 1.0. OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

358

The supported [min, max] range of line widths may be queried as ALIASED_LINE_WIDTH_RANGE, as described in table 21.40. The maximum line width supported must be at least one. Errors An INVALID_VALUE error is generated if width is less than or equal to zero.

13.6.1

Basic Line Segment Rasterization

Line segment rasterization begins by characterizing the segment as either x-major or y-major. x-major line segments have slope in the closed interval [−1, 1]; all other line segments are y-major (slope is determined by the segment’s endpoints). We shall specify rasterization only for x-major segments except in cases where the modifications for y-major segments are not self-evident. Ideally, the GL uses a “diamond-exit” rule to determine those fragments that are produced by rasterizing a line segment. For each fragment f with center at window coordinates xf and yf , define a diamond-shaped region that is the intersection of four half planes: 1 Rf = { (x, y) | |x − xf | + |y − yf | < .} 2 Essentially, a line segment starting at pa and ending at pb produces those fragments f for which the segment intersects Rf , except if pb is contained in Rf . See figure 13.2. To avoid difficulties when an endpoint lies on a boundary of Rf we (in principle) perturb the supplied endpoints by a tiny amount. Let pa and pb have window coordinates (xa , ya ) and (xb , yb ), respectively. Obtain the perturbed endpoints p0a given by (xa , ya ) − (, 2 ) and p0b given by (xb , yb ) − (, 2 ). Rasterizing the line segment starting at pa and ending at pb produces those fragments f for which the segment starting at p0a and ending on p0b intersects Rf , except if p0b is contained in Rf .  is chosen to be so small that rasterizing the line segment produces the same fragments when δ is substituted for  for any 0 < δ ≤ . When pa and pb lie on fragment centers, this characterization of fragments reduces to Bresenham’s algorithm with one modification: lines produced in this description are “half-open,” meaning that the final fragment (corresponding to pb ) is not drawn. This means that when rasterizing a series of connected line segments, shared endpoints will be produced only once rather than twice (as would occur with Bresenham’s algorithm). OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

359

















































































































































































































































































































































































































































































































































































































































































































































Figure 13.2. Visualization of Bresenham’s algorithm. A portion of a line segment is shown. A diamond shaped region of height 1 is placed around each fragment center; those regions that the line segment exits cause rasterization to produce corresponding fragments.

Because the initial and final conditions of the diamond-exit rule may be difficult to implement, other line segment rasterization algorithms are allowed, subject to the following rules: 1. The coordinates of a fragment produced by the algorithm may not deviate by more than one unit in either x or y window coordinates from a corresponding fragment produced by the diamond-exit rule. 2. The total number of fragments produced by the algorithm may differ from that produced by the diamond-exit rule by no more than one. 3. For an x-major line, no two fragments may be produced that lie in the same window-coordinate column (for a y-major line, no two fragments may appear in the same row). 4. If two line segments share a common endpoint, and both segments are either x-major (both left-to-right or both right-to-left) or y-major (both bottom-totop or both top-to-bottom), then rasterizing both segments may not produce duplicate fragments, nor may any fragments be omitted so as to interrupt continuity of the connected segments. OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

360

Next we must specify how the data associated with each rasterized fragment are obtained. Let the window coordinates of a produced fragment center be given by pr = (xd , yd ) and let pa = (xa , ya ) and pb = (xb , yb ). Set t=

(pr − pa ) · (pb − pa ) . kpb − pa k2

(13.3)

(Note that t = 0 at pa and t = 1 at pb .) The value of an associated datum f for the fragment, whether it be a shader output or the clip w coordinate, is found as f=

(1 − t)fa /wa + tfb /wb (1 − t)/wa + t/wb

(13.4)

where fa and fb are the data associated with the starting and ending endpoints of the segment, respectively; wa and wb are the clip w coordinates of the starting and ending endpoints of the segments, respectively. However, depth values for lines must be interpolated by z = (1 − t)za + tzb

(13.5)

where za and zb are the depth values of the starting and ending endpoints of the segment, respectively. Although the interpolation formula given above is preferred, the formula given in equation 13.3 may be approximated by replacing (pr , pa , pb ) with (xr , xa , xb ) for x-major lines, or with (yr , ya , yb ) for y-major lines, respectively. The flat keyword used to declare shader outputs affects how they are interpolated. When it is not specified, interpolation is performed as described in equation 13.4. When the flat keyword is specified, no interpolation is performed, and outputs are taken from the corresponding input value of the provoking vertex corresponding to that primitive (see section 12.4).

13.6.2

Other Line Segment Features

We have just described the rasterization of line segments of width one. We now describe the rasterization of line segments for general values of line width. 13.6.2.1

Wide Lines

The actual width of lines is determined by rounding the supplied width to the nearest integer, then clamping it to the implementation-dependent maximum line width. This implementation-dependent value must be no less than 1. If rounding the specified width results in the value 0, then it is as if the value were 1.

OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

width = 2

361

width = 3

Figure 13.3. Rasterization of wide lines. x-major line segments are shown. The heavy line segment is the one specified to be rasterized; the light segment is the offset segment used for rasterization. x marks indicate the fragment centers produced by rasterization.

Line segments of width other than one are rasterized by offsetting them in the minor direction (for an x-major line, the minor direction is y, and for a y-major line, the minor direction is x) and producing a row or column of fragments in the minor direction (see figure 13.3). Let w be the width rounded to the nearest integer (if w = 0, then it is as if w = 1). If the line segment has endpoints given by (x0 , y0 ) and (x1 , y1 ) in window coordinates, the segment with endpoints (x0 , y0 − (w − 1)/2) and (x1 , y1 − (w − 1)/2) is rasterized, but instead of a single fragment, a column of fragments of height w (a row of fragments of length w for a y-major segment) is produced at each x (y for y-major) location. The lowest fragment of this column is the fragment that would be produced by rasterizing the segment of width 1 with the modified coordinates. The preferred method of attribute interpolation for a wide line is to generate the same attribute values for all fragments in the row or column described above, as if the adjusted line were used for interpolation and those values replicated to the other fragments, except for gl_FragCoord which is interpolated as usual. An implementation may instead interpolate each fragment according to the formula in ¡¡primsrast-lines-basic,Basic Line Segment Rasterization¿¿, using the original line segment endpoints.

OpenGL ES 3.2 (May 5, 2022)

13.6. LINE SEGMENTS

362

Figure 13.4. The region used in rasterizing a multisampled line segment (an x-major line segment is shown).

13.6.3

Line Rasterization State

The state required for line rasterization consists of the floating-point line width. The initial value of the line width is 1.0.

13.6.4

Line Multisample Rasterization

If the value of SAMPLE_BUFFERS is one, then lines are rasterized using the following algorithm. Line rasterization produces a fragment for each framebuffer pixel with one or more sample points that intersect a rectangle centered on the line segment (see figure 13.4). Two of the edges are parallel to the specified line segment; each is at a distance of one-half the line width from that segment: one above the segment and one below it. The other two edges pass through the line endpoints and are perpendicular to the direction of the specified line segment. Coverage bits that correspond to sample points that intersect a rectangle are 1, other coverage bits are 0. Each depth value and set of associated data is produced by substituting the corresponding sample location into equation 13.3, then using the result to evaluate equation 13.4. Note that the approximate form of equation 13.3 described in section 13.6.1 may not be used during multisampled line rasterization. An implementation may choose to assign the associated data to more than one sample by evaluating equation 13.3 at any location within the pixel including the fragment center or any one of the sample locations, then substituting

OpenGL ES 3.2 (May 5, 2022)

13.7. POLYGONS

363

into equation 13.4. The different associated data values need not be evaluated at the same location. The supported [min, max] range of multisampled line widths, and the width of evenly-spaced gradations within that range are implementation-dependent and may be queried as MULTISAMPLE_LINE_WIDTH_RANGE and MULTISAMPLE_LINE_WIDTH_GRANULARITY respectively, as described in table 21.40. If, for instance, the width range is from 0.1 to 2.0 and the gradation width is 0.1, then the widths 0.1, 0.2, . . . , 1.9, 2.0 are supported. Additional line widths may also be supported; there is no requirement that these widths be evenly spaced. If an unsupported width is requested, the nearest supported width is used instead. Width 1.0 segments must be supported.

13.7

Polygons

A polygon results from a triangle arising from a triangle strip, triangle fan, or series of separate triangles.

13.7.1

Basic Polygon Rasterization

The first step of polygon rasterization is to determine if the polygon is back-facing or front-facing. This determination is made based on the sign of the (clipped or unclipped) polygon’s area computed in window coordinates. One way to compute this area is n−1

a=

1 X i i⊕1 i⊕1 i x w yw − x w yw 2

(13.6)

i=0

i are the x and y window coordinates of the ith vertex of the where xiw and yw

n-vertex polygon (vertices are numbered starting at zero for purposes of this computation) and i ⊕ 1 is (i + 1) mod n. The interpretation of the sign of this value is controlled with void FrontFace( enum dir ); Setting dir to CCW (corresponding to counter-clockwise orientation of the projected polygon in window coordinates) uses a as computed above. Setting dir to CW (corresponding to clockwise orientation) indicates that the sign of a should be reversed prior to use. Front face determination requires one bit of state, and is initially set to CCW.

OpenGL ES 3.2 (May 5, 2022)

13.7. POLYGONS

364

Errors An INVALID_ENUM error is generated if dir is not CW or CCW. If the sign of a (including the possible reversal of this sign as determined by FrontFace) is positive, the polygon is front-facing; otherwise, it is back-facing. This determination is used in conjunction with the CullFace enable bit and mode value to decide whether or not a particular polygon is rasterized. The CullFace mode is set by calling void CullFace( enum mode ); mode is a symbolic constant: one of FRONT, BACK or FRONT_AND_BACK. Culling is enabled or disabled with Enable or Disable using the symbolic constant CULL_FACE. Front-facing polygons are rasterized if either culling is disabled or the CullFace mode is BACK while back-facing polygons are rasterized only if either culling is disabled or the CullFace mode is FRONT. The initial setting of the CullFace mode is BACK. Initially, culling is disabled. Errors An INVALID_ENUM error is generated if mode is not FRONT, BACK, or FRONT_AND_BACK. The rule for determining which fragments are produced by polygon rasterization is called point sampling. The two-dimensional projection obtained by taking the x and y window coordinates of the polygon’s vertices is formed. Fragment centers that lie inside of this polygon are produced by rasterization. Special treatment is given to a fragment whose center lies on a polygon edge. In such a case we require that if two polygons lie on either side of a common edge (with identical endpoints) on which a fragment center lies, then exactly one of the polygons results in the production of the fragment during rasterization. As for the data associated with each fragment produced by rasterizing a polygon, we begin by specifying how these values are produced for fragments in a triangle. Define barycentric coordinates for a triangle. Barycentric coordinates are a set of three numbers, a, b, and c, each in the range [0, 1], with a + b + c = 1. These coordinates uniquely specify any point p within the triangle or on the triangle’s boundary as p = apa + bpb + cpc , where pa , pb , and pc are the vertices of the triangle. a, b, and c can be found as a=

A(ppb pc ) , A(pa pb pc )

b=

A(ppa pc ) , A(pa pb pc )

c=

OpenGL ES 3.2 (May 5, 2022)

A(ppa pb ) , A(pa pb pc )

13.7. POLYGONS

365

where A(lmn) denotes the area in window coordinates of the triangle with vertices l, m, and n. Denote an associated datum at pa , pb , or pc as fa , fb , or fc , respectively. Then the value f of a datum at a fragment produced by rasterizing a triangle is given by f=

afa /wa + bfb /wb + cfc /wc a/wa + b/wb + c/wc

(13.7)

where wa , wb and wc are the clip w coordinates of pa , pb , and pc , respectively. a, b, and c are the barycentric coordinates of the fragment for which the data are produced. a, b, and c must correspond precisely to the exact coordinates of the center of the fragment. Another way of saying this is that the data associated with a fragment must be sampled at the fragment’s center. However, depth values for polygons must be interpolated by z = aza + bzb + czc

(13.8)

where za , zb , and zc are the depth values of pa , pb , and pc , respectively. When outputs are declared with the flat keyword, the method used to compute associated data depends on the type of the output. For floating point outputs, equation 13.7 is used, but fa , fb , and fc are replaced with fp , the value of the datum at the provoking vertex. The equation collapses to f = fp , but note that since interpolation is still performed, precision may be lost. For a polygon with more than three edges, such as may be produced by clipping a triangle, we require only that a convex combination of the values of the datum at the polygon’s vertices can be used to obtain the value assigned to each fragment produced by the rasterization algorithm. That is, it must be the case that at every fragment n X f= ai fi i=1

where n is the number of vertices P in the polygon, fi is the value of the f at vertex i; for each i 0 ≤ ai ≤ 1 and ni=1 ai = 1. The values of the ai may differ from fragment to fragment, but at vertex i, aj = 0, j 6= i and ai = 1. One algorithm that achieves the required behavior is to triangulate a polygon (without adding any vertices) and then treat each triangle individually as already discussed. A scan-line rasterizer that linearly interpolates data along each edge and then linearly interpolates data across each horizontal span from edge to edge also satisfies the restrictions (in this case, the numerator and denominator of equation 13.7 should be iterated independently and a division performed for each fragment).

OpenGL ES 3.2 (May 5, 2022)

13.7. POLYGONS

13.7.2

366

Depth Offset

The depth values of all fragments generated by the rasterization of a polygon may be offset by a single value that is computed for that polygon. The function that determines this value is specified by calling void PolygonOffset( float factor, float units ); factor scales the maximum depth slope of the polygon, and units scales an implementation-dependent constant that relates to the usable resolution of the depth buffer. The resulting values are summed to produce the polygon offset value. Both factor and units may be either positive or negative. The maximum depth slope m of a triangle is s    ∂zw 2 ∂zw 2 m= + (13.9) ∂xw ∂yw where (xw , yw , zw ) is a point on the triangle. m may be approximated as   ∂zw ∂zw m = max , . (13.10) ∂xw ∂yw The minimum resolvable difference r is an implementation-dependent parameter that depends on the depth buffer representation. It is the smallest difference in window coordinate z values that is guaranteed to remain distinct throughout polygon rasterization and in the depth buffer. All pairs of fragments generated by the rasterization of two polygons with otherwise identical vertices, but zw values that differ by r, will have distinct depth values. For fixed-point depth buffer representations, r is constant throughout the range of the entire depth buffer. For floating-point depth buffers, there is no single minimum resolvable difference. In this case, the minimum resolvable difference for a given polygon is dependent on the maximum exponent, e, in the range of z values spanned by the primitive. If n is the number of bits in the floating-point mantissa, the minimum resolvable difference, r, for the given primitive is defined as r = 2e−n . If no depth buffer is present, r is undefined. The offset value o for a polygon is o = m × f actor + r × units.

OpenGL ES 3.2 (May 5, 2022)

(13.11)

13.8. EARLY PER-FRAGMENT TESTS

367

m is computed as described above. If the depth buffer uses a fixed-point representation, m is a function of depth values in the range [0, 1], and o is applied to depth values in the same range. Boolean state value POLYGON_OFFSET_FILL determines whether o is applied during the rasterization of polygons. This boolean state value is enabled and disabled with the commands Enable and Disable. For fixed-point depth buffers, fragment depth values are always limited to the range [0, 1] by clamping after offset addition is performed. Fragment depth values are clamped even when the depth buffer uses a floating-point representation.

13.7.3

Polygon Multisample Rasterization

If the value of SAMPLE_BUFFERS is one, then polygons are rasterized using the following algorithm. Polygon rasterization produces a fragment for each framebuffer pixel with one or more sample points that satisfy the point sampling criteria described in section 13.7.1. If a polygon is culled, based on its orientation and the CullFace mode, then no fragments are produced during rasterization. Coverage bits that correspond to sample points that satisfy the point sampling criteria are 1, other coverage bits are 0. Each associated datum is produced as described in section 13.7.1, but using the corresponding sample location instead of the fragment center. An implementation may choose to assign the same associated data values to more than one sample by barycentric evaluation using any location within the pixel including the fragment center or one of the sample locations. The flat qualifier affects how shader outputs are interpolated in the same fashion as described for basic polygon rasterization in section 13.7.1.

13.7.4

Polygon Rasterization State

The state required for polygon rasterization consists of whether polygon offsets are enabled or disabled, and the factor and bias values of the polygon offset equation. The initial polygon offset factor and bias values are both 0; initially polygon offset is disabled.

13.8

Early Per-Fragment Tests

Once fragments are produced by rasterization, a number of per-fragment operations are performed prior to fragment shader execution. If a fragment is discarded during any of these operations, it will not be processed by any subsequent stage, including fragment shader execution.

OpenGL ES 3.2 (May 5, 2022)

13.8. EARLY PER-FRAGMENT TESTS

368

Three fragment operations are performed, and a further three are optionally performed on each fragment, in the following order: • the pixel ownership test (see section 13.8.1); • the scissor test (see section 13.8.2); • multisample fragment operations (see section 13.8.3); If early per-fragment operations are enabled, these tests are also performed: • the stencil test (see section 15.1.2); • the depth buffer test (see section 15.1.3); and • occlusion query sample counting (see section 15.1.4).

13.8.1

Pixel Ownership Test

The first test is to determine if the pixel at location (xw , yw ) in the framebuffer is currently owned by the GL (more precisely, by this GL context). If it is not, the window system decides the fate of the incoming fragment. Possible results are that the fragment is discarded or that some subset of the subsequent per-fragment operations are applied to the fragment. This test allows the window system to control the GL’s behavior, for instance, when a GL window is obscured. If the draw framebuffer is a framebuffer object (see section 15.2.1), the pixel ownership test always passes, since the pixels of framebuffer objects are owned by the GL, not the window system. If the draw framebuffer is the default framebuffer, the window system controls pixel ownership.

13.8.2

Scissor Test

The scissor test determines if (xw , yw ) lies within the scissor rectangle defined by four values for each viewport. These values are set with void Scissor( int left, int bottom, sizei width, sizei height ); If left ≤ xw < left +width and bottom ≤ yw < bottom +height, then the scissor test passes. Otherwise, the test fails and the fragment is discarded. The test is enabled or disabled using Enable or Disable with target SCISSOR_TEST. When disabled, it is as if the scissor test always passes.

OpenGL ES 3.2 (May 5, 2022)

13.8. EARLY PER-FRAGMENT TESTS

369

Errors An INVALID_VALUE error is generated if width or height is negative. The state required consists of four integer values and a bit indicating whether the test is enabled or disabled. In the initial state, left = bottom = 0. width and height are set to the width and height, respectively, of the window into which the GL is to do its rendering. If the default framebuffer is bound but no default framebuffer is associated with the GL context (see chapter 9), then width and height are initially set to zero. Initially, the scissor test is disabled.

13.8.3

Multisample Fragment Operations

This step modifies fragment coverage values based on the values of SAMPLE_COVERAGE, SAMPLE_COVERAGE_VALUE, SAMPLE_COVERAGE_INVERT, SAMPLE_MASK, and SAMPLE_MASK_VALUE. If the value of SAMPLE_BUFFERS is not one, this step is skipped. All alpha values in this section refer only to the alpha component of the fragment shader output linked to color number zero (see section 14.2.3). If the fragment shader does not write to this output, the alpha value is undefined. Sample coverage and sample mask operations are enabled or disabled by calling Enable and Disable with targets SAMPLE_COVERAGE or SAMPLE_MASK, respectively. If SAMPLE_COVERAGE is enabled, the fragment coverage value is ANDed with a temporary coverage mask generated from the value of SAMPLE_COVERAGE_VALUE. If the value of SAMPLE_COVERAGE_INVERT is TRUE, this mask is inverted (all bit values are inverted) before it is ANDed with the fragment coverage. Finally, if SAMPLE_MASK is enabled, the fragment coverage is ANDed with the value of SAMPLE_MASK_VALUE. This updated coverage becomes the new fragment coverage value. No specific algorithm is required for converting the sample coverage value to a temporary coverage mask. It is intended that the number of 1’s in this value be proportional to the sample coverage value, with all 1’s corresponding to a value of 1.0 and all 0’s corresponding to 0.0. It is also intended that the algorithm be pseudo-random in nature, to avoid image artifacts due to regular coverage sample locations. The algorithm can and probably should be different at different pixel locations. If it does differ, it should be defined relative to window, not screen, coordinates, so that rendering results are invariant with respect to window position. The values of SAMPLE_COVERAGE_VALUE and SAMPLE_COVERAGE_INVERT are specified by calling OpenGL ES 3.2 (May 5, 2022)

13.8. EARLY PER-FRAGMENT TESTS

370

void SampleCoverage( float value, boolean invert ); with value set to the desired coverage value, and invert set to TRUE or FALSE. value is clamped to [0, 1] before being stored as SAMPLE_COVERAGE_VALUE. These values may be queried as described in table 21.8. The value of SAMPLE_MASK_VALUE is specified using void SampleMaski( uint maskNumber, bitfield mask ); with mask set to the desired mask for mask word maskNumber. Bit B of mask word M corresponds to sample 32M + B as described in section 13.4. The sample mask value is queried by calling GetIntegeri v with target SAMPLE_MASK_VALUE and index set to maskNumber. Errors An INVALID_VALUE error is generated if maskNumber is greater than or equal to the value of MAX_SAMPLE_MASK_WORDS.

13.8.4

The Early Fragment Test Qualifier

The stencil test, depth buffer test and occlusion query sample counting are performed if and only if early fragment tests are enabled in the active fragment shader (see section 14.2.4). When early per-fragment operations are enabled, these operations are performed prior to fragment shader execution, and the stencil buffer, depth buffer, and occlusion query sample counts will be updated accordingly; these operations will not be performed again after fragment shader execution. When there is no active program, the active program has no fragment shader, or the active program was linked with early fragment tests disabled, these operations are performed only after fragment program execution, in the order described in section 15.1. If early fragment tests are enabled, the depth buffer, stencil buffer, and occlusion query sample counts may be updated even for fragments or samples that would be discarded after fragment shader execution due to per-fragment operations such as alpha-to-coverage tests.

OpenGL ES 3.2 (May 5, 2022)

Chapter 14

Programmable Fragment Processing When the program object currently in use for the fragment stage (see section 7.3) includes a fragment shader, its shader is considered active and is used to process fragments resulting from rasterization (see section 13). If the current fragment stage program object has no fragment shader, or no fragment program object is current for the fragment stage, the results of fragment shader execution are undefined. The processed fragments resulting from fragment shader execution are then further processed and written to the framebuffer as described in chapter 15.

14.1

Fragment Shader Variables

Fragment shaders can access uniforms belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. Fragment shaders also have access to samplers to perform texturing operations, as described in section 7.9. Fragment shaders can read input variables or inputs that correspond to the attributes of the fragments produced by rasterization. The OpenGL ES Shading Language Specification defines a set of built-in inputs that can be be accessed by a fragment shader. These built-in inputs include data associated with a fragment such as the fragment’s position. Additionally, the previous active shader stage may define one or more output variables (see section 11.1.2.1 and the OpenGL ES Shading Language Specification). The values of these user-defined outputs are, if not flat shaded, interpolated 371

14.2. SHADER EXECUTION

372

across the primitive being rendered. The results of these interpolations are available when inputs of the same name are defined in the fragment shader. When interpolating input variables, the default screen-space location at which these variables are sampled is defined in previous rasterization sections. The default location may be overriden by interpolation qualifiers. When interpolating variables declared using centroid in, the variable is sampled at a location within the pixel covered by the primitive generating the fragment. When interpolating variables declared using sample in when the value of SAMPLE_BUFFERS is one, the fragment shader will be invoked separately for each covered sample and the variable will be sampled at the corresponding sample point. Additionally, built-in fragment shader functions provide further fine-grained control over interpolation. The built-in functions interpolateAtCentroid and interpolateAtSample will sample variables as though they were declared with the centroid or sample qualifiers, respectively. The built-in function interpolateAtOffset will sample variables at a specified (x, y) offset relative to the center of the pixel. The range and granularity of offsets supported by this function is implementation-dependent. If either component of the specified offset is less than MIN_FRAGMENT_INTERPOLATION_OFFSET or greater than MAX_FRAGMENT_INTERPOLATION_OFFSET, the position used to interpolate the variable is undefined. Not all values of offset may be supported; x and y offsets may be rounded to fixed-point values with the number of fraction bits given by the implementation-dependent constant FRAGMENT_INTERPOLATION_OFFSET_BITS. A fragment shader can also write to output variables. Values written to these outputs are used in the subsequent per-fragment operations. Output variables can be used to write floating-point, integer or unsigned integer values destined for buffers attached to a framebuffer object, or destined for color buffers attached to the default framebuffer. Section 14.2.3 describes how to direct these values to buffers.

14.2

Shader Execution

The executable version of the fragment shader is used to process incoming fragment values that are the result of rasterization. Following shader execution, the fixed-function operations described in chapter 15 are performed. Special considerations for fragment shader execution are described in the following sections.

OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION

14.2.1

373

Texture Access

Section 11.1.3.1 describes texture lookup functionality accessible to a vertex shader. The texel fetch and texture size query functionality described there also applies to fragment shaders. When a texture lookup is performed in a fragment shader, the GL computes the filtered texture value τ in the manner described in sections 8.14 and 8.15, and converts it to a texture base color Cb as shown in table 14.1, followed by swizzling the components of Cb , controlled by the values of the texture parameters TEXTURE_SWIZZLE_R, TEXTURE_SWIZZLE_G, TEXTURE_SWIZZLE_B, and TEXTURE_SWIZZLE_A. If the value of TEXTURE_SWIZZLE_R is denoted by swizzler , swizzling computes the first component of Cs according to if (swizzler == RED) Cs [0] = Cb [0]; else if (swizzler == GREEN) Cs [0] = Cb [1]; else if (swizzler == BLUE) Cs [0] = Cb [2]; else if (swizzler == ALPHA) Cs [0] = Ab ; else if (swizzler == ZERO) Cs [0] = 0; else if (swizzler == ONE) Cs [0] = 1; // float or int depending on texture component type Swizzling of Cs [1], Cs [2], and As are similarly controlled by the values of TEXTURE_SWIZZLE_G, TEXTURE_SWIZZLE_B, and TEXTURE_SWIZZLE_A, re-

spectively. The resulting four-component vector (Rs , Gs , Bs , As ) is returned to the fragment shader. For the purposes of level-of-detail calculations, the derivatives du dx , du dv dv dw dw , , , and may be approximated by a differencing algorithm as dedy dx dy dx dy scribed in section 8.8 (“Texture Functions”) of the OpenGL ES Shading Language Specification. Texture lookups involving textures with depth and/or stencil component data are performed as described in section 11.1.3.5.

14.2.2

Shader Inputs

The OpenGL ES Shading Language Specification describes the values that are available as inputs to the fragment shader. OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION Texture Base Internal Format RED RG RGB RGBA LUMINANCE ALPHA LUMINANCE_ALPHA

374 Texture base color Cb Ab (Rt , 0, 0) 1 (Rt , Gt , 0) 1 (Rt , Gt , Bt ) 1 (Rt , Gt , Bt ) At (Lt , Lt , Lt ) 1 (0, 0, 0) At (Lt , Lt , Lt ) At

Table 14.1: Correspondence of filtered texture components to texture base components. The values Rt , Gt , Bt , At , and Lt are respectively the red, green, blue, alpha, and luminance components of the filtered texture value τ (see table 8.8).

The built-in variable gl_FragCoord holds the fragment coordinate

xw yw zw w1c for the fragment where xw yw zw is the fragment’s window-space position and wc is the w component of the fragment’s clip-space position (see section 12.6). The zw component of gl_FragCoord undergoes an implied conversion to floating-point. This conversion must leave the values 0 and 1 invariant. Note that zw already has a polygon offset added in, if enabled (see section 13.7.2). The built-in variable gl_FrontFacing is set to TRUE if the fragment is generated from a front-facing primitive, and FALSE otherwise. For fragments generated from triangle primitives, the determination is made by examining the sign of the area computed by equation 13.6 of section 13.7.1 (including the possible reversal of this sign controlled by FrontFace). If the sign is positive, fragments generated by the primitive are front-facing; otherwise, they are back-facing. All other fragments are considered front-facing. If a geometry shader is active, the built-in variable gl_PrimitiveID contains the ID value emitted by the geometry shader for the provoking vertex. If no geometry shader is active, gl_PrimitiveID contains the number of primitives processed by the rasterizer since the last drawing command was called. The first primitive generated by a drawing command is numbered zero, and the primitive ID counter is incremented after every individual point, line, or polygon primitive is processed. The counter is reset to zero between each instance drawn. Restarting a primitive using the primitive restart index (see section 10.3) has no effect on the primitive ID counter. gl_PrimitiveID is only defined under the same conditions that gl_-

OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION

375

VertexID is defined, as described under “Shader Inputs” in section 11.1.3.9. The built-in read-only variable gl_SampleID is filled with the sample number of the sample currently being processed. This variable is in the range zero to gl_NumSamples minus one, where gl_NumSamples is the total number of samples

in the framebuffer, or one if rendering to a non-multisample framebuffer. Using gl_SampleID in a fragment shader causes the entire shader to be executed persample. When rendering to a non-multisample buffer, gl_SampleID will always be zero. gl_NumSamples is the sample count of the framebuffer regardless of whether the framebuffer is multisampled or not. The built-in read-only variable gl_SamplePosition contains the position of the current sample within the multi-sample draw buffer. The x and y components of gl_SamplePosition contain the sub-pixel coordinate of the current sample and will have values in the range [0, 1]. The sub-pixel coordinate of the center of the pixel is always (0.5, 0.5). Using gl_SamplePosition in a fragment shader causes the entire shader to be executed per-sample. When rendering to a nonmultisample buffer, gl_SamplePosition will always be (0.5, 0.5). The built-in variable gl_SampleMaskIn is an integer array holding bitfields indicating the set of fragment samples covered by the primitive corresponding to the fragment shader invocation. The number of elements in the array is lsm , 32 where s is the value of the maximum number of color samples supported by the implementation for any renderable internal format. Bit n of element w in the array is set if and only if the sample numbered 32w + n is considered covered for this fragment shader invocation. When rendering to a non-multisample buffer, all bits are zero except for bit zero of the first array element. That bit will be one if the pixel is covered and zero otherwise. Bits in the sample mask corresponding to covered samples that will be killed due to SAMPLE_COVERAGE or SAMPLE_MASK will not be set (see section 13.8.3). When per-sample shading is active due to the use of a fragment input qualified by sample or due to the use of the gl_SampleID or gl_SamplePosition variables, only the bit for the current sample is set in gl_SampleMaskIn. When state specifies multiple fragment shader invocations for a given fragment, the bit corresponding to each covered sample will be set in exactly one fragment shader invocation. Similarly to the limit on geometry shader output components (see section 11.3.4.4), there is a limit on the number of components of built-in and user-defined input variables that can be read by the fragment shader, given by the value of the implementation-dependent constant MAX_FRAGMENT_INPUT_COMPONENTS. OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION

376

When a program is linked, all components of any input variables read by a fragment shader will count against this limit. A program whose fragment shader exceeds this limit may fail to link, unless device-dependent optimizations are able to make the program fit within available hardware resources. Component counting rules for different variable types and variable declarations are the same as for MAX_VERTEX_OUTPUT_COMPONENTS. (see section 11.1.2.1).

14.2.3

Shader Outputs

The OpenGL ES Shading Language Specification describes the values that may be output by a fragment shader. These outputs are split into two categories, userdefined outputs and the built-in outputs gl_FragColor, gl_FragData[n] (both available only in OpenGL ES Shading Language version 1.00), gl_FragDepth and gl_SampleMask. For fixed-point depth buffers, the final fragment depth written by a fragment shader is first clamped to [0, 1] and then converted to fixed-point as if it were a window z value (see section 12.6.1). For floating-point depth buffers, conversion is not performed but clamping is. Note that the depth range computation is not applied here, only the conversion to fixed-point. The built-in integer array gl_SampleMask can be used to change the sample coverage for a fragment from within the shader. The number of elements in the array is lsm , 32 where s is the value of the maximum number of color samples supported by the implementation for any renderable internal format. If bit n of element w in the array is set to zero, sample 32w + n should be considered uncovered for the purposes of additional multisample fragment operations, as described in section 15.1.8, and the corresponding bits in the fragment coverage mask are set to zero. Modifying the sample mask in this way may exclude covered samples from being processed further at a per-fragment granularity. However, setting sample mask bits to one will never enable samples not covered by the original primitive. If the fragment shader is being executed at any frequency other than per-fragment, bits of the sample mask not corresponding to the current fragment shader invocation do not affect the fragment coverage mask. If a fragment shader does not statically assign a value to gl_SampleMask, the fragment coverage mask is not modified. If a value is not assigned to gl_SampleMask due to flow of control, the affected bits of the sample mask are undefined. If there is only a single output variable, it does not need to be explicitly bound to a fragment color within the shader text, in which case it is implicitly bound to OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION

377

fragment color zero. If there is more than one output variable, all output variables must be explicitly bound to fragment colors within the shader text. Missing or conflicting binding assignments will cause CompileShader to fail. Color values written by a fragment shader may be floating-point, signed integer, or unsigned integer. If the color buffer has a signed or unsigned normalized fixed-point format, color values are assumed to be floating-point and are converted to fixed-point as described in equations 2.4 or 2.3, respectively; otherwise no type conversion is applied. If the values written by the fragment shader do not match the format(s) of the corresponding color buffer(s), the result is undefined. Writing to gl_FragColor specifies the fragment color (color number zero) that will be used by subsequent stages of the pipeline. Writing to gl_FragData[n] specifies the value of fragment color number n. Writing to a user-defined output variable specifies the value of the fragment color it is explicitly or implicitly bound to. Any fragment colors which are not written by the fragment shader are undefined. A fragment shader may not statically assign values to both gl_FragColor and gl_FragData[n]. In this case, a compile or link error will result. A shader statically assigns a value to a variable if, after pre-processing, it contains a statement that would write to the variable, whether or not run-time flow of control will cause that statement to be executed. Writing to gl_FragDepth specifies the depth value for the fragment being processed. If the active fragment shader does not statically assign a value to gl_FragDepth, then the depth value generated during rasterization is used by subsequent stages of the pipeline. Otherwise, the value assigned to gl_FragDepth is used, and is undefined for any fragments where statements assigning a value to gl_FragDepth are not executed. Thus, if a shader statically assigns a value to gl_FragDepth, then it is responsible for always writing it. To determine the set of fragment shader output attribute variables used by a program, applications can query the properties and active resources of the PROGRAM_OUTPUT interface of a program including a fragment shader. Additionally, the command int GetFragDataLocation( uint program, const char *name ); is provided to query the location assigned to a fragment shader output variable. Errors If program has been linked successfully but contains no fragment shader,

OpenGL ES 3.2 (May 5, 2022)

14.2. SHADER EXECUTION

378

no error is generated but -1 will be returned. An INVALID_OPERATION error is generated and -1 is returned if program has not been linked successfully. Otherwise, the command is equivalent to GetProgramResourceLocation(program, PROGRAM_OUTPUT, name);

14.2.4

Early Fragment Tests

An explicit control is provided to allow fragment shaders to enable early fragment tests. If the fragment shader specifies the early_fragment_tests layout qualifier, the per-fragment tests described in section 13.8 will be performed prior to fragment shader execution. Otherwise, they will be performed after fragment shader execution.

OpenGL ES 3.2 (May 5, 2022)

Chapter 15

Writing Fragments and Samples to the Framebuffer After programmable fragment processing, per-fragment operations are performed as described in section 15.1, followed by writing to the framebuffer, which is the final set of operations performed as a result of drawing primitives. Additional commands controlling the framebuffer as a whole are described in section 15.2.

15.1

Per-Fragment Operations

A fragment is produced by rasterization with window coordinates of (xw , yw ) and depth z, as described in chapter 13. The fragment is then modified by programmable fragment processing, which adds associated data as described in chapter 14. The fragment is then further modified, and possibly discarded by the perfragment operations described in this chapter. These operations are diagrammed in figure 15.1, in the order in which they are performed. Finally, if the fragment was not discarded, it is used to update the framebuffer at the fragment’s window coordinates. The stencil test, depth test, and occlusion query operations described in sections 15.1.2, 15.1.3, and 15.1.4 may instead be performed prior to fragment processing, as described in section 13.8, if requested by the fragment program.

15.1.1

Alpha To Coverage

This step modifies fragment alpha and coverage values based on the value of SAMPLE_ALPHA_TO_COVERAGE. If the value of SAMPLE_BUFFERS is not one, or 379

15.1. PER-FRAGMENT OPERATIONS

Fragment (or sample) from Rasterization

Fragment Shader

Blending

380

Fragment (or sample) and Associated Data

Alpha To Coverage Operation

Occlusion Query (*)

Depth Buffer Test (*)

Framebuffer

Framebuffer

SRGB Conversion

Dithering

Additional Multisample Fragment Operations

Stencil Test (*)

Framebuffer

Write To Framebuffer

Figure 15.1. Per-fragment operations. The boxes labelled with “(*)” may instead be performend during early per-fragment operations, as described in section 13.8.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

381

if draw buffer zero is not NONE and the buffer it references has an integer format, the operation is skipped. Alpha to coverage is enabled or disabled by calling Enable and Disable with target SAMPLE_ALPHA_TO_COVERAGE All alpha values in this section refer only to the alpha component of the fragment shader output linked to color number zero (see section 14.2.3). If SAMPLE_ALPHA_TO_COVERAGE is enabled, a temporary coverage value is generated where each bit is determined by the alpha value at the corresponding sample location (see section 13.4) of draw buffer zero (see section 14.2.3). The temporary coverage value is then ANDed with the fragment coverage value to generate a new fragment coverage value. This temporary coverage is generated in the same manner as for sample coverage (see section 13.8.3), but as a function of the fragment’s alpha value, clamped to the range [0, 1]. The function need not be identical, but it must have the same properties of proportionality and invariance.

15.1.2

Stencil Test

The stencil test conditionally discards a fragment based on the outcome of a comparison between the value in the stencil buffer at location (xw , yw ) and a reference value. The test is enabled or disabled with the Enable and Disable commands, using the symbolic constant STENCIL_TEST. When disabled, the stencil test and associated modifications are not made, and the fragment is always passed. The stencil test is controlled with void StencilFunc( enum func, int ref, uint mask ); void StencilFuncSeparate( enum face, enum func, int ref, uint mask ); void StencilOp( enum sfail, enum dpfail, enum dppass ); void StencilOpSeparate( enum face, enum sfail, enum dpfail, enum dppass ); There are two sets of stencil-related state, the front stencil state set and the back stencil state set. Stencil tests and writes use the front set of stencil state when processing fragments rasterized from non-polygon primitives (points and lines) and front-facing polygon primitives while the back set of stencil state is used when processing fragments rasterized from back-facing polygon primitives. Whether a polygon is front- or back-facing is determined in the same manner used for face culling (see section 13.7.1).

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

382

StencilFuncSeparate and StencilOpSeparate take a face argument which can be FRONT, BACK, or FRONT_AND_BACK and indicates which set of state is affected. StencilFunc and StencilOp set front and back stencil state to identical values. StencilFunc and StencilFuncSeparate take three arguments that control whether the stencil test passes or fails. ref is an integer reference value that is used in the unsigned stencil comparison. Stencil comparison operations and queries of ref clamp its value to the range [0, 2s − 1], where s is the number of bits in the stencil buffer attached to the draw framebuffer. The s least significant bits of mask are bitwise ANDed with both the reference and the stored stencil value, and the resulting masked values are those that participate in the comparison controlled by func. func is a symbolic constant that determines the stencil comparison function; the eight symbolic constants are NEVER, ALWAYS, LESS, LEQUAL, EQUAL, GEQUAL, GREATER, or NOTEQUAL. Accordingly, the stencil test passes never, always, and if the masked reference value is less than, less than or equal to, equal to, greater than or equal to, greater than, or not equal to the masked stored value in the stencil buffer. StencilOp and StencilOpSeparate take three arguments that indicate what happens to the stored stencil value if this or certain subsequent tests fail or pass. sfail indicates what action is taken if the stencil test fails. The symbolic constants are KEEP, ZERO, REPLACE, INCR, DECR, INVERT, INCR_WRAP, and DECR_WRAP. These correspond to keeping the current value, setting to zero, replacing with the reference value, incrementing with saturation, decrementing with saturation, bitwise inverting it, incrementing without saturation, and decrementing without saturation. For purposes of increment and decrement, the stencil bits are considered as an unsigned integer. Incrementing or decrementing with saturation clamps the stencil value at 0 and the maximum representable value. Incrementing or decrementing without saturation will wrap such that incrementing the maximum representable value results in 0, and decrementing 0 results in the maximum representable value. The same symbolic values are given to indicate the stencil action if the depth buffer test (see section 15.1.3) fails (dpfail), or if it passes (dppass). If the stencil test fails, the incoming fragment is discarded. The state required consists of the most recent values passed to StencilFunc or StencilFuncSeparate and to StencilOp or StencilOpSeparate, and a bit indicating whether stencil testing is enabled or disabled. In the initial state, stenciling is disabled, the front and back stencil reference value are both zero, the front and back stencil comparison functions are both ALWAYS, and the front and back stencil mask are both set to the value 2s − 1, where s is greater than or equal to the number of bits in the deepest stencil buffer supported by the GL implementation. Initially, all three front and back stencil operations are KEEP. OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

383

If there is no stencil buffer, no stencil modification can occur, and it is as if the stencil tests always pass, regardless of any calls to StencilFunc.

15.1.3

Depth Buffer Test

The depth buffer test discards the incoming fragment if a depth comparison fails. The comparison is enabled or disabled by calling Enable and Disable with target DEPTH_TEST. When disabled, the depth comparison and subsequent possible updates to the depth buffer value are bypassed and the fragment is passed to the next operation. The stencil value, however, is modified as indicated below as if the depth buffer test passed. If enabled, the comparison takes place and the depth buffer and stencil value may subsequently be modified. The comparison is specified with void DepthFunc( enum func ); This command takes a single symbolic constant: one of NEVER, ALWAYS, LESS, LEQUAL, EQUAL, GREATER, GEQUAL, NOTEQUAL. Accordingly, the depth buffer test passes never, always, if the incoming fragment’s zw value is less than, less than or equal to, equal to, greater than, greater than or equal to, or not equal to the depth value stored at the location given by the incoming fragment’s (xw , yw ) coordinates. If the depth buffer test fails, the incoming fragment is discarded. The stencil value at the fragment’s (xw , yw ) coordinates is updated according to the function currently in effect for depth buffer test failure. Otherwise, the fragment continues to the next operation and the value of the depth buffer at the fragment’s (xw , yw ) location is set to the fragment’s zw value. In this case the stencil value is updated according to the function currently in effect for depth buffer test success. The necessary state is an eight-valued integer and a single bit indicating whether depth buffering is enabled or disabled. In the initial state the function is LESS and the test is disabled. If there is no depth buffer, it is as if the depth buffer test always passes.

15.1.4

Occlusion Queries

Occlusion queries use query objects to track the number of fragments that are not discarded by earlier stages. An occlusion query can be started and finished by calling BeginQuery and EndQuery, respectively, with a target ANY_SAMPLES_PASSED or ANY_SAMPLES_PASSED_CONSERVATIVE.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

384

When an occlusion query is started with the target ANY_SAMPLES_PASSED, the samples-boolean state maintained by the GL is set to FALSE. While that occlusion query is active, the samples-boolean state is set to TRUE for each fragment still being processed after the depth test. When the target is ANY_SAMPLES_PASSED_CONSERVATIVE, an implementation may choose to use a less precise version of the test which can additionally set the samples-boolean state to TRUE in some other implementation-dependent cases. This may offer better performance on some implementations at the expense of false positives. When the occlusion query finishes, the samples-boolean state of FALSE or TRUE is written to the corresponding query object as the query result value, and the query result for that object is marked as available.

15.1.5

Blending

Blending combines the incoming source fragment’s R, G, B, and A values with the destination R, G, B, and A values stored in the framebuffer at the fragment’s (xw , yw ) location. Source and destination values are combined according to the blend equation, quadruplets of source and destination weighting factors determined by the blend functions, and a constant blend color to obtain a new set of R, G, B, and A values, as described below. If the color buffer is fixed-point, the components of the source and destination values and blend factors are each clamped to [0, 1] for an unsigned normalized color buffer prior to evaluating the blend equation. If the color buffer is floatingpoint, no clamping occurs. The resulting four values are sent to the next operation. Blending applies only if the color buffer has a fixed-point or floating-point format. If the color buffer has an integer format, proceed to the next operation. Blending is enabled or disabled for an individual draw buffer with the commands void Enablei( enum target, uint index ); void Disablei( enum target, uint index ); target is the symbolic constant BLEND and index is an integer i specifying the draw buffer associated with the symbolic constant DRAW_BUFFERi. Blending can be enabled or disabled for all draw buffers using Enable or Disable with target BLEND. If blending is disabled for a particular draw buffer, proceed to the next operation. If one or more fragment colors are being written to multiple buffers (see section 15.2.1), blending is computed and applied separately for each fragment color and the corresponding buffer. OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

385

Errors An INVALID_VALUE error is generated by Enablei, Disablei and IsEnabledi if target is BLEND and index is greater than or equal to the value of MAX_DRAW_BUFFERS. An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if blending is enabled (see below) and any draw buffer has 32-bit floating-point format components. 15.1.5.1

Blend Equation

Blending is controlled by the blend equation. This equation can be simultaneously set to the same value for all draw buffers using the commands void BlendEquation( enum mode ); void BlendEquationSeparate( enum modeRGB, enum modeAlpha ); or for an individual draw buffer using the indexed commands void BlendEquationi( uint buf, enum mode ); void BlendEquationSeparatei( uint buf, enum modeRGB, enum modeAlpha ); BlendEquationSeparate and BlendEquationSeparatei argument modeRGB determines the RGB blend function while modeAlpha determines the alpha blend equation. BlendEquation and BlendEquationi argument mode determines both the RGB and alpha blend equations. BlendEquation and BlendEquationSeparate modify the blend equations for all draw buffers. BlendEquationi and BlendEquationSeparatei modify the blend equations associated with an individual draw buffer. The buf argument is an integer i that indicates that the blend equations should be modified for DRAW_BUFFERi. Errors An INVALID_VALUE error is generated if buf is not in the range zero to the value of MAX_DRAW_BUFFERS minus one. An INVALID_ENUM error is generated by BlendEquation if mode is not one of the blend equation modes in tables 15.1, 15.3, and 15.4. An INVALID_ENUM error is generated by BlendEquationSeparate if either modeRGB or modeAlpha is not one of the blend equation modes in ta-

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

386

ble 15.1. Unsigned normalized fixed-point destination (framebuffer) components are represented as described in section 2.3.5. Constant color components, floatingpoint destination components, and source (fragment) components are taken to be floating-point values. If source components are represented internally by the GL as fixed-point values, they are also interpreted according to section 2.3.5. Prior to blending, unsigned normalized fixed-point color components undergo an implied conversion to floating-point using equation 2.1. This conversion must leave the values zero and one invariant. Blending computations are treated as if carried out in floating-point. For the equations in table 15.1, computations will be performed with a precision and dynamic range no lower than that used to represent destination components. For the equations in tables 15.3 and 15.4, computations will be performed with a precision and dynamic range no lower than the smaller of that used to represent destination components or that used to represent 16-bit floating-point values, as described in section 2.3.4.2. If the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the destination buffer is SRGB (see section 9.2.3), the R, G, and B destination color values (after conversion from fixedpoint to floating-point) are considered to be encoded for the sRGB color space and hence must be linearized prior to their use in blending. Each R, G, and B component is converted in the same fashion described for sRGB texture components in section 8.21. If the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING is not SRGB, no linearization is performed. The resulting linearized R, G, and B and unmodified A values are recombined as the destination color used in blending computations. Table 15.1 provides the corresponding per-component blend equations for each mode, whether acting on RGB components for modeRGB or the alpha component for modeAlpha. In the table, the s subscript on a color component abbreviation (R, G, B, or A) refers to the source color component for an incoming fragment, the d subscript on a color component abbreviation refers to the destination color component at the corresponding framebuffer location, and the c subscript on a color component abbreviation refers to the constant blend color component. A color component abbreviation without a subscript refers to the new color component resulting from blending. Additionally, Sr , Sg , Sb , and Sa are the red, green, blue, and alpha components of the source weighting factors determined by the source blend function, and Dr , Dg , Db , and Da are the red, green, blue, and alpha components of the destination weighting factors determined by the destination blend function. Blend OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS Mode FUNC_ADD

FUNC_SUBTRACT

FUNC_REVERSE_SUBTRACT

MIN

MAX

RGB Components R = Rs ∗ Sr + Rd ∗ Dr G = Gs ∗ S g + Gd ∗ D g B = Bs ∗ Sb + Bd ∗ Db R = Rs ∗ Sr − Rd ∗ Dr G = Gs ∗ S g − Gd ∗ D g B = Bs ∗ Sb − Bd ∗ Db R = Rd ∗ Dr − Rs ∗ Sr G = Gd ∗ Dg − Gs ∗ Sg B = Bd ∗ Db − Bs ∗ Sb R = min(Rs , Rd ) G = min(Gs , Gd ) B = min(Bs , Bd ) R = max(Rs , Rd ) G = max(Gs , Gd ) B = max(Bs , Bd )

387 Alpha Component A = As ∗ Sa + Ad ∗ Da

A = As ∗ Sa − Ad ∗ Da

A = Ad ∗ D a − As ∗ S a

A = min(As , Ad )

A = max(As , Ad )

Table 15.1: RGB and alpha blend equations.

functions are described below. 15.1.5.2

Blend Functions

The weighting factors used by the blend equation are determined by the blend functions. There are three possible sources for weighting factors. These are the constant color (Rc , Gc , Bc , Ac ) set with BlendColor (see below), the source color (Rs , Gs , Bs , As ), and the destination color (the existing content of the draw buffer). Additionally the special constants ZERO and ONE are available as weighting factors. Blend functions are simultaneously specified for all draw buffers using the commands void BlendFunc( enum src, enum dst ); void BlendFuncSeparate( enum srcRGB, enum dstRGB, enum srcAlpha, enum dstAlpha ); or for an individual draw buffer using the indexed commands void BlendFunci( uint buf, enum src, enum dst ); OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS Function ZERO ONE SRC_COLOR ONE_MINUS_SRC_COLOR DST_COLOR ONE_MINUS_DST_COLOR SRC_ALPHA ONE_MINUS_SRC_ALPHA DST_ALPHA ONE_MINUS_DST_ALPHA CONSTANT_COLOR ONE_MINUS_CONSTANT_COLOR CONSTANT_ALPHA ONE_MINUS_CONSTANT_ALPHA SRC_ALPHA_SATURATE

RGB Blend Factors (Sr , Sg , Sb ) or (Dr , Dg , Db ) (0, 0, 0) (1, 1, 1) (Rs , Gs , Bs ) (1, 1, 1) − (Rs , Gs , Bs ) (Rd , Gd , Bd ) (1, 1, 1) − (Rd , Gd , Bd ) (As , As , As ) (1, 1, 1) − (As , As , As ) (Ad , Ad , Ad ) (1, 1, 1) − (Ad , Ad , Ad ) (Rc , Gc , Bc ) (1, 1, 1) − (Rc , Gc , Bc ) (Ac , Ac , Ac ) (1, 1, 1) − (Ac , Ac , Ac ) (f, f, f )1

388 Alpha Blend Factor Sa or Da 0 1 As 1 − As Ad 1 − Ad As 1 − As Ad 1 − Ad Ac 1 − Ac Ac 1 − Ac 1

Table 15.2: RGB and ALPHA source and destination blending functions and the corresponding blend factors. Addition and subtraction of triplets is performed component-wise. 1 f = min(A , 1 − A ). s d

void BlendFuncSeparatei( uint buf, enum srcRGB, enum dstRGB, enum srcAlpha, enum dstAlpha ); BlendFuncSeparate and BlendFuncSeparatei arguments srcRGB and dstRGB determine the source and destination RGB blend functions, respectively, while srcAlpha and dstAlpha determine the source and destination alpha blend functions. BlendFunc and BlendFunci argument src determines both RGB and alpha source functions, while dst determines both RGB and alpha destination functions. BlendFuncSeparate and BlendFunc modify the blend functions for all draw buffers. BlendFuncSeparatei and BlendFunci modify the blend functions associated with an individual draw buffer. The buf argument is an integer i that indicates that the blend functions should be modified for DRAW_BUFFERi. The possible source and destination blend functions and their corresponding computed blend factors are summarized in table 15.2.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

389

Errors An INVALID_VALUE error is generated if buf is not in the range zero to the value of MAX_DRAW_BUFFERS minus one. An INVALID_ENUM error is generated if any of src, dst, srcRGB, dstRGB, srcAlpha, or dstAlpha are not one of the blend functions in table 15.2. 15.1.5.3

Advanced Blend Equations

The advanced blend equations are those listed in tables 15.3 and 15.4. When using one of these equations, blending is performed according to the following equations: 0

0

0

0

0

0

0

0

0

0

0

0

R = f (Rs , Rd )p0 (As , Ad ) + Y Rs p1 (As , Ad ) + ZRd p2 (As , Ad ) G = f (Gs , Gd )p0 (As , Ad ) + Y Gs p1 (As , Ad ) + ZGd p2 (As , Ad ) B = f (Bs , Bd )p0 (As , Ad ) + Y Bs p1 (As , Ad ) + ZBd p2 (As , Ad ) A = Xp0 (As , Ad ) + Y p1 (As , Ad ) + Zp2 (As , Ad ) where the function f and terms X, Y , and Z are specified in the tables. The R, G, and B components of the source color used for blending are considered to have been premultiplied by the A component prior to blending. The base source color 0 0 0 (Rs , Gs , Bs ) is obtained by dividing through by the A component: ( (0, 0, 0), (Rs , Gs , Bs ) =  Rs Gs Bs  As , As , As , 0

0

0

As = 0 otherwise

The destination color components are always considered to have been premultiplied by the destination A component, and the base destination color 0 0 0 (Rd , Gd , Bd ) is obtained by dividing through by the A component: ( (0, 0, 0), (Rd , Gd , Bd ) =  Rd Gd Bd  Ad , Ad , Ad , 0

0

0

Ad = 0 otherwise

When blending using advanced blend equations, we expect that the R, G, and B components of premultiplied source and destination color inputs be stored as the product of non-premultiplied R, G, and B components and the A component of the color. If any R, G, or B component of a premultiplied input color is non-zero and the A component is zero, the color is considered ill-formed, and the corresponding component of the blend result will be undefined. OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

390

The weighting functions p0 , p1 , and p2 are defined as follows: p0 (As , Ad ) = As Ad p1 (As , Ad ) = As (1 − Ad ) p2 (As , Ad ) = Ad (1 − As ) In these functions, the A components of the source and destination colors are taken to indicate the portion of the pixel covered by the fragment (source) and the fragments previously accumulated in the pixel (destination). The functions p0 , p1 , and p2 approximate the relative portion of the pixel covered by the intersection of the source and destination, covered only by the source, and covered only by the destination, respectively. The equations defined here assume that there is no correlation between the source and destination coverage. When using one of the HSL blend equations in table 15.4 as the blend equation, the RGB color components produced by the function f are effectively obtained by converting both the non-premultiplied source and destination colors to the HSL (hue, saturation, luminosity) color space, generating a new HSL color by selecting H, S, and L components from the source or destination according to the blend equation, and then converting the result back to RGB. The HSL blend equations are only well defined when the values of the input color components are in the range [0, 1] In the equations below, a blended RGB color is produced according to the following pseudocode: float minv3(vec3 c) { return min(min(c.r, c.g), c.b); } float maxv3(vec3 c) { return max(max(c.r, c.g), c.b); } float lumv3(vec3 c) { return dot(c, vec3(0.30, 0.59, 0.11)); } float satv3(vec3 c) { return maxv3(c) - minv3(c); } // If any color components are outside [0,1], adjust the color to // get the components in range. vec3 ClipColor(vec3 color) { float lum = lumv3(color); OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

Mode MULTIPLY SCREEN

(X, Y, Z) (1, 1, 1) (1, 1, 1)

OVERLAY

(1, 1, 1)

DARKEN LIGHTEN

(1, 1, 1) (1, 1, 1)

COLORDODGE

(1, 1, 1)

COLORBURN

(1, 1, 1)

HARDLIGHT

(1, 1, 1)

SOFTLIGHT

(1, 1, 1)

DIFFERENCE EXCLUSION

(1, 1, 1) (1, 1, 1)

391

Blend Coefficients f (Cs , Cd ) Cs Cd C (s + Cd − Cs Cd 2Cs Cd , Cd ≤ 0.5 1 − 2(1 − Cs )(1 − Cd ) otherwise min(Cs , Cd ) max(Cs , Cd )   Cd ≤ 0 0, Cd min(1, 1−Cs ), Cd > 0, Cs < 1   1, Cd > 0, Cs ≥ 1   Cd ≥ 1 1, 1−Cd 1 − min(1, Cs ), Cd < 1, Cs > 0   0, Cd < 1, Cs ≤ 0 ( 2Cs Cd , Cs ≤ 0.5 1 − 2(1 − Cs )(1 − Cd ), otherwise   Cs ≤ 0.5 Cd − (1 − 2Cs )Cd (1 − Cd ), Cd + (2Cs − 1)Cd ((16Cd − 12)Cd + 3), Cs > 0.5, Cd ≤ 0.25  √  Cd + (2Cs − 1)( Cd − Cd ), Cs > 0.5, Cd > 0.25 |Cd − Cs | Cs + Cd − 2Cs Cd

Table 15.3: Advanced Blend Equations

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

392

float mincol = minv3(color); float maxcol = maxv3(color); if (mincol < 0.0) { color = lum + ((color-lum)*lum) / (lum-mincol); } if (maxcol > 1.0) { color = lum + ((color-lum)*(1-lum)) / (maxcol-lum); } return color; } // Take the base RGB color cbase and override its // luminosity with that of the RGB color clum. vec3 SetLum(vec3 cbase, vec3 clum) { float lbase = lumv3(cbase); float llum = lumv3(clum); float ldiff = llum - lbase; vec3 color = cbase + vec3(ldiff); return ClipColor(color); } // Take the base RGB color cbase and override its saturation // with that of the RGB color csat. Then override the // luminosity of the result with that of the RGB color clum. vec3 SetLumSat(vec3 cbase, vec3 csat, vec3 clum) { float minbase = minv3(cbase); float sbase = satv3(cbase); float ssat = satv3(csat); vec3 color; if (sbase > 0) { // Equivalent (modulo rounding errors) to setting // the smallest (R,G,B) component to 0, the largest // to ssat, and interpolating the "middle" // component based on its original value relative // to the smallest/largest. color = (cbase - minbase) * ssat / sbase; } else { color = vec3(0.0); } return SetLum(color, clum); OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS Mode HSL_HUE HSL_SATURATION HSL_COLOR HSL_LUMINOSITY

393

Blend Coefficients (X, Y, Z) f (Cs , Cd ) (1, 1, 1) SetLumSat(Cs , Cd , Cd ) (1, 1, 1) SetLumSat(Cd , Cs , Cd ) (1, 1, 1) SetLum(Cs , Cd ) (1, 1, 1) SetLum(Cd , Cs )

Table 15.4: Hue-Saturation-Luminosity Advanced Blend Equations

} Advanced blending equations are supported only when rendering to a single color buffer using fragment color zero. Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if any non-NONE draw buffer uses a blend equation found in table 15.3 or 15.4, and • the draw buffer for color output zero selects multiple color buffers; or • the draw buffer for any other color output is not NONE. When using advanced blending equations, applications should split their rendering into a collection of blending passes, none of which touch an individual sample in the framebuffer more than once. The results of blending are undefined if the sample being blended has been touched previously in the same pass. The command void BlendBarrier( void ); specifies a boundary between passes when using advanced blend equations. Any command that causes the value of a sample to be modified using the framebuffer is considered to touch the sample, including clears, blended or unblended primitives, and BlitFramebuffer copies. Advanced blending equations require the use of a fragment shader with a matching blend_support layout qualifier.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

394

Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if blending is enabled, the current blend equation is found in table 15.3 or 15.4, and the active fragment shader does not include the layout qualifier matching the blend equation or blend_support_all_equations. The set of layout qualifiers supported in fragment shaders is specified in section 4.3.8.2 (“Texture Functions”) of the OpenGL ES Shading Language Specification. 15.1.5.4

Blend Color

The constant color Cc to be used in blending is specified with the command void BlendColor( float red, float green, float blue, float alpha ); The constant color can be used in both the source and destination blending functions. If destination framebuffer components use an unsigned normalized fixed-point representation, the constant color components are clamped to the range [0, 1] when computing blend factors. 15.1.5.5

Blending State

The state required for blending, for each draw buffer, is two integers for the RGB and alpha blend equations, four integers indicating the source and destination RGB and alpha blending functions, and a bit indicating whether blending is enabled or disabled. Additionally, four floating-point values to store the RGBA constant blend color are required. For all draw buffers, the initial blend equations for RGB and alpha are both FUNC_ADD, and the initial blending functions are ONE for the source RGB and alpha functions and ZERO for the destination RGB and alpha functions. Initially, blending is disabled for all draw buffers. The initial constant blend color is (R, G, B, A) = (0, 0, 0, 0). The value of the blend enable for draw buffer i may be queried by calling IsEnabledi with target BLEND and index i, and the values of the blend equations and functions may be queried by calling GetIntegeri v with the corresponding target as shown in table 21.13 and index i.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

395

The value of the blend enable, or the blend equations and functions for draw buffer zero may also be queried by calling IsEnabled or GetIntegerv respectively, with the same target but no index parameter. Blending occurs once for each color buffer currently enabled for blending and for writing (see section 15.2.1), using each buffer’s color for Cd . If a color buffer has no A value, then Ad is taken to be 1.

15.1.6

sRGB Conversion

If the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the destination buffer is SRGB1 (see section 9.2.3), the R, G, and B values after blending are converted into the non-linear sRGB color space by computing   0.0, cl ≤ 0    12.92c , 0 < cl < 0.0031308 l cs = (15.1) 0.41666  − 0.055, 0.0031308 ≤ cl < 1 1.055cl    1.0, cl ≥ 1 where cl is the R, G, or B element and cs is the result (effectively converted into an sRGB color space). If FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING is not SRGB, then cs = cl . The resulting cs values for R, G, and B, and the unmodified A form a new RGBA color value. If the color buffer is fixed-point, each component is clamped to the range [0, 1] and then converted to a fixed-point value using equation 2.3. The resulting four values are sent to the subsequent dithering operation.

15.1.7

Dithering

Dithering selects between two representable color values or indices. A representable value is a value that has an exact representation in the color buffer. Dithering selects, for each color component, either the largest representable color value (for that particular color component) that is less than or equal to the incoming color component value, c, or the smallest representable color value that is greater than or equal to c. The selection may depend on the xw and yw coordinates of the pixel, 1 Note that only unsigned normalized fixed-point color buffers may be SRGB-encoded. Signed normalized fixed-point + SRGB encoding is not defined.

OpenGL ES 3.2 (May 5, 2022)

15.1. PER-FRAGMENT OPERATIONS

396

as well as on the exact value of c. If one of the two values does not exist, then the selection defaults to the other value. Many dithering selection algorithms are possible, but an individual selection must depend only on the incoming component value and the fragment’s x and y window coordinates. If dithering is disabled, then one of the two values above is selected, in an implementation-dependent manner that must not depend on the xw and yw coordinates of the pixel. Dithering is enabled and disabled by calling Enable or Disable with target DITHER. The state required is a single bit. Initially, dithering is enabled.

15.1.8

Additional Multisample Fragment Operations

If the DrawBuffer mode (see section 15.2.1) is NONE, no change is made to any multisample or color buffer. Otherwise, fragment processing is as described below. If the value of SAMPLE_BUFFERS is one, the stencil test, depth test, blending, and dithering are performed for each pixel sample, rather than just once for each fragment. Failure of the stencil or depth test results in termination of the processing of that sample, rather than discarding of the fragment. All operations are performed on the color, depth, and stencil values stored in the multisample renderbuffer attachments if a draw framebuffer object is bound, or otherwise in the multisample buffer of the default framebuffer. The contents of the color buffers are not modified at this point. Stencil, depth, blending, dithering, and logical operations are performed for a pixel sample only if that sample’s fragment coverage bit is a value of 1. If the corresponding coverage bit is 0, no operations are performed for that sample. If a draw framebuffer object is not bound, after all operations have been completed on the multisample buffer, the sample values for each color in the multisample buffer are combined to produce a single color value, and that value is written into the corresponding color buffer selected by DrawBuffers. An implementation may defer the writing of the color buffers until a later time, but the state of the framebuffer must behave as if the color buffers were updated as each fragment was processed. The method of combination is not specified. If the framebuffer contains sRGB values, then it is recommended that the an average of sample values is computed in a linearized space, as for blending (see section 15.1.5). Otherwise, a simple average computed independently for each color component is recommended.

OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS Symbolic Constant NONE COLOR_ATTACHMENTi (see caption)

397

Meaning No buffer Output fragment color to image attached at color attachment point i

Table 15.5: Arguments to DrawBuffers and ReadBuffer when the context is bound to a framebuffer object, and the buffers they indicate. i in COLOR_ATTACHMENTi may range from zero to the value of MAX_COLOR_ATTACHMENTS minus one.

15.2

Whole Framebuffer Operations

The preceding sections described the operations that occur as individual fragments are sent to the framebuffer. This section describes operations that control or affect the whole framebuffer.

15.2.1

Selecting Buffers for Writing

The first such operation is controlling the color buffers into which each of the fragment color values is written. This is accomplished with DrawBuffers. The command void DrawBuffers( sizei n, const enum *bufs ); defines the draw buffers to which all fragment colors are written. n specifies the number of buffers in bufs. bufs is a pointer to an array of symbolic constants specifying the buffer to which each fragment color is written. Each buffer listed in bufs must be BACK, NONE, or one of the values from table 15.5. Further, acceptable values for the constants in bufs depend on whether the GL is using the default framebuffer (the value of DRAW_FRAMEBUFFER_BINDING is zero), or a framebuffer object (the value of DRAW_FRAMEBUFFER_BINDING is non-zero). For more information about framebuffer objects, see section 9. If the GL is bound to the default framebuffer, then n must be 1 and the constant must be BACK or NONE. When draw buffer zero is BACK, color values are written into the sole buffer for single-buffered contexts, or into the back buffer for doublebuffered contexts. If the GL is bound to a draw framebuffer object, then each of the constants must be one of the values listed in table 15.5. Calling DrawBuffers with 0 as the value of n is equivalent to setting all the draw buffers to NONE.

OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

398

In both cases, the draw buffers being defined correspond in order to the respective fragment colors. The draw buffer for fragment colors beyond n is set to NONE. The maximum number of draw buffers is implementation-dependent. The number of draw buffers supported can be queried by calling GetIntegerv with the symbolic constant MAX_DRAW_BUFFERS. If the GL is bound to a draw framebuffer object, the ith buffer listed in bufs must be COLOR_ATTACHMENTi or NONE. Fragment color(s) are written by the fragment shader as described in section 14.2.3. DrawBuffers specifies a set of draw buffers into which the fragment color(s) are written. If a fragment color is undefined after fragment shader execution, or if the value written for a fragment color does not match the format of a draw buffer it is to be written to, the resulting value in the draw buffer is undefined. The order of writes to user-defined output variables is undefined. If the same image is attached to multiple attachment points of a framebuffer object and different values are written to outputs bound to those attachments, the resulting value in the attached image is undefined. Similarly undefined behavior results during any other per-fragment operations where a multiply-attached image may be written to by more than one output, such as during blending. Errors An INVALID_VALUE error is generated if n is negative, or greater than the value of MAX_DRAW_BUFFERS. An INVALID_ENUM error is generated if any value in bufs is not one of the values in tables 15.5, BACK, or NONE. An INVALID_OPERATION error is generated if the GL is bound to the default framebuffer and n is not 1, or *bufs is a value other than BACK or NONE. An INVALID_OPERATION error is generated if the GL is bound to a draw framebuffer object and the ith argument is a value other than COLOR_ATTACHMENTi or NONE. Indicating a buffer or buffers using DrawBuffers causes subsequent pixel color value writes to affect the indicated buffers. If the GL is bound to a draw framebuffer object and a draw buffer selects an attachment that has no image attached, then that fragment color is not written. Specifying NONE as the draw buffer for a fragment color will inhibit that fragment color from being written. The state required to handle color buffer selection for each framebuffer is an

OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

399

integer for each supported fragment color. For the default framebuffer, in the initial state the draw buffer for fragment color zero is BACK if there is a default framebuffer associated with the context, otherwise NONE. For framebuffer objects, in the initial state the draw buffer for fragment color zero is COLOR_ATTACHMENT0. For both the default framebuffer and framebuffer objects, the initial state of draw buffers for fragment colors other than zero is NONE. The draw buffer of the currently bound draw framebuffer selected for fragment color i can be queried by calling GetIntegerv with pname set to DRAW_BUFFERi.

15.2.2

Fine Control of Buffer Updates

Writing of bits to each of the logical buffers after all per-fragment operations have been performed may be masked. The commands void ColorMask( boolean r, boolean g, boolean b, boolean a ); void ColorMaski( uint buf, boolean r, boolean g, boolean b, boolean a ); control writes to the active draw buffers. ColorMask and ColorMaski are used to mask the writing of R, G, B and A values to the draw buffer or buffers. ColorMaski sets the mask for a particular draw buffer. The mask for DRAW_BUFFERi is modified by passing i as the parameter buf. r, g, b, and a indicate whether R, G, B, or A values, respectively, are written or not (a value of TRUE means that the corresponding value is written). The mask specified by r, g, b, and a is applied to the color buffer associated with DRAW_BUFFERi. ColorMask sets the mask for all draw buffers to the same values as specified by r, g, b, and a. Errors An INVALID_VALUE error is generated by ColorMaski if buf is greater than the value of MAX_DRAW_BUFFERS minus one. In the initial state, all color values are enabled for writing for all draw buffers. The value of the color writemask for draw buffer i may be queried by calling GetBooleani v with target COLOR_WRITEMASK and index i. The value of the color writemask for draw buffer zero may also be queried by calling GetBooleanv with pname COLOR_WRITEMASK. The depth buffer can be enabled or disabled for writing zw values using OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

400

void DepthMask( boolean mask ); If mask is non-zero, the depth buffer is enabled for writing; otherwise, it is disabled. In the initial state, the depth buffer is enabled for writing. The commands void StencilMask( uint mask ); void StencilMaskSeparate( enum face, uint mask ); control the writing of particular bits into the stencil planes. The least significant s bits of mask, where s is the number of bits in the stencil buffer, specify an integer mask. Where a 1 appears in this mask, the corresponding bit in the stencil buffer is written; where a 0 appears, the bit is not written. The face parameter of StencilMaskSeparate can be FRONT, BACK, or FRONT_AND_BACK and indicates whether the front or back stencil mask state is affected. StencilMask sets both front and back stencil mask state to identical values. Fragments generated by front-facing primitives use the front mask and fragments generated by back-facing primitives use the back mask (see section 15.1.2). The clear operation always uses the front stencil write mask when clearing the stencil buffer. The state required for the various masking operations is two integers for the front and back stencil values, and a bit for depth values. A set of four bits is also required indicating which color components of an RGBA value should be written. In the initial state, the integer masks are all ones, as are the bits controlling depth value and RGBA component writing. 15.2.2.1

Fine Control of Multisample Buffer Updates

When a framebuffer object is not bound and the value of SAMPLE_BUFFERS is one, ColorMask, DepthMask, and StencilMask or StencilMaskSeparate control the modification of values in the multisample buffer. The color mask has no effect on modifications to the color buffers. If the color mask is entirely disabled, the color sample values must still be combined (as described above) and the result used to replace the color values of the buffers enabled by DrawBuffer.

15.2.3

Clearing the Buffers

The GL provides a means for setting portions of every pixel in a particular buffer to the same value. The argument to void Clear( bitfield buf ); OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS is zero or the bitwise OR of one or more values indicating which buffers are to be cleared. The values are COLOR_BUFFER_BIT, DEPTH_BUFFER_BIT, and STENCIL_BUFFER_BIT, indicating the buffers currently enabled for color writing, the depth buffer, and the stencil buffer (see below), respectively. The value to which each buffer is cleared depends on the setting of the clear value for that buffer. If buf is zero, no buffers are cleared. Errors An INVALID_VALUE error is generated if buf contains any bits other than COLOR_BUFFER_BIT, DEPTH_BUFFER_BIT, or STENCIL_BUFFER_BIT. void ClearColor( float r, float g, float b, float a ); sets the clear value for fixed-point and floating-point color buffers. The specified components are stored as floating-point values. Unsigned normalized fixed-point RGBA color buffers are cleared to color values derived by clamping each component of the clear color to the range [0, 1], then converting the (possibly sRGB converted and/or dithered) color to fixed-point using equations 2.3 or 2.4, respectively. The result of clearing integer color buffers with Clear is undefined. The command void ClearDepthf( float d ); sets the depth value used when clearing the depth buffer. d is clamped to the range [0, 1] when specified. When clearing a fixed-point depth buffer, d is converted to fixed-point according to the rules for a window z value given in section 12.6.1. No conversion is applied when clearing a floating-point depth buffer. The command void ClearStencil( int s ); takes a single integer argument that is the value to which to clear the stencil buffer. s is masked to the number of bitplanes in the stencil buffer. When Clear is called, the only per-fragment operations that are applied (if enabled) are the pixel ownership test, the scissor test, sRGB conversion (see section 15.1.6), and dithering. The masking operations described in section 15.2.2 are also applied. If a buffer is not present, then a Clear directed at that buffer has no effect. The state required for clearing is a clear value for each of the color buffer, the depth buffer, and the stencil buffer. Initially, the RGBA color clear value is (0.0, 0.0, 0.0, 0.0), the depth buffer clear value is 1.0, and the stencil buffer clear index is 0. OpenGL ES 3.2 (May 5, 2022)

401

15.2. WHOLE FRAMEBUFFER OPERATIONS 15.2.3.1

402

Clearing Individual Buffers

Individual buffers of the currently bound draw framebuffer may be cleared with the command void ClearBuffer{if ui}v( enum buffer, int drawbuffer, const T *value ); where buffer and drawbuffer identify a buffer to clear, and value specifies the value or values to clear it to. ClearBufferfv, ClearBufferiv, and ClearBufferuiv should be used to clear fixed- and floating-point, signed integer, and unsigned integer color buffers respectively. If buffer is COLOR, a particular draw buffer DRAW_BUFFERi is specified by passing i as the parameter drawbuffer, and value points to a four-element vector specifying the R, G, B, and A color to clear that draw buffer to. If the value of DRAW_BUFFERi is NONE, the command has no effect. Otherwise, the value of DRAW_BUFFERi is BACK or one of the possible values in tables 15.5 identifying the color buffer to clear. Clamping and conversion for fixed-point color buffers are performed in the same fashion as ClearColor. If buffer is DEPTH, drawbuffer must be zero, and value points to the single depth value to clear the depth buffer to. Clamping and type conversion for fixedpoint depth buffers are performed in the same fashion as ClearDepth. Only ClearBufferfv should be clear depth buffers; neither ClearBufferiv nor ClearBufferuiv accept a buffer of DEPTH. If buffer is STENCIL, drawbuffer must be zero, and value points to the single stencil value to clear the stencil buffer to. Masking is performed in the same fashion as ClearStencil. Only ClearBufferiv should be used to clear stencil buffers; neither ClearBufferfv nor ClearBufferuiv accept a buffer of STENCIL. The command void ClearBufferfi( enum buffer, int drawbuffer, float depth, int stencil ); clears both depth and stencil buffers of the currently bound draw framebuffer. buffer must be DEPTH_STENCIL and drawbuffer must be zero. depth and stencil are the values to clear the depth and stencil buffers to, respectively. Clamping and type conversion of depth for fixed-point depth buffers is performed in the same fashion as ClearDepth. Masking of stencil for stencil buffers is performed in the same fashion as ClearStencil. ClearBufferfi is equivalent to clearing the depth and stencil buffers separately, but may be faster when a buffer of internal format DEPTH_STENCIL is being cleared. OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

403

The result of these commands is undefined if no conversion between the type of the specified value and the type of the buffer being cleared is defined (for example, if ClearBufferiv is called for a fixed- or floating-point buffer, or if ClearBufferfv is called for a signed or unsigned integer buffer). This is not an error. When ClearBuffer* is called, the same per-fragment and masking operations defined for Clear are applied. Errors An INVALID_ENUM error is generated by ClearBufferiv if buffer is not COLOR or STENCIL. An INVALID_ENUM error is generated by ClearBufferuiv if buffer is not COLOR. An INVALID_ENUM error is generated by ClearBufferfv if buffer is not COLOR or DEPTH. An INVALID_ENUM error is generated by ClearBufferfi if buffer is not DEPTH_STENCIL. An INVALID_VALUE error is generated if buffer is COLOR and drawbuffer is negative, or greater than the value of MAX_DRAW_BUFFERS minus one; or if buffer is DEPTH, STENCIL, or DEPTH_STENCIL and drawbuffer is not zero. 15.2.3.2

Clearing the Multisample Buffer

The color samples of the multisample buffer are cleared when one or more color buffers are cleared, as specified by the Clear mask bit COLOR_BUFFER_BIT and the DrawBuffer mode. If the DrawBuffer mode is NONE, the color samples of the multisample buffer cannot be cleared using Clear. If the Clear mask bits DEPTH_BUFFER_BIT or STENCIL_BUFFER_BIT are set, then the corresponding depth or stencil samples, respectively, are cleared. The ClearBuffer* commands also clear color, depth, or stencil samples of multisample buffers corresponding to the specified buffer. Masking and scissoring affect clearing the multisample buffer in the same way as they affect clearing the corresponding color, depth, and stencil buffers.

15.2.4

Invalidating Framebuffer Contents

The GL provides a means for invalidating portions of every pixel or a subregion of pixels in a particular buffer, effectively leaving their contents undefined. The command

OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

404

void InvalidateSubFramebuffer( enum target, sizei numAttachments, const enum *attachments, int x, int y, sizei width, sizei height ); signals the GL that it need not preserve all contents of a bound framebuffer object. target must be FRAMEBUFFER, DRAW_FRAMEBUFFER, or READ_FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. numAttachments indicates how many attachments are supplied in the attachments list. If an attachment is specified that does not exist in the framebuffer bound to target, it is ignored. x and y are the origin (with lower left-hand corner at (0, 0)) and width and height are the width and height, respectively, of the pixel rectangle to be invalidated. Any of these pixels lying outside of the window allocated to the current GL context, or outside of the attachments of the currently bound framebuffer object, are ignored. If a framebuffer object is bound to target, then including DEPTH_STENCIL_ATTACHMENT in the attachments list is a special case causing both the depth and stencil attachments of the framebuffer object to be invalidated. Note that if a specified attachment has base internal format DEPTH_STENCIL but the attachments list does not include DEPTH_STENCIL_ATTACHMENT or both DEPTH_ATTACHMENT and STENCIL_ATTACHMENT, then only the specified portion of every pixel in the subregion of pixels of the DEPTH_STENCIL buffer may be invalidated, and the other portion must be preserved. If the framebuffer object is not complete, InvalidateSubFramebuffer may be ignored. Errors An INVALID_ENUM error is generated if target is not FRAMEBUFFER, DRAW_FRAMEBUFFER, or READ_FRAMEBUFFER. An INVALID_ENUM error is generated if a framebuffer object is bound to target and any elements of attachments are not one of the attachments in table 9.1. An INVALID_ENUM error is generated if attachments contains a value other than COLOR_ATTACHMENTm for m in the range [0, 31]. An INVALID_OPERATION error is generated if attachments contains a value COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_VALUE error is generated if numAttachments, width, or height is negative. An INVALID_ENUM error is generated if the default framebuffer is bound

OpenGL ES 3.2 (May 5, 2022)

15.2. WHOLE FRAMEBUFFER OPERATIONS

405

to target and any elements of attachments are not one of • COLOR, identifying the color buffer • DEPTH, identifying the depth buffer • STENCIL, identifying the stencil buffer. The command void InvalidateFramebuffer( enum target, sizei numAttachments, const enum *attachments ); is equivalent to InvalidateSubFramebuffer(target, numAttachments, attachments, 0, 0, vw, vh); where vw and vh are equal to the maximum viewport width and height, respctively, obtained by querying MAX_VIEWPORT_DIMS (for the default framebuffer) or the largest framebuffer object’s attachments’ width and height, respectively (for a framebuffer object).

OpenGL ES 3.2 (May 5, 2022)

Chapter 16

Reading and Copying Pixels Pixels may be read from the framebuffer using ReadPixels. BlitFramebuffer can be used to copy a block of pixels from one portion of the framebuffer to another.

16.1

Reading Pixels

The method for reading pixels from the framebuffer and placing them in pixel pack buffer or client memory is diagrammed in figure 16.1. We describe the stages of the pixel reading process in the order in which they occur.

16.1.1

Selecting Buffers for Reading

When reading pixels from a color buffer, the buffer selected for reading is termed the read buffer, and is controlled with the command void ReadBuffer( enum src ); If the GL is bound to the default framebuffer (see section 9), src must be BACK or NONE. BACK refers to the back buffer of a double-buffered context or the sole buffer of a single-buffered context. The initial value of the read framebuffer for the default framebuffer is BACK if there is a default framebuffer associated with the context, otherwise it is NONE. If the GL is bound to a read framebuffer object, src must be one of the values listed in table 15.5, including NONE. Specifying COLOR_ATTACHMENTi enables reading from the image attached to the framebuffer at that attachment point. The initial value of the read framebuffer for framebuffer objects is COLOR_ATTACHMENT0. The read buffer of the currently bound read framebuffer can be queried by calling GetIntegerv with pname set to READ_BUFFER. 406

16.1. READING PIXELS

407

RGBA pixel data in

Convert to float

Clamp to [0,1]

Pixel Storage Operations

Pack

byte, short, int, float, or packed pixel component data stream

Figure 16.1. Operation of ReadPixels. Operations in dashed boxes are not performed for all data formats.

Errors An INVALID_ENUM error is generated if src is not BACK or one of the values from table 15.5. An INVALID_OPERATION error is generated if src is BACK and there is no default framebuffer associated with the context. An INVALID_OPERATION error is generated if the GL is bound to the default framebuffer and src is COLOR_ATTACHMENTm for any m. An INVALID_OPERATION error is generated if the GL is bound to a draw framebuffer object and src is BACK or COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS.

16.1.2

ReadPixels

Initially, zero is bound for the PIXEL_PACK_BUFFER, indicating that image read and query commands such as ReadPixels return pixel results into client memory pointer parameters. However, if a non-zero buffer object is bound as the current pixel pack buffer, then the pointer parameter is treated as an offset into the designated buffer object. OpenGL ES 3.2 (May 5, 2022)

16.1. READING PIXELS Parameter Name PACK_ROW_LENGTH PACK_SKIP_ROWS PACK_SKIP_PIXELS PACK_ALIGNMENT

408 Type integer integer integer integer

Initial Value 0 0 0 4

Valid Range [0, ∞) [0, ∞) [0, ∞) 1,2,4,8

Table 16.1: PixelStorei parameters pertaining to ReadPixels.

Pixels are read using void ReadPixels( int x, int y, sizei width, sizei height, enum format, enum type, void *data ); void ReadnPixels( int x, int y, sizei width, sizei height, enum format, enum type, sizei bufSize, void *data ); The arguments after x and y to ReadPixels are described in section 8.4.2. The pixel storage modes that apply to ReadPixels and other commands that query images (see section 8.11) are summarized in table 16.1. Only two combinations of format and type are accepted in most cases. The first varies depending on the format of the currently bound rendering surface. For normalized fixed-point rendering surfaces, the combination format RGBA and type UNSIGNED_BYTE is accepted. For floating-point rendering surfaces, the combination format RGBA and type FLOAT is accepted. For signed integer rendering surfaces, the combination format RGBA_INTEGER and type INT is accepted. For unsigned integer rendering surfaces, the combination format RGBA_INTEGER and type UNSIGNED_INT is accepted. The second is an implementation-chosen format from among those defined in table 8.2, excluding formats DEPTH_COMPONENT, DEPTH_STENCIL, and STENCIL_INDEX. The values of format and type for this format may be determined by calling GetIntegerv with the symbolic constants IMPLEMENTATION_COLOR_READ_FORMAT and IMPLEMENTATION_COLOR_READ_TYPE, respectively. The implementation-chosen format may vary depending on the format of the selected read buffer of the currently bound read framebuffer. Additionally, when the internal format of the rendering surface is RGB10_A2, a third combination of format RGBA and type UNSIGNED_INT_2_10_10_10_REV is accepted.

OpenGL ES 3.2 (May 5, 2022)

16.1. READING PIXELS

409

Errors An INVALID_OPERATION error is generated if the combination of format and type is unsupported. An INVALID_OPERATION error is generated if the read framebuffer is not framebuffer complete. An INVALID_OPERATION error is generated if the value of READ_FRAMEBUFFER_BINDING (see section 9) is non-zero, the read framebuffer is framebuffer complete, and the effective value of SAMPLE_BUFFERS for the read framebuffer is one. An INVALID_VALUE error is generated if width, height, or bufSize is negative. An INVALID_OPERATION error is generated by ReadnPixels if the buffer size required to store the requested data is greater than bufSize. An INVALID_OPERATION error is generated by GetIntegerv if pname is IMPLEMENTATION_COLOR_READ_FORMAT or IMPLEMENTATION_COLOR_READ_TYPE and any of: • the read framebuffer is not framebuffer complete • the read framebuffer is a framebuffer object, and the selected read buffer (see section 16.1.1) has no image attached • the selected read buffer is NONE Additional errors for ReadPixels are described in the following sections.

16.1.3

Obtaining Pixels from the Framebuffer

Values are obtained from the color buffer selected by the read buffer (see section 16.1.1). ReadPixels obtains values from the selected buffer from each pixel with lower left hand corner at (x + i, y + j) for 0 ≤ i < width and 0 ≤ j < height; this pixel is said to be the ith pixel in the jth row. If any of these pixels lies outside of the window allocated to the current GL context, or outside of the image attached to the currently bound read framebuffer object, then the values obtained for those pixels are undefined. When READ_FRAMEBUFFER_BINDING is zero, values are also undefined for individual pixels that are not owned by the current context. Otherwise, ReadPixels obtains values from the selected buffer, regardless of how those values were placed there. If format is one of RED, RG, RGB, or RGBA, then red, green, blue, and alpha OpenGL ES 3.2 (May 5, 2022)

16.1. READING PIXELS

410

values are obtained from the selected buffer at each pixel location. Errors An INVALID_OPERATION error is generated if: • format is an integer format and the color buffer is not an integer format; • format is not an and the color buffer is an integer format; not an integer format; • format is an integer format and type is FLOAT, HALF_FLOAT, or UNSIGNED_INT_10F_11F_11F_REV; or • the color buffer is a floating-point format and type is not FLOAT, HALF_FLOAT, or UNSIGNED_INT_10F_11F_11F_REV. When READ_FRAMEBUFFER_BINDING is non-zero, the red, green, blue, and alpha values are obtained by first reading the internal component values of the corresponding value in the image attached to the selected logical buffer. Internal components are converted to an RGBA color by taking each R, G, B, and A component present according to the base internal format of the buffer (as shown in table 8.8). If G, B, or A values are not present in the internal format, they are taken to be zero, zero, and one respectively.

16.1.4

Conversion of RGBA values

The R, G, B, and A values form a group of elements. For a normalized fixed-point color buffer, each element is converted to floating-point using equation 2.1. For an integer or floating-point color buffer, the elements are unmodified.

16.1.5

Final Conversion

For a floating-point RGBA color, if type is not one of FLOAT, HALF_FLOAT, or UNSIGNED_INT_10F_11F_11F_REV, each component is first clamped to [0, 1]. Then the appropriate conversion table 16.2 is applied to the component. In the special case of calling ReadPixels with type of UNSIGNED_INT_10F_11F_11F_REV and format of RGB, conversion is performed as follows: the returned data are packed into a series of uint values. The red, green, and blue components are converted to unsigned 11-bit floating-point, unsigned 11-bit floating-point, and unsigned 10-bit floating point as described in sections 2.3.4.3 and 2.3.4.4. The resulting red 11 bits, green 11 bits, and blue 10 bits are then packed as the 1st, 2nd, OpenGL ES 3.2 (May 5, 2022)

16.1. READING PIXELS type Parameter

411 GL Data Type

UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT HALF_FLOAT FLOAT UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_5_5_5_1 UNSIGNED_INT_2_10_10_10_REV UNSIGNED_INT_10F_11F_11F_REV

ubyte byte ushort short uint int half float ushort ushort ushort uint uint

Component Conversion Formula Equation 2.3, b = 8 Equation 2.4, b = 8 Equation 2.3, b = 16 Equation 2.4, b = 16 Equation 2.3, b = 32 Equation 2.4, b = 32 c=f c=f Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Special

Table 16.2: Reversed component conversions, used when component data are being returned to client memory. Color components are converted from the internal floating-point representation (f ) to a datum of the specified GL data type (c) using the specified equation. All arithmetic is done in the internal floating point format. These conversions apply to component data returned by GL query commands and to components of pixel data returned to client memory. The equations remain the same even if the implemented ranges of the GL data types are greater than the minimum required ranges. (See table 2.2.)

and 3rd components of the UNSIGNED_INT_10F_11F_11F_REV format as shown in figure 8.4. For an integer RGBA color, each component is clamped to the representable range of type.

16.1.6

Placement in Pixel Pack Buffer or Client Memory

If a pixel pack buffer is bound (as indicated by a non-zero value of PIXEL_PACK_BUFFER_BINDING), data is an offset into the pixel pack buffer and the pixels are packed into the buffer relative to this offset; otherwise, data is a pointer to a block client memory and the pixels are packed into the client memory relative to the pointer.

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

412

Errors An INVALID_OPERATION error is generated if a pixel pack buffer object is bound and packing the pixel data according to the pixel pack storage state would access memory beyond the size of the pixel pack buffer’s memory size. An INVALID_OPERATION error is generated if a pixel pack buffer object is bound and data is not evenly divisible by the number of basic machine units needed to store in memory the corresponding GL data type from table 8.4 for the type parameter. Groups of elements are placed in memory just as they are taken from memory when transferring pixel rectangles to the GL. That is, the ith group of the jth row (corresponding to the ith pixel in the jth row) is placed in memory just where the ith group of the jth row would be taken from when transferring pixels. See Unpacking under section 8.4.2.1. The only difference is that the storage mode parameters whose names begin with PACK_ are used instead of those whose names begin with UNPACK_. If the format is RED, only the corresponding single element is written. Likewise if the format is RG or RGB, only the corresponding two or three elements are written. Otherwise all the elements of each group are written.

16.2

Copying Pixels

Several commands copy pixel data between regions of the framebuffer (see section 16.2.1), or between regions of textures and renderbuffers (see section 16.2.2).

16.2.1

Blitting Pixel Rectangles

To transfer a rectangle of pixel values from one region of a source framebuffer to another region of a destination framebuffer, use the command void BlitFramebuffer( int srcX0, int srcY0, int srcX1, int srcY1, int dstX0, int dstY0, int dstX1, int dstY1, bitfield mask, enum filter ); The source and destination framebuffers are those bound to READ_The source and destination must not be the same framebuffer. If no framebuffer is bound to READ_FRAMEBUFFER or DRAW_FRAMEBUFFER, then the default read or draw framebuffer is used as the corresponding source or destination framebuffer, respectively. FRAMEBUFFER and DRAW_FRAMEBUFFER respectively.

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

413

mask is zero or the bitwise OR of one or more values indicating which buffers are to be copied. The values are COLOR_BUFFER_BIT, DEPTH_BUFFER_BIT, and STENCIL_BUFFER_BIT, which are described in section 15.2.3. The pixels corresponding to these buffers are copied from the source rectangle bounded by the locations (srcX0, srcY 0) and (srcX1, srcY 1) to the destination rectangle bounded by the locations (dstX0, dstY 0) and (dstX1, dstY 1). Pixels have half-integer center coordinates. Only pixels whose centers lie within the destination rectangle are written by BlitFramebuffer. Linear filter sampling (see below) may result in pixels outside the source rectangle being read. If mask is zero, no buffers are copied. When the color buffer is transferred, values are taken from the read buffer of the read framebuffer and written to each of the draw buffers of the draw framebuffer. The actual region taken from the read framebuffer is limited to the intersection of the source buffers being transferred, which may include the color buffer selected by the read buffer, the depth buffer, and/or the stencil buffer depending on mask. The actual region written to the draw framebuffer is limited to the intersection of the destination buffers being written, which may include multiple draw buffers, the depth buffer, and/or the stencil buffer depending on mask. Whether or not the source or destination regions are altered due to these limits, the scaling and offset applied to pixels being transferred is performed as though no such limits were present. If the source and destination rectangle dimensions do not match, the source image is stretched to fit the destination rectangle. filter must be LINEAR or NEAREST, and specifies the method of interpolation to be applied if the image is stretched. LINEAR filtering is allowed only for the color buffer. If the source and destination dimensions are identical, no filtering is applied. If either the source or destination rectangle specifies a negative width or height (X1 < X0 or Y 1 < Y 0), the image is reversed in the corresponding direction. If both the source and destination rectangles specify a negative width or height for the same direction, no reversal is performed. If a linear filter is selected and the rules of LINEAR sampling would require sampling outside the bounds of a source buffer, it is as though CLAMP_TO_EDGE texture sampling were being performed. If a linear filter is selected and sampling would be required outside the bounds of the specified source region, but within the bounds of a source buffer, the implementation may choose to clamp while sampling or not. When values are taken from the read buffer, if the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the read buffer is SRGB (see section 9.2.3), the red, green, and blue components are converted from the non-linear sRGB color space according to equation 8.14. When values are written to the draw buffers, blit operations bypass most of the OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

414

fragment pipeline. The only fragment operations which affect a blit are the pixel ownership test, the scissor test, and sRGB conversion (see section 15.1.6). Color, depth, and stencil masks (see section 15.2.2) are ignored. If the read framebuffer is layered (see section 9.8), pixel values are read from layer zero. If the draw framebuffer is layered, pixel values are written to layer zero. If both read and draw framebuffers are layered, the blit operation is still performed only on layer zero. If a buffer is specified in mask and does not exist in both the read and draw framebuffers, the corresponding bit is silently ignored. If the color formats of the read and draw buffers do not match, and mask includes COLOR_BUFFER_BIT, pixel groups are converted to match the destination format. However, colors are clamped only if all draw color buffers have fixed-point components. Format conversion is not supported for all data types, as described below. If the read framebuffer is multisampled (its effective value of SAMPLE_BUFFERS is one) and the draw framebuffer is not (its value of SAMPLE_BUFFERS is zero), the samples corresponding to each pixel location in the source are converted to a single sample before being written to the destination. The filter parameter is ignored. If the source formats are integer types or stencil values, a single sample’s value is selected for each pixel. If the source formats are floating-point or normalized types, the sample values for each pixel are resolved in an implementationdependent manner. If the source formats are depth values, sample values are resolved in an implementation-dependent manner where the result will be between the minimum and maximum depth values in the pixel. Errors An INVALID_VALUE error is generated if mask contains any bits other than COLOR_BUFFER_BIT, DEPTH_BUFFER_BIT, or STENCIL_BUFFER_BIT. An INVALID_ENUM error is generated if filter is not LINEAR or NEAREST. An INVALID_OPERATION error is generated if mask includes DEPTH_BUFFER_BIT or STENCIL_BUFFER_BIT, and filter is not NEAREST. An INVALID_OPERATION error is generated if the source and destination buffers are identical. An INVALID_FRAMEBUFFER_OPERATION error is generated if either the read framebuffer or the draw framebuffer is not framebuffer complete (section 9.4.2). An INVALID_OPERATION error is generated if mask includes DEPTH_-

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

415

BUFFER_BIT or STENCIL_BUFFER_BIT, and the source and destination

depth and stencil buffer formats do not match. An INVALID_OPERATION error is generated if filter is LINEAR and the read buffer contains integer data. An INVALID_OPERATION error is generated if the read framebuffer is multisampled, and the source and destination rectangles are not defined with the same (X0, Y 0) and (X1, Y 1) bounds. An INVALID_OPERATION error is generated if the read framebuffer is multisampled, and the formats of the read and draw framebuffers are not identical. An INVALID_OPERATION error is generated if the draw framebuffer is multisampled. An INVALID_OPERATION error is generated if format conversions are not supported, which occurs under any of the following conditions: • The read buffer contains fixed-point or floating-point values and any draw buffer contains neither fixed-point nor floating-point values. • The read buffer contains unsigned integer values and any draw buffer does not contain unsigned integer values. • The read buffer contains signed integer values and any draw buffer does not contain signed integer values.

16.2.2

Copying Between Images

The command void CopyImageSubData( uint srcName, enum srcTarget, int srcLevel, int srcX, int srcY, int srcZ, uint dstName, enum dstTarget, int dstLevel, int dstX, int dstY, int dstZ, sizei srcWidth, sizei srcHeight, sizei srcDepth ); may be used to copy a region of texel data between two image objects. An image object may be either a texture or a renderbuffer. CopyImageSubData does not perform general-purpose conversions such as scaling, resizing, blending, color-space, or format conversions. It should be considered to operate in a manner similar to a CPU memcpy. CopyImageSubData can copy between images with different internal formats, provided the formats are compatible. OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

416

CopyImageSubData also allows copying between certain types of compressed and uncompressed internal formats as described in table 16.3. This copy does not perform on-the-fly compression or decompression. When copying from an uncompressed internal format to a compressed internal format, each texel of uncompressed data becomes a single block of compressed data. When copying from a compressed internal format to an uncompressed internal format, a block of compressed data becomes a single texel of uncompressed data. The texel size of the uncompressed format must be the same size as the block size of the compressed formats. Thus it is permitted to copy between a 128-bit uncompressed format and a compressed format which uses 8-bit 4 × 4 blocks, or between a 64-bit uncompressed format and a compressed format which uses 4-bit 4 × 4 blocks. The source object is identified by srcName and srcTarget. Similarly the destination object is identified by dstName and dstTarget. The interpretation of the name depends on the value of the corresponding target parameter. If the target parameter is RENDERBUFFER, the name is interpreted as the name of a renderbuffer object. If the target parameter is a texture target, the name is interpreted as a texture object. All texture targets are accepted, with the exception of TEXTURE_BUFFER and the cubemap face selectors described in table 8.20. srcLevel and dstLevel identify the source and destination level-of-detail. For textures, this must be a valid level-of-detail in the texture object. For renderbuffers, this value must be zero. srcX, srcY, and srcZ specify the lower left texel coordinates of a srcWidth-wide by srcHeight-high by srcDepth-deep rectangular subregion of the source texture image. Similarly, dstX, dstY and dstZ specify the coordinates of a subregion of the destination texture image. The source and destination subregions must be contained entirely within the specified level of the corresponding image objects. The dimensions are always specified in texels, even for compressed texture formats. But it should be noted that if only one of the source and destination textures is compressed then the number of texels touched in the compressed image will be a factor of the block size larger than in the uncompressed image. Slices of a two-dimensional array, cube map array, or three dimensional texture, or faces of a cube map texture are all compatible provided they share a compatible internal format, and multiple slices or faces may be copied between these objects with a single call by specifying the starting slice with srcZ and dstZ, and the number of slices to be copied with srcDepth. Cubemap textures always have six faces which are selected by a zero-based face index, according to the order specified in table 8.20. For the purposes of CopyImageSubData, two internal formats are considered compatible if any of the following conditions are met:

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

417

• the formats are the same, • the formats are both listed in the same entry of table 16.4, or • one format is compressed and the other is uncompressed and table 16.3 lists the two formats in the same row. Texel / Block Size 128-bit

Uncompressed internal format RGBA32UI, RGBA32I, RGBA32F

64-bit

RGBA16F, RG32F, RGBA16UI, RG32UI, RGBA16I, RG32I

Compressed internal format COMPRESSED_RGBA8_ETC2_EAC, COMPRESSED_SRGB8_ALPHA8_ETC2_EAC, COMPRESSED_RG11_EAC, COMPRESSED_SIGNED_RG11_EAC, and all COMPRESSED_RGBA_ASTC* and COMPRESSED_SRGB8_ALPHA8_ASTC*

formats from table 8.17 COMPRESSED_RGB8_ETC2, COMPRESSED_SRGB8_ETC2, COMPRESSED_R11_EAC, COMPRESSED_SIGNED_R11_EAC, COMPRESSED_RGB8_PUNCHTHROUGH_ALPHA1_ETC2, COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2

Table 16.3: Compatible internal formats for copying between compressed and uncompressed internal formats with CopyImageSubData. Formats in the same row can be copied between each other.

Class VIEW_CLASS_128_BITS VIEW_CLASS_96_BITS VIEW_CLASS_64_BITS VIEW_CLASS_48_BITS VIEW_CLASS_32_BITS

Internal formats RGBA32F, RGBA32UI, RGBA32I RGB32F, RGB32UI, RGB32I RGBA16F, RG32F, RGBA16UI, RG32UI, RGBA16I, RG32I, RGB16F, RGB16UI, RGB16I RG16F, R11F_G11F_B10F, R32F, RGB10_A2UI, RGBA8UI, RG16UI, R32UI, RGBA8I, RG16I, R32I, RGB10_A2, RGBA8, RG16, RGBA8_SNORM, SRGB8_ALPHA8, RGB9_E5 (Continued on next page)

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

418

Compatible internal formats for CopyImageSubData (continued) Class Internal formats VIEW_CLASS_24_BITS RGB8, RGB8_SNORM, SRGB8, RGB8UI, RGB8I VIEW_CLASS_16_BITS R16F, RG8UI, R16UI, RG8I, R16I, RG8, RG8_SNORM VIEW_CLASS_8_BITS R8UI, R8I, R8, R8_SNORM VIEW_CLASS_EAC_R11 COMPRESSED_R11_EAC, COMPRESSED_SIGNED_R11_EAC VIEW_CLASS_EAC_RG11 COMPRESSED_RG11_EAC, COMPRESSED_SIGNED_RG11_EAC VIEW_CLASS_ETC2_RGB COMPRESSED_RGB8_ETC2, COMPRESSED_SRGB8_ETC2 VIEW_CLASS_ETC2_RGBA COMPRESSED_RGB8_PUNCHTHROUGH_ALPHA1_ETC2, VIEW_CLASS_ETC2_EAC_RGBA VIEW_CLASS_ASTC_4x4_RGBA VIEW_CLASS_ASTC_5x4_RGBA VIEW_CLASS_ASTC_5x5_RGBA VIEW_CLASS_ASTC_6x5_RGBA VIEW_CLASS_ASTC_6x6_RGBA VIEW_CLASS_ASTC_8x5_RGBA VIEW_CLASS_ASTC_8x6_RGBA VIEW_CLASS_ASTC_8x8_RGBA VIEW_CLASS_ASTC_10x5_RGBA VIEW_CLASS_ASTC_10x6_RGBA VIEW_CLASS_ASTC_10x8_RGBA VIEW_CLASS_ASTC_10x10_RGBA

COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_RGBA8_ETC2_EAC, COMPRESSED_SRGB8_ALPHA8_ETC2_EAC COMPRESSED_RGBA_ASTC_4x4, COMPRESSED_SRGB8_ALPHA8_ASTC_4x4 COMPRESSED_RGBA_ASTC_5x4, COMPRESSED_SRGB8_ALPHA8_ASTC_5x4 COMPRESSED_RGBA_ASTC_5x5, COMPRESSED_SRGB8_ALPHA8_ASTC_5x5 COMPRESSED_RGBA_ASTC_6x5, COMPRESSED_SRGB8_ALPHA8_ASTC_6x5 COMPRESSED_RGBA_ASTC_6x6, COMPRESSED_SRGB8_ALPHA8_ASTC_6x6 COMPRESSED_RGBA_ASTC_8x5, COMPRESSED_SRGB8_ALPHA8_ASTC_8x5 COMPRESSED_RGBA_ASTC_8x6, COMPRESSED_SRGB8_ALPHA8_ASTC_8x6 COMPRESSED_RGBA_ASTC_8x8, COMPRESSED_SRGB8_ALPHA8_ASTC_8x8 COMPRESSED_RGBA_ASTC_10x5, COMPRESSED_SRGB8_ALPHA8_ASTC_10x5 COMPRESSED_RGBA_ASTC_10x6, COMPRESSED_SRGB8_ALPHA8_ASTC_10x6 COMPRESSED_RGBA_ASTC_10x8, COMPRESSED_SRGB8_ALPHA8_ASTC_10x8 COMPRESSED_RGBA_ASTC_10x10, COMPRESSED_SRGB8_ALPHA8_ASTC_10x10

(Continued on next page)

OpenGL ES 3.2 (May 5, 2022)

16.2. COPYING PIXELS

419

Compatible internal formats for CopyImageSubData (continued) Class Internal formats VIEW_CLASS_ASTC_12x10_RGBA COMPRESSED_RGBA_ASTC_12x10, COMPRESSED_SRGB8_VIEW_CLASS_ASTC_12x12_RGBA

ALPHA8_ASTC_12x10 COMPRESSED_RGBA_ASTC_12x12, ALPHA8_ASTC_12x12

COMPRESSED_SRGB8_-

Table 16.4: Compatible internal formats for CopyImageSubData. Formats in the same row may be cast to each other.

If the source and destination buffers are identical, and the source and destination subregions overlap, texel values resulting from the copy operation are undefined. If the internal format does not exactly match the internal format of the original texture, the contents of the memory are reinterpreted in the same manner as for image bindings described in section 8.23. Errors An INVALID_OPERATION error is generated if the texel size of the uncompressed image is not equal to the block size of the compressed image. An INVALID_ENUM error is generated if either target is not RENDERBUFFER or a valid texture target; is TEXTURE_BUFFER or one of the cubemap face selectors described in table 8.20; or if the target does not match the type of the object. An INVALID_OPERATION error is generated if either object is a texture and the texture is not complete (as defined in section 8.17, but ignoring format-based completeness rules), or if the source and destination internal formats are not compatible, or if the number of samples do not match. An INVALID_VALUE error is generated if either name does not correspond to a valid renderbuffer or texture object according to the corresponding target parameter. An INVALID_VALUE error is generated if srcLevel and dstLevel are not valid levels for the corresponding images. An INVALID_VALUE error is generated if srcWidth, srcHeight, or srcDepth is negative. An INVALID_VALUE error is generated if the dimensions of either sub-

OpenGL ES 3.2 (May 5, 2022)

16.3. PIXEL DRAW AND READ STATE

420

region exceeds the boundaries of the corresponding image object, or if the image format is compressed and the dimensions of the subregion fail to meet the alignment constraints of the format. An INVALID_OPERATION error is generated if the formats are not compatible.

16.3

Pixel Draw and Read State

The state required for pixel operations consists of the parameters that are set with PixelStorei. This state has been summarized in table 8.1. Additional state includes an integer indicating the current setting of ReadBuffer. State set with PixelStorei is GL client state.

OpenGL ES 3.2 (May 5, 2022)

Chapter 17

Compute Shaders In addition to graphics-oriented shading operations such as vertex, tessellation, geometry, and fragment shading, generic computation may be performed by the GL through the use of compute shaders. The compute pipeline is a form of singlestage machine that runs generic shaders. Compute shaders are created as described in section 7.1 using a type parameter of COMPUTE_SHADER. They are attached to and used in program objects as described in section 7.3. Compute workloads are formed from groups of work items called workgroups and processed by the executable code for a compute program. A workgroup is a collection of shader invocations that execute the same code, potentially in parallel. An invocation within a workgroup may share data with other members of the same workgroup through shared variables (see section 4.3.7 (“Shared Variables”) of the OpenGL ES Shading Language Specification) and issue memory and control barriers to synchronize with other members of the same workgroup. One or more workgroups is launched by calling: void DispatchCompute( uint num groups x, uint num groups y, uint num groups z ); Each workgroup is processed by the active program object for the compute shader stage. The active program for the compute shader stage will be determined in the same manner as the active program for other pipeline stages, as described in section 7.3. While the individual shader invocations within a workgroup are executed as a unit, workgroups are executed completely independently and in unspecified order. num groups x, num groups y and num groups z specify the number of workgroups that will be dispatched in the X, Y and Z dimensions, respectively. The built-in vector variable gl_NumWorkGroups will be initialized with the contents 421

422 of the num groups x, num groups y and num groups z parameters. The maximum number of workgroups that may be dispatched at one time may be determined by calling GetIntegeri v with target set to MAX_COMPUTE_WORK_GROUP_COUNT and index set to zero, one, or two, representing the X, Y, and Z dimensions respectively. If the workgroup count in any dimension is zero, no workgroups are dispatched. The workgroup size in each dimension is specified at compile time using an input layout qualifier in the compute shader attached to the program (see section 4 (“Compute Shader Inputs”) of the OpenGL ES Shading Language Specification). After the program has been linked, the workgroup size of the program may be queried by calling GetProgramiv with pname COMPUTE_WORK_GROUP_SIZE. This will return an array of three integers containing the workgroup size of the compute program as specified by its input layout qualifier(s). The maximum workgroup size may be determined by calling GetIntegeri v with target set to MAX_COMPUTE_WORK_GROUP_SIZE and index set to 0, 1, or 2 to retrieve the maximum workgroup size in the X, Y and Z dimension, respectively. Furthermore, the maximum number of invocations in a single workgroup (i.e., the product of the three dimensions) may be determined by calling GetIntegerv with pname set to MAX_COMPUTE_WORK_GROUP_INVOCATIONS. Errors An INVALID_OPERATION error is generated if there is no active program object for the compute shader stage. An INVALID_VALUE error is generated if any of num groups x, num groups y and num groups z are greater than the maximum workgroup count for the corresponding dimension. The command void DispatchComputeIndirect( intptr indirect ); is equivalent to calling DispatchCompute with num groups x, num groups y and num groups z initialized with the three uint values contained in the buffer currently bound to the DISPATCH_INDIRECT_BUFFER binding at an offset, in basic machine units, specified in indirect. If any of num groups x, num groups y or num groups z is greater than the value of MAX_COMPUTE_WORK_GROUP_COUNT for the corresponding dimension then the results are undefined. Errors An INVALID_OPERATION error is generated if there is no active program

OpenGL ES 3.2 (May 5, 2022)

17.1. COMPUTE SHADER VARIABLES

423

for the compute shader stage. An INVALID_VALUE error is generated if indirect is negative or is not a multiple of the size, in basic machine units, of uint. An INVALID_OPERATION error is generated if the command would source data beyond the end of the buffer object. An INVALID_OPERATION error is generated if zero is bound to the DISPATCH_INDIRECT_BUFFER binding.

17.1

Compute Shader Variables

Compute shaders can access uniform variables belonging to the current program object. Limits on uniform storage and methods for manipulating uniforms are described in section 7.6. There is a limit to the total size of all variables declared as shared in a single program object. This limit, expressed in units of basic machine units, may be queried as the value of MAX_COMPUTE_SHARED_MEMORY_SIZE. The total size required for a given program object is implementation-dependent. However, this total size may not exceed the largest block size that would be obtained if all active variables declared as shared were instead declared as arbitrarily ordered members of a shader storage block qualified with std430.

OpenGL ES 3.2 (May 5, 2022)

Chapter 18

Debug Output Application developers can obtain details about errors, undefined behavior, implementation-dependent performance warnings, or other useful hints from the GL in the form of debug output. This information is communicated through a stream of debug messages that are generated as GL commands are executed. The application can choose to receive these messages either through a callback routine, or by querying for them from a message log. Controls are provided for disabling messages that the application does not care about, and for inserting application-generated messages into the stream. Different levels of debug output may be provided, depending on how the context was created. If the context is not a debug context1 (e.g. if it was created without the CONTEXT_FLAG_DEBUG_BIT set in the CONTEXT_FLAGS state, as described in section 20.2), then the GL may optionally not generate any debug messages, but the commands described in this chapter will otherwise operate without error. Debug output functionality is enabled or disabled by calling Enable or Disable with target DEBUG_OUTPUT. If the context is a debug context (if it was created with the CONTEXT_FLAG_DEBUG_BIT set in CONTEXT_FLAGS) then the initial value of DEBUG_OUTPUT is TRUE; otherwise the initial value is FALSE. In a debug context, if DEBUG_OUTPUT is disabled the GL will not generate any debug output logs or callbacks. Enabling DEBUG_OUTPUT again will enable full debug output functionality. In a non-debug context, if DEBUG_OUTPUT is later enabled, the level of debug output logging is defined by the GL implementation, which may have zero debug output. 1

Debug contexts are specified at context creation time, using window system binding APIs such as those specified by EGL 1.5.

424

18.1. DEBUG MESSAGES Debug Output Message Source DEBUG_SOURCE_API DEBUG_SOURCE_SHADER_COMPILER DEBUG_SOURCE_WINDOW_SYSTEM DEBUG_SOURCE_THIRD_PARTY DEBUG_SOURCE_APPLICATION DEBUG_SOURCE_OTHER

425 Messages Generated by The GL The GLSL shader compiler or compilers for other extension-provided languages The window system, such as EGL, GLX or WGL External debuggers or third-party middleware libraries The application Sources that do not fit to any of the ones listed above

Table 18.1: Sources of debug output messages. Each message must originate from a source listed in this table.

Full debug output support is guaranteed only in a debug context.

18.1

Debug Messages

A debug message is uniquely identified by the source that generated it, a type within that source, and an unsigned integer ID identifying the message within that type. The message source is one of the symbolic constants listed in table 18.1. The message type is one of the symbolic constants listed in table 18.2. Each message source and type pair contains its own namespace of messages with every message being associated with an ID. The assignment of IDs to messages within a namespace is implementation-dependent. There can potentially be overlap between the namespaces of two different pairs of source and type, so messages can only be uniquely distinguished from each other by the full combination of source, type and ID. Each message is also assigned a severity level that roughly describes its importance across all sources and types along a single global axis. The severity of a message is one of the symbolic constants defined in table 18.3. Because messages can be disabled by their severity, this allows for quick control the global volume of debug output. Every message also has a null-terminated string representation that is used to describe the message. The contents of the string can change slightly between different instances of the same message (e.g. which parameter value caused a specific GL error to occur). The format of a message string is left as implementationOpenGL ES 3.2 (May 5, 2022)

18.1. DEBUG MESSAGES

426

Debug Output Message Type DEBUG_TYPE_ERROR DEBUG_TYPE_DEPRECATED_BEHAVIOR DEBUG_TYPE_UNDEFINED_BEHAVIOR DEBUG_TYPE_PERFORMANCE DEBUG_TYPE_PORTABILITY DEBUG_TYPE_MARKER DEBUG_TYPE_PUSH_GROUP DEBUG_TYPE_POP_GROUP DEBUG_TYPE_OTHER

Informs about Events that generated an error Behavior that has been marked for deprecation Behavior that is undefined according to the specification Implementation-dependent performance warnings Use of extensions or shaders in a way that is highly vendor-specific Annotation of the command stream Entering a debug group Leaving a debug group Types of events that do not fit any of the ones listed above

Table 18.2: Types of debug output messages. Each message is associated with one of these types that describes the nature of the message.

Severity Level Token DEBUG_SEVERITY_HIGH

DEBUG_SEVERITY_MEDIUM

DEBUG_SEVERITY_LOW DEBUG_SEVERITY_NOTIFICATION

Suggested examples of messages Any GL error; dangerous undefined behavior; any shader compiler and linker errors; Severe performance warnings; GLSL or other shader compiler and linker warnings; use of currently deprecated behavior Performance warnings from redundant state changes; trivial undefined behavior Any message which is not an error or performance concern

Table 18.3: Severity levels of messages. Each debug output message is associated with one of these severity levels.

OpenGL ES 3.2 (May 5, 2022)

18.2. DEBUG MESSAGE CALLBACK

427

dependent, although it should at least represent a concise description of the event that caused the message to be generated. Messages with different IDs should also have sufficiently distinguishable string representations to warrant their separation. The lengths of all messages, including their null terminators, must be guaranteed to be less or equal to the value of the implementation-dependent constant MAX_DEBUG_MESSAGE_LENGTH. Messages can be either enabled or disabled. Messages that are disabled will not be generated. All messages are initially enabled unless their assigned severity is DEBUG_SEVERITY_LOW. The enabled state of messages can be changed using the command DebugMessageControl.

18.2

Debug Message Callback

Applications can provide a callback function for receiving debug messages using the command void DebugMessageCallback( DEBUGPROC callback, const void *userParam ); with callback storing the address of the callback function. callback must be a function whose prototype is of the form void callback( enum source, enum type, uint id, enum severity, sizei length, const char *message, const void *userParam ); Additionally, callback must be declared with the same platform-dependent calling convention used in the definition of the type DEBUGPROC. Anything else will result in undefined behavior. Only one debug callback can be specified for the current context, and further calls overwrite the previous callback. Specifying NULL as the value of callback clears the current callback and disables message output through callbacks. Applications can provide user-specified data through the pointer userParam. The context will store this pointer and will include it as one of the parameters in each call to the callback function. If the application has specified a callback function for receiving debug output, the implementation will call that function whenever any enabled message is generated. The source, type, ID, and severity of the message are specified by the DEBUGPROC parameters source, type, id, and severity, respectively. The string representation of the message is stored in message and its length (excluding the OpenGL ES 3.2 (May 5, 2022)

18.3. DEBUG MESSAGE LOG null-terminator) is stored in length. The parameter userParam is the user-specified parameter that was given when calling DebugMessageCallback. Applications that specify a callback function must be aware of certain special conditions when executing code inside a callback when it is called by the GL, regardless of the debug source. The memory for message is owned and managed by the GL, and should only be considered valid for the duration of the function call. The behavior of calling any GL or window system function from within the callback function is undefined and may lead to program termination. Care must also be taken in securing debug callbacks for use with asynchronous debug output by multi-threaded GL implementations. Section 18.8 describes this in further detail. If the DEBUG_OUTPUT state is disabled then the GL will not call the callback function.

18.3

Debug Message Log

If DEBUG_CALLBACK_FUNCTION is NULL, then debug messages are instead stored in an internal message log up to some maximum number of messages as defined by the value of MAX_DEBUG_LOGGED_MESSAGES. Each context stores its own message log and will only store messages generated by commands operating in that context. If the message log fills up, then any subsequently generated messages will not be placed in the log until the message log is cleared, and will instead be discarded. Applications can query the number of messages currently in the log by obtaining the value of DEBUG_LOGGED_MESSAGES, and the string length (including its null terminator) of the oldest message in the log through the value of DEBUG_NEXT_LOGGED_MESSAGE_LENGTH. To fetch message data stored in the log, the command GetDebugMessageLog can be used. If DEBUG_CALLBACK_FUNCTION is not NULL, no generated messages will be stored in the log but will instead be passed to the debug callback routine as described in section 18.2. If the DEBUG_OUTPUT state is disabled then no messages are added to the message log.

OpenGL ES 3.2 (May 5, 2022)

428

18.4. CONTROLLING DEBUG MESSAGES

18.4

429

Controlling Debug Messages

Applications can control the volume of debug output in the active debug group (see section 18.6) by disabling specific groups of messages with the command void DebugMessageControl( enum source, enum type, enum severity, sizei count, const uint *ids, boolean enabled ); If enabled is TRUE, the referenced subset of messages will be enabled. If FALSE, then those messages will be disabled. This command can reference different subsets of messages by first considering the set of all messages, and filtering out messages based on the following ways: • If source, type, or severity is DONT_CARE, then messages from all sources, of all types, or of all severities are referenced respectively. • When values other than DONT_CARE are specified, all messages whose source, type, or severity match the specified source, type, or severity respectively will be referenced. • If count is greater than zero, then ids is an array of count message IDs for the specified combination of source and type. In this case, source and type must not be DONT_CARE, and severity must be DONT_CARE, Unrecognized message IDs in ids are ignored. If count is zero, the value if ids is ignored. Although messages are grouped into an implicit hierarchy by their sources and types, there is no explicit per-source, per-type or per-severity enabled state. Instead, the enabled state is stored individually for each message. There is no difference between disabling all messages from one source in a single call, and individually disabling all messages from that source using their types and IDs. If DEBUG_OUTPUT is disabled, then it is as if messages of every source, type, or severity are disabled. Errors An INVALID_ENUM error is generated if any of source, type, and severity is neither DONT_CARE nor one of the symbols from, respectively, tables 18.1, 18.2, and 18.3. An INVALID_VALUE error is generated if count is negative, An INVALID_OPERATION error is generated if count is greater than zero

OpenGL ES 3.2 (May 5, 2022)

18.5. EXTERNALLY GENERATED MESSAGES

430

and either source or type is DONT_CARE, or severity is not DONT_CARE.

18.5

Externally Generated Messages

To support applications and third-party libraries generating their own messages, such as ones containing timestamp information or signals about specific render system events, the following function can be called void DebugMessageInsert( enum source, enum type, uint id, enum severity, int length, const char *buf ); The value of id specifies the ID for the message and severity indicates its severity level as defined by the caller. The string buf contains the string representation of the message. The parameter length contains the number of characters in buf. If length is negative, it is implied that buf contains a null terminated string. Errors If DEBUG_OUTPUT is disabled, then calls to DebugMessageInsert are discarded, but do not generate an error. An INVALID_ENUM error is generated if type is not one of the values from table 18.2, or if source is not DEBUG_SOURCE_APPLICATION or DEBUG_SOURCE_THIRD_PARTY. An INVALID_ENUM error is generated if severity is not one of the severity levels listed in table 18.3. An INVALID_VALUE error is generated if the number of characters in buf, excluding the null terminator when length is negative, is not less than the value of MAX_DEBUG_MESSAGE_LENGTH.

18.6

Debug Groups

Debug groups provide a method for annotating a command stream with discrete groups of commands using a descriptive text. Debug output messages, either generated by the implementation or inserted by the application with DebugMessageInsert are written to the active debug group (the top of the debug group stack). Debug groups are strictly hierarchical. Their sequences may be nested within other debug groups but can not overlap. If no debug group has been pushed by the application then the active debug group is the default debug group. The command

OpenGL ES 3.2 (May 5, 2022)

18.6. DEBUG GROUPS

431

void PushDebugGroup( enum source, uint id, sizei length, const char *message ); pushes a debug group described by the string message into the command stream. The value of id specifies the ID of messages generated. The parameter length contains the number of characters in message. If length is negative, it is implied that message contains a null terminated string. The message has the specified source and id, type DEBUG_TYPE_PUSH_GROUP, and severity DEBUG_SEVERITY_NOTIFICATION. The GL will put a new debug group on top of the debug group stack which inherits control of the volume of debug output of the debug group previously residing on the top of the debug group stack. Because debug groups are strictly hierarchical, any additional control of the debug output volume will only apply within the active debug group and the debug groups pushed on top of the active debug group. Errors An INVALID_ENUM error is generated if the value of source is neither DEBUG_SOURCE_APPLICATION nor DEBUG_SOURCE_THIRD_PARTY. An INVALID_VALUE error is generated if length is negative and the number of characters in message, excluding the null-terminator, is not less than the value of MAX_DEBUG_MESSAGE_LENGTH. A STACK_OVERFLOW error is generated if PushDebugGroup is called and the stack contains the value of MAX_DEBUG_GROUP_STACK_DEPTH minus one elements. The command void PopDebugGroup( void ); pops the active debug group. After popping a debug group, the GL will also generate a debug output message describing its cause based on the message string, the source, and an id submitted to the associated PushDebugGroup command. DEBUG_TYPE_PUSH_GROUP and DEBUG_TYPE_POP_GROUP share a single namespace for message id. severity has the value DEBUG_SEVERITY_NOTIFICATION and type has the value DEBUG_TYPE_POP_GROUP. Popping a debug group restores the debug output volume control of the parent debug group. Errors A STACK_UNDERFLOW error is generated if PopDebugGroup is called and

OpenGL ES 3.2 (May 5, 2022)

18.7. DEBUG LABELS

432 Identifier

BUFFER FRAMEBUFFER PROGRAM_PIPELINE PROGRAM QUERY RENDERBUFFER SAMPLER SHADER TEXTURE TRANSFORM_FEEDBACK VERTEX_ARRAY

Object Type buffer framebuffer program pipeline program query renderbuffer sampler shader texture transform feedback vertex array

Table 18.4: Object namespace identifiers and the corresponding object types.

only the default debug group is on the stack.

18.7

Debug Labels

Debug labels provide a method for annotating any object (texture, buffer, shader, etc.) with a descriptive text label. These labels may then be used by the debug output (see section 5.5) or an external tool such as a debugger or profiler to describe labelled objects. The command void ObjectLabel( enum identifier, uint name, sizei length, const char *label ); labels the object identified by name and its namespace identifier. identifier must be one of the tokens in table 18.4, indicating the type of the object corresponding to name. label contains a string used to label an object. length contains the number of characters in label. If length is negative, then label contains a null-terminated string. If label is NULL, any debug label is effectively removed from the object. Errors An INVALID_ENUM error is generated if identifier is not one of the object

OpenGL ES 3.2 (May 5, 2022)

18.8. ASYNCHRONOUS AND SYNCHRONOUS DEBUG OUTPUT

433

types listed in table 18.4. An INVALID_VALUE error is generated if name is not the name of a valid object of the type specified by identifier. An INVALID_VALUE error is generated if the number of characters in label, excluding the null terminator when length is negative, is not less than the value of MAX_LABEL_LENGTH. The command void ObjectPtrLabel( void *ptr, sizei length, const char *label ); labels the sync object identified by ptr. length and label match the corresponding arguments of ObjectLabel. Errors An INVALID_VALUE error is generated if ptr is not the name of a sync object. An INVALID_VALUE error is generated if the number of characters in label, excluding the null terminator when length is negative, is not less than the value of MAX_LABEL_LENGTH. A label is part of the state of the object to which it is associated. The initial state of an object’s label is the empty string. Labels need not be unique.

18.8

Asynchronous and Synchronous Debug Output

The behavior of how and when the GL driver is allowed to generate debug messages, and subsequently either call back to the application or place the message in the debug message log, is affected by the state DEBUG_OUTPUT_SYNCHRONOUS. This state can be modified by the Enable and Disable commands. Its initial value is FALSE. When DEBUG_OUTPUT_SYNCHRONOUS is disabled, the driver is optionally allowed to concurrently call the debug callback routine from potentially multiple threads, including threads that the context that generated the message is not currently bound to. The implementation may also call the callback routine asynchronously after the GL command that generated the message has already returned. The application is fully responsible for ensuring thread safety due to debug callbacks under these circumstances. In this situation the userParam value may be

OpenGL ES 3.2 (May 5, 2022)

18.9. DEBUG OUTPUT QUERIES

434

helpful in identifying which application thread’s command originally generated the debug callback. When DEBUG_OUTPUT_SYNCHRONOUS is enabled, the driver guarantees synchronous calls to the callback routine by the context. When synchronous callbacks are enabled, all calls to the callback routine will be made by the thread that owns the current context; all such calls will be made serially by the current context; and each call will be made before the GL command that generated the debug message is allowed to return. When no callback is specified and DEBUG_OUTPUT_SYNCHRONOUS is disabled, the driver can still asynchronously place messages in the debug message log, even after the context thread has returned from the GL function that generated those messages. When DEBUG_OUTPUT_SYNCHRONOUS is enabled, the driver guarantees that all messages are added to the log before the GL function returns. Enabling synchronous debug output greatly simplifies the responsibilities of the application for making its callback functions thread-safe, but may potentially result in drastically reduced driver performance. DEBUG_OUTPUT_SYNCHRONOUS only guarantees intra-context synchronization for the callbacks of messages generated by that context, and does not guarantee synchronization across multiple contexts. If multiple contexts are concurrently used by the application, it is allowed for those contexts to also concurrently call their designated callbacks, and the application is responsible for handling thread safety in that situation even if DEBUG_OUTPUT_SYNCHRONOUS is enabled in all contexts.

18.9

Debug Output Queries

Pointers set with debug output commands are queried with the generic GetPointerv command (see section 20.2). pnames DEBUG_CALLBACK_FUNCTION and DEBUG_CALLBACK_USER_PARAM respectively query the current callback function and the user parameter to that function set with DebugMessageCallback. When no debug callback is set, debug messages are stored in a debug message log as described in section 18.3. Messages may be queried from the log by calling uint GetDebugMessageLog( uint count, sizei bufSize, enum *sources, enum *types, uint *ids, enum *severities, sizei *lengths, char *messageLog ); GetDebugMessageLog fetches a maximum of count messages from the message log, and will return the number of messages successfully fetched.

OpenGL ES 3.2 (May 5, 2022)

18.9. DEBUG OUTPUT QUERIES

435

Messages will be fetched from the log in order of oldest to newest. Those messages that were fetched will be removed from the log. The sources, types, severities, IDs, and string lengths of fetched messages will be stored in the application-provided arrays sources, types, severities, ids, and lengths, respectively. The application is responsible for allocating enough space for each array to hold up to count elements. The string representations of all fetched messages are stored in the messageLog array. If multiple messages are fetched, their strings are concatenated into the same messageLog array and will be separated by single null terminators. The last string in the array will also be null-terminated. The maximum size of messageLog, including the space used by all null terminators, is given by bufSize. If a message’s string, including its null terminator, can not fully fit within the messageLog array’s remaining space, then that message and any subsequent messages will not be fetched and will remain in the log. The string lengths stored in the array lengths include the space for the null terminator of each string. Any or all of the arrays sources, types, ids, severities, lengths and messageLog can also be NULL pointers, which causes attributes for such arrays to be discarded when messages are fetched. However, those messages will still be removed from the log. Thus to simply delete up to count messages from the message log while ignoring their attributes, the application can call GetDebugMessageLog with NULL pointers for all attribute arrays. If the context is not a debug context, then the GL can opt to never add messages to the message log, so that GetDebugMessageLog will always return zero. Errors An INVALID_VALUE error is generated if bufSize is negative and messageLog is not NULL. The command void GetObjectLabel( enum identifier, uint name, sizei bufSize, sizei *length, char *label ); returns in label the string labelling an object. identifier and name specify the namespace and name of the object, and match the corresponding arguments of ObjectLabel (see section 18.7). label will be null-terminated. The actual number of characters written into label, excluding the null terminator, is returned in length. If length is NULL, no length is returned. The maximum number of characters that may be written into label, including the null terminator, is specified by bufSize. If no debug label was OpenGL ES 3.2 (May 5, 2022)

18.9. DEBUG OUTPUT QUERIES

436

specified for the object then label will contain a null-terminated empty string, and zero will be returned in length. If label is NULL and length is non-NULL then no string will be returned and the length of the label will be returned in length. Errors An INVALID_ENUM error is generated is identifier is not one of the object types listed in table 18.4 other than SYNC An INVALID_VALUE error is generated if name is not the name of a valid object of the type specified by identifier. An INVALID_VALUE error is generated if bufSize is negative. The command void GetObjectPtrLabel( void *ptr, sizei bufSize, size *length, char *label ); returns in label the string labelling the sync object identified by ptr. bufSize, length, and label match the corresponding arguments of GetObjectLabel. Errors An INVALID_VALUE error is generated if ptr is not the name of a sync object. An INVALID_VALUE error is generated if bufSize is negative.

OpenGL ES 3.2 (May 5, 2022)

Chapter 19

Special Functions This chapter describes additional functionality that does not fit easily into any of the preceding chapters, including hints influencing GL behavior (see section 19.1).

19.1

Hints

Certain aspects of GL behavior, when there is room for variation, may be controlled with hints. A hint is specified using void Hint( enum target, enum hint ); target is a symbolic constant indicating the behavior to be controlled, and hint is a symbolic constant indicating what type of behavior is desired. The possible targets are described in table 19.1. For each target, hint must be one of FASTEST, indicating that the most efficient option should be chosen; NICEST, indicating that the highest quality option should be chosen; and DONT_CARE, indicating no preference in the matter. Target GENERATE_MIPMAP_HINT FRAGMENT_SHADER_DERIVATIVE_HINT

Hint description Quality and performance of automatic mipmap level generation Derivative accuracy for fragment processing built-in functions dFdx, dFdy and fwidth

Table 19.1: Hint targets and descriptions.

437

19.1. HINTS

438

The interpretation of hints is implementation-dependent. An implementation may ignore them entirely. The initial value of all hints is DONT_CARE. Errors An INVALID_ENUM error is generated if target is not one of the values in table 19.1. An INVALID_ENUM error is generated if hint is not FASTEST, NICEST, or DONT_CARE.

OpenGL ES 3.2 (May 5, 2022)

Chapter 20

Context State Queries The state required to describe the GL machine is enumerated in chapter 21, and is set using commands described in previous chapters. State that is part of GL objects can usually be queried using commands described together with the commands to set that state. Such commands operate either directly on a named object, or indirectly through a binding in the GL context (such as a currently bound framebuffer object). The commands in this chapter describe queries for state directly associated with the context, rather than with an object. Data conversions may be done when querying context state, as described in section 2.2.2.

20.1

Simple Queries

Much of the GL state is completely identified by symbolic constants. The values of these state variables can be obtained using a set of Get* commands. Valid values of the symbolic constants allowed as parameter names to the various queries in this section are not summarized here, because there are many allowed parameters. Instead they are described elsewhere in the Specification together with the commands such state is relevant to, as well as in the state tables in chapter 21. There are four commands for obtaining simple state variables: void GetBooleanv( enum pname, boolean *data ); void GetIntegerv( enum pname, int *data ); void GetInteger64v( enum pname, int64 *data ); void GetFloatv( enum pname, float *data );

439

20.1. SIMPLE QUERIES

440

The commands obtain boolean, integer, 64-bit integer, or floating-point state variables. pname is a symbolic constant indicating the state variable to return. data is a pointer to a scalar or array of the indicated type in which to place the returned data. Errors An INVALID_ENUM error is generated if pname is not state queriable with these commands. Indexed simple state variables are queried with the commands void GetBooleani v( enum target, uint index, boolean *data ); void GetIntegeri v( enum target, uint index, int *data ); void GetInteger64i v( enum target, uint index, int64 *data ); target is the name of the indexed state and index is the index of the particular element being queried. data is a pointer to a scalar or array of the indicated type in which to place the returned data. Errors An INVALID_ENUM error is generated if target is not indexed state queriable with these commands. An INVALID_VALUE error is generated if index is outside the valid range for the indexed state target. State which is controlled with Enable and Disable is queried with the command boolean IsEnabled( enum cap ); TRUE or FALSE is returned if cap is currently enabled or disabled, respectively.

Errors An INVALID_ENUM error is generated if cap is not enable state queriable with IsEnabled. boolean IsEnabledi( enum target, uint index ); OpenGL ES 3.2 (May 5, 2022)

20.2. POINTER, STRING, AND RELATED CONTEXT QUERIES

441

can be used to determine if the indexed state corresponding to target and index is enabled or disabled. Errors An INVALID_ENUM error is generated if target is not indexed enable state queriable with IsEnabledi. An INVALID_VALUE error is generated if index is outside the valid range for the indexed state target.

20.2

Pointer, String, and Related Context Queries

Pointers in the current GL context are queried with the command void GetPointerv( enum pname, void **params ); pname is a symbolic constant indicating the pointer to return. params is a pointer to a variable in which to place the single returned pointer value. pnames of DEBUG_CALLBACK_FUNCTION and DEBUG_CALLBACK_USER_PARAM, return debug output state as described in section 18.9. Errors An INVALID_ENUM error is generated if pname is not DEBUG_CALLBACK_FUNCTION or DEBUG_CALLBACK_USER_PARAM. String queries return pointers to UTF-8 encoded, null-terminated static strings describing properties of the current GL context1 . The command ubyte *GetString( enum name ); accepts name values of RENDERER, VENDOR, EXTENSIONS, VERSION, and SHADING_LANGUAGE_VERSION. The format of the RENDERER and VENDOR strings is implementation-dependent. The EXTENSIONS string contains a space separated list of extension names (the extension names themselves do not contain any spaces). The VERSION string is laid out as follows: 1 Applications making copies of these static strings should never use a fixed-length buffer, because the strings may grow unpredictably between releases, resulting in buffer overflow when copying. This is particularly true of the EXTENSIONS string, which has become extremely long in some GL implementations.

OpenGL ES 3.2 (May 5, 2022)

20.2. POINTER, STRING, AND RELATED CONTEXT QUERIES

442

"OpenGL ES N.M vendor-specific information" The SHADING_LANGUAGE_VERSION string is laid out as follows: "OpenGL ES GLSL ES N.M vendor-specific information" The version number is either of the form major number.minor number or major number.minor number.release number, where the numbers all have one or more digits. The minor number for SHADING_LANGUAGE_VERSION is always two digits, matching the OpenGL ES Shading Language Specification release number. For example, this query might return the string "3.10" while the corresponding VERSION query returns "3.1". The release number and vendor specific information are optional. However, if present, then they pertain to the server and their format and contents are implementation-dependent. GetString returns the version number (in the VERSION string) and the extension names (in the EXTENSIONS string) that can be supported by the current GL context. Thus, if the client and server support different versions and/or extensions, a compatible version and list of extensions is returned. Errors An INVALID_ENUM error is generated if name is not RENDERER, VENDOR, EXTENSIONS, VERSION, or SHADING_LANGUAGE_VERSION. The context version may also be queried by calling GetIntegerv with pname MAJOR_VERSION and MINOR_VERSION, which respectively return the same values as major number and minor number in the VERSION string.

Flags defining additional properties of the context may be queried by calling GetIntegerv with pname CONTEXT_FLAGS. If CONTEXT_FLAG_DEBUG_BIT is set in CONTEXT_FLAGS, then the context is a debug context, enabling full support for debug output as described in chapter 18. If CONTEXT_FLAG_ROBUST_ACCESS_BIT is set in CONTEXT_FLAGS, then robust buffer access will be enabled for drawing commands using vertex arrays, as described in section 10.3.5. Indexed strings are queried with the command ubyte *GetStringi( enum name, uint index ); name is the name of the indexed state and index is the index of the particular element being queried. OpenGL ES 3.2 (May 5, 2022)

20.3. INTERNAL FORMAT QUERIES Target TEXTURE_2D_MULTISAMPLE TEXTURE_2D_MULTISAMPLE_ARRAY RENDERBUFFER

443 Usage 2D multisample texture 2D multisample array texture renderbuffer

Table 20.1: Possible targets that internalformat can be used with and the corresponding usage meaning.

If name is EXTENSIONS, the extension name corresponding to the indexth supported extension will be returned. index may range from zero to the value of NUM_EXTENSIONS minus one. All extension names, and only the extension names returned in GetString(EXTENSIONS) will be returned as individual names, but there is no defined relationship between the order in which names appear in the non-indexed string and the order in which they appear in the indexed query. There is no defined relationship between any particular extension name and the index values; an extension name may correspond to a different index in different GL contexts and/or implementations. Errors An INVALID_ENUM error is generated if name is not EXTENSIONS. An INVALID_VALUE error is generated if index is outside the valid range for the indexed state name.

20.3

Internal Format Queries

Information about implementation-dependent support for internal formats can be queried with the command void GetInternalformativ( enum target, enum internalformat, enum pname, sizei count, int *params ); internalformat must be a sized internal format that is color-, depth- or stencilrenderable, as defined in section 9.4. target indicates the usage of the internalformat, and must be one of the targets listed in table 20.1. No more than count integers will be written into params. If more data are available, they will be ignored and no error will be generated.

OpenGL ES 3.2 (May 5, 2022)

20.3. INTERNAL FORMAT QUERIES

444

pname indicates the information to query. The following subsection lists the valid values for pname and defines their meaning and the values that may be returned.

20.3.1

Internal Format Query Parameters

Supported values for pname, their meanings, and their possible return values include: • NUM_SAMPLE_COUNTS: The number of sample counts that would be returned by querying SAMPLES is returned in params. – If target does not support multiple samples (is not TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY or RENDERBUFFER), zero is returned. – If internalformat is RGBA16F, R32F, RG32F, or RGBA32F, zero may be returned. • SAMPLES: The sample counts supported for internalformat and target are written into params, in descending numeric order. Only positive values are returned. – Note that querying SAMPLES with a count of one will will return just the maximum supported number of samples for this format. – The maximum value in SAMPLES is guaranteed to be at least the lowest of the following: * The value of MAX_INTEGER_SAMPLES, if internalformat is a signed or unsigned integer format. * The value of MAX_DEPTH_TEXTURE_SAMPLES, if internalformat is a depth/stencil-renderable format and target is TEXTURE_2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY. * The value of MAX_COLOR_TEXTURE_SAMPLES, if internalformat is a color-renderable format and target is TEXTURE_2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY. * A value less than or equal to the value of MAX_SAMPLES, if internalformat is RGBA16F, R32F, RG32F, or RGBA32F. * The value of MAX_SAMPLES, otherwise.

OpenGL ES 3.2 (May 5, 2022)

20.3. INTERNAL FORMAT QUERIES

445

Errors An INVALID_ENUM error is generated if target is not one of the targets in table 20.1, or if pname is not SAMPLES or NUM_SAMPLES_COUNTS. An INVALID_ENUM error is generated if internalformat is not a sized internal format that is color-, depth- or stencil-renderable. An INVALID_VALUE error is generated if count is negative.

OpenGL ES 3.2 (May 5, 2022)

Chapter 21

State Tables The tables on the following pages indicate which state variables are obtained with what commands. State variables that can be obtained using any of GetBooleanv, GetIntegerv, GetInteger64v, or GetFloatv are listed with just one of these commands – the one that is most appropriate given the type of the data to be returned. These state variables cannot be obtained using IsEnabled. However, state variables for which IsEnabled is listed as the query command can also be obtained using GetBooleanv, GetIntegerv, GetInteger64v, and GetFloatv. State variables for which any other command is listed as the query command can be obtained by using that command or any of its typed variants, although information may be lost when not using the listed command. Unless otherwise specified, when floatingpoint state is returned as integer values or integer state is returned as floating-point values it is converted in the fashion described in section 2.2.2. State table entries indicate a type for each variable. Table 21.1 explains these types. The type actually identifies all state associated with the indicated description; in certain cases only a portion of this state is returned. This is the case with textures, where only the selected texture or texture parameter is returned. The abbreviations max, min, and no. are used interchangeably with maximum, minimum, and number, respectively, to help fit tables without overflowing pages.

446

447

Type code B BM U C E Z Z+ Zk , Zk∗ R R+ R[a,b] Rk S I Y n × type

Explanation Boolean Basic machine units Color (floating-point R, G, B, and A values) Enumerated value (as described in spec body) Integer Non-negative integer or enumerated token value k-valued integer (k∗ indicates k is minimum) Floating-point number Non-negative floating-point number Floating-point number in the range [a, b] k-tuple of floating-point numbers null-terminated string Image Pointer (data type unspecified) n copies of type type (n∗ indicates n is minimum) Table 21.1: State Variable Types

OpenGL ES 3.2 (May 5, 2022)

Table 21.2: Current Values and Associated Data

OpenGL ES 3.2 (May 5, 2022)

Get value PATCH VERTICES

Type Z+

Get Command GetIntegerv

Initial Value 3

Description No. of vertices in input patch

Sec. 10.1.12

448

VERTEX ATTRIB ARRAY TYPE

Type 16 ∗ ×B 16 ∗ ×Z5 16 ∗ ×Z + 16 ∗ ×E

Get Command GetVertexAttribiv GetVertexAttribiv GetVertexAttribiv GetVertexAttribiv

Initial Value FALSE 4 0 FLOAT

VERTEX ATTRIB ARRAY NORMALIZED

16 ∗ ×B

GetVertexAttribiv

FALSE

VERTEX ATTRIB ARRAY INTEGER

16 ∗ ×B

GetVertexAttribiv

FALSE

VERTEX ATTRIB ARRAY DIVISOR

16 ∗ ×Z +

GetVertexAttribiv

0

VERTEX ATTRIB ARRAY POINTER

16 ∗ ×Y

GetVertexAttribPointerv

NULL

ELEMENT ARRAY BUFFER BINDING

Z+

GetIntegerv

0

Get value VERTEX ATTRIB ARRAY ENABLED VERTEX ATTRIB ARRAY SIZE VERTEX ATTRIB ARRAY STRIDE

† The ith attribute defaults to a value of i.

Table 21.3: Vertex Array Object State

OpenGL ES 3.2 (May 5, 2022)

16 ∗ ×Z +

GetVertexAttribiv

0

VERTEX ATTRIB BINDING

16 × Z16∗

GetVertexAttribiv

i†

VERTEX ATTRIB RELATIVE OFFSET

16 × Z +

GetVertexAttribiv

0

VERTEX BINDING OFFSET

16 × Z

GetInteger64i v

0

VERTEX BINDING STRIDE

16 × Z

GetIntegeri v

16

VERTEX BINDING DIVISOR

16 × Z +

GetIntegeri v

0

VERTEX BINDING BUFFER

16 × Z +

GetIntegeri v

0

S

GetObjectLabel

empty

–

Sec. 10.3 10.3 10.3 10.3 10.3 10.3 10.5 10.3 10.3.8 6 10.3 10.3

10.3

10.3 10.3 10.3 18.9

449

VERTEX ATTRIB ARRAY BUFFER BINDING

Description Vertex attrib array enable Vertex attrib array size Vertex attrib array stride Vertex attrib array type Vertex attrib array normalized Vertex attrib array has unconverted integers Vertex attrib array instance divisor Vertex attrib array pointer Element array buffer binding Attribute array buffer binding Vertex buffer binding used by vertex attrib i Byte offset added to vertex binding offset for this attribute Byte offset of the first element in data store of the buffer bound to vertex binding i Stride between elements in vertex binding i Instance divisor used for vertex binding i Name of buffer bound to vertex binding i Debug label

Table 21.4: Vertex Array Data (not in vertex array objects)

OpenGL ES 3.2 (May 5, 2022)

ARRAY BUFFER BINDING

Type Z+

Get Command GetIntegerv

Initial Value 0

DRAW INDIRECT BUFFER BINDING

Z+

GetIntegerv

0

VERTEX ARRAY BINDING

Z+

GetIntegerv

0

PRIMITIVE RESTART FIXED INDEX

B

IsEnabled

FALSE

Get value

Description Current buffer binding Indirect command buffer binding Current vertex array object binding Primitive restart with fixed index enable

Sec. 6 10.3.9 10.4 10.3

450

Table 21.5: Buffer Object State

BUFFER SIZE BUFFER USAGE BUFFER ACCESS FLAGS BUFFER MAPPED BUFFER MAP POINTER BUFFER MAP OFFSET BUFFER MAP LENGTH –

Type n × Z+ n×E n × Z+ n×B n×Y n × Z+ n × Z+ S

Get Command GetBufferParameteri64v GetBufferParameteriv GetBufferParameteriv GetBufferParameteriv GetBufferPointerv GetBufferParameteri64v GetBufferParameteri64v GetObjectLabel

Initial Value 0 STATIC_DRAW 0 FALSE NULL 0 0 empty

Description Buffer data size † Buffer usage pattern Extended buffer access flag Buffer map flag Mapped buffer pointer Start of mapped buffer range Size of mapped buffer range Debug label

Sec. 6 6 6 6 6 6 6 18.9

451

† This state may be queried with GetBufferParameteriv, in which case values greater than or equal to 231 will be clamped to 231 − 1.

OpenGL ES 3.2 (May 5, 2022)

Get value

Table 21.6: Transformation State

OpenGL ES 3.2 (May 5, 2022)

Get value VIEWPORT DEPTH RANGE TRANSFORM FEEDBACK BINDING

TRANSFORM FEEDBACK BUFFER BINDING

PRIMITIVE BOUNDING BOX

Type 4×Z 2 × R+

Get Command GetIntegerv GetFloatv

Initial Value see 12.6.1 0,1

Z+

GetIntegerv

0

Z+

GetIntegerv

0

8×R

GetFloatv

(−1, −1, −1, 1, 1, 1, 1, 1)

Description Viewport origin & extent Depth range near & far Object bound for transform feedback operations Buffer object bound to generic bind point for transform feedback Default primitive bounding box

Sec. 12.6.1 12.6.1 12.2 6.6 13.2

452

Table 21.7: Rasterization

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Initial Value

B

IsEnabled

FALSE

CULL FACE MODE

R+ B E

GetFloatv IsEnabled GetIntegerv

1.0 FALSE BACK

FRONT FACE

E

GetIntegerv

CCW

POLYGON OFFSET FACTOR

R R B

GetFloatv GetFloatv IsEnabled

0 0 FALSE

Get value RASTERIZER DISCARD LINE WIDTH CULL FACE

POLYGON OFFSET UNITS POLYGON OFFSET FILL

Description Discard primitives before rasterization Line width Polygon culling enabled Cull front-/back-facing polygons Polygon frontface CW/CCW indicator Polygon offset factor Polygon offset units Polygon offset enable

Sec. 13.1 13.6 13.7.1 13.7.1 13.7.1 13.7.2 13.7.2 13.7.2

453

Table 21.8: Multisampling

† n is the value of MAX_SAMPLE_MASK_WORDS.

OpenGL ES 3.2 (May 5, 2022)

SAMPLE SHADING

Type B B R+ B B

Get Command IsEnabled IsEnabled GetFloatv GetBooleanv IsEnabled

Initial Value FALSE FALSE 1 FALSE FALSE

MIN SAMPLE SHADING VALUE

R+

GetFloatv

0

B n × Z+ †

IsEnabled GetIntegeri v

FALSE All bits of all words set

Get value SAMPLE ALPHA TO COVERAGE SAMPLE COVERAGE SAMPLE COVERAGE VALUE SAMPLE COVERAGE INVERT

SAMPLE MASK SAMPLE MASK VALUE

Description Modify coverage from alpha Mask to modify coverage Coverage mask value Invert coverage mask value Sample shading enable Fraction of multisamples to use for sample shading Additional sample mask Additional sample mask value

Sec. 15.1.1 13.8.3 13.8.3 13.8.3 13.8.3 13.4.1 13.8.3 13.8.3

454

Table 21.9: Textures (selector, state per texture unit)

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Initial Value

E

GetIntegerv

TEXTURE0

96 ∗ ×2 × Z +

GetIntegerv

0

TEXTURE BINDING 2D ARRAY

96 ∗ ×Z +

GetIntegerv

0

TEXTURE BINDING BUFFER

96 ∗ ×Z +

GetIntegerv

0

TEXTURE BINDING CUBE MAP

96 ∗ ×Z +

GetIntegerv

0

TEXTURE BINDING CUBE MAP ARRAY

96 ∗ ×Z +

GetIntegerv

0

TEXTURE BINDING 2D MULTISAMPLE

96 ∗ ×Z +

GetIntegerv

0

96 ∗ ×Z +

GetIntegerv

0

96 ∗ ×Z +

GetIntegerv

0

Get value ACTIVE TEXTURE

TEXTURE BINDING xD

TEXTURE BINDING 2D MULTISAMPLE ARRAY

SAMPLER BINDING

Description Active texture unit selector Texture object bound to TEXTURE_xD Texture object bound to TEXTURE_2D_ARRAY Texture object bound to TEXTURE_BUFFER Texture object bound to TEXTURE_CUBE_MAP Texture object bound to TEXTURE_CUBE_MAP_ARRAY Texture object bound to TEXTURE_2D_MULTISAMPLE Texture object bound to TEXTURE_2D_MULTISAMPLE_ARRAY Sampler object bound to active texture unit

Sec. 10.2 8.1 8.1 8.1 8.1 8.1

8.1

8.1

8.2

455

Get Command GetTexParameter

Initial Value RED

TEXTURE SWIZZLE G

E

GetTexParameter

GREEN

TEXTURE SWIZZLE B

E

GetTexParameter

BLUE

TEXTURE SWIZZLE A

E

GetTexParameter

ALPHA

TEXTURE BORDER COLOR

TEXTURE WRAP S

C E E E

GetTexParameterfv GetTexParameter GetTexParameter GetTexParameter

0.0,0.0,0.0,0.0 see sec. 8.19 LINEAR see sec. 8.19

TEXTURE WRAP T

E

GetTexParameter

see sec. 8.19

TEXTURE WRAP R

E

GetTexParameter

see sec. 8.19

TEXTURE MIN LOD

TEXTURE MAX LEVEL

R R Z+ Z+

GetTexParameterfv GetTexParameterfv GetTexParameterfv GetTexParameterfv

-1000 1000 0 1000

DEPTH STENCIL TEXTURE MODE

E

GetTexParameteriv

DEPTH_COMPONENT

TEXTURE COMPARE MODE TEXTURE COMPARE FUNC

E E

GetTexParameteriv GetTexParameteriv

NONE LEQUAL

TEXTURE IMMUTABLE FORMAT

B

GetTexParameter

FALSE

TEXTURE IMMUTABLE LEVELS

Z+

GetTexParameter

0

–

S

GetObjectLabel

empty

Table 21.10: Textures (state per texture object)

OpenGL ES 3.2 (May 5, 2022)

TEXTURE MIN FILTER TEXTURE MAG FILTER

TEXTURE MAX LOD TEXTURE BASE LEVEL

Description Red component swizzle Green component swizzle Blue component swizzle Alpha component swizzle Border color Minification function Magnification function Texcoord s wrap mode Texcoord t wrap mode (2D, 3D, cube map textures only) Texcoord r wrap mode (3D textures only) Min. level of detail Max. level of detail Base texture array Max. texture array level Depth stencil texturemode Comparison mode Comparison function Size and format immutable No. of levels in immutable textures Debug label

Sec. 8.10 8.10 8.10 8.10 8 8.14 8.15 8.14.2 8.14.2 8.14.2 8 8 8 8 8.16 8.20 8.20 8.18 8.18 18.9

456

TEXTURE SWIZZLE R

Type E

Get value

Get Command GetTexLevelParameter GetTexLevelParameter GetTexLevelParameter GetTexLevelParameter

Initial Value 0 0 0 0

TEXTURE FIXED SAMPLE LOCATIONS

B

GetTexLevelParameter

TRUE

TEXTURE INTERNAL FORMAT

E

GetTexLevelParameteriv

RGBA or R8

6 × Z+

GetTexLevelParameter

0

TEXTURE SHARED SIZE

Z+

GetTexLevelParameter

0

TEXTURE x TYPE

E

GetTexLevelParameter

NONE

TEXTURE COMPRESSED

B

GetTexLevelParameter

FALSE

TEXTURE BUFFER DATA STORE BIND-

Z+

GetTexLevelParameteriv

0

TEXTURE BUFFER OFFSET

n×Z

GetTexLevelParameteriv

0

TEXTURE BUFFER SIZE

n×Z

GetTexLevelParameteriv

0

TEXTURE WIDTH TEXTURE HEIGHT TEXTURE DEPTH

OpenGL ES 3.2 (May 5, 2022)

Table 21.11: Textures (state per texture image)

TEXTURE SAMPLES

TEXTURE x SIZE

ING

Description Specified width Specified height (2D/3D) Specified depth (3D) No. of samples per texel Whether the image uses a fixed sample pattern Internal format (see section 8.19) Component resolution (x is RED, GREEN, BLUE, ALPHA, DEPTH, or STENCIL) Shared exponent field resolution Component type (x is RED, GREEN, BLUE, ALPHA, or DEPTH) True if image has a compressed internal format Buffer object bound as the data store for the active image unit’s buffer texture Offset into buffer’s data store used for the active image unit’s buffer texture Size of the buffer’s data store used for the active image unit’s buffer texture

Sec. 8 8 8 8.8 8.8 8

8

8 8.11 8.7

8.9

8.9

457

Type Z+ Z+ Z+ Z+

Get value

8.9

Table 21.12: Textures (state per sampler object)

OpenGL ES 3.2 (May 5, 2022)

TEXTURE WRAP S

Type C E E E

Get Command GetSamplerParameterfv GetSamplerParameter GetSamplerParameter GetSamplerParameter

Initial Value 0.0,0.0,0.0,0.0 NEAREST_MIPMAP_LINEAR LINEAR REPEAT

TEXTURE WRAP T

E

GetSamplerParameter

REPEAT

TEXTURE WRAP R

E

GetSamplerParameter

REPEAT

TEXTURE MIN LOD

R R E E S

GetSamplerParameterfv GetSamplerParameterfv GetSamplerParameteriv GetSamplerParameteriv GetObjectLabel

-1000 1000 NONE LEQUAL empty

Get value TEXTURE BORDER COLOR TEXTURE MIN FILTER TEXTURE MAG FILTER

TEXTURE MAX LOD TEXTURE COMPARE MODE TEXTURE COMPARE FUNC –

Description Border color Minification function Magnification function Texcoord s wrap mode Texcoord t wrap mode (2D, 3D, cube map textures only) Texcoord r wrap mode (3D textures only) Min. level of detail Max. level of detail Comparison mode Comparison function Debug label

Sec. 8 8.14 8.15 8.14.2 8.14.2 8.14.2 8 8 8.20 8.20 18.9

458

Get Command IsEnabled GetIntegerv IsEnabled GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv

Initial Value FALSE see 13.8.2 FALSE ALWAYS see 15.1.2 0 KEEP KEEP

STENCIL PASS DEPTH PASS

E

GetIntegerv

KEEP

BLEND

E Z+ Z+ E E E B E 4 ∗ ×B

GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv IsEnabled GetIntegerv IsEnabledi

ALWAYS see 15.1.2 0 KEEP KEEP KEEP FALSE LESS FALSE

BLEND SRC RGB

4 ∗ ×E

GetIntegeri v

ONE

BLEND SRC ALPHA

4 ∗ ×E

GetIntegeri v

ONE

BLEND DST RGB

4 ∗ ×E

GetIntegeri v

ZERO

BLEND DST ALPHA

4 ∗ ×E

GetIntegeri v

ZERO

BLEND EQUATION RGB

4 ∗ ×E

GetIntegeri v

FUNC_ADD

BLEND EQUATION ALPHA

4 ∗ ×E

GetIntegeri v

FUNC_ADD

C B

GetFloatv IsEnabled

0.0,0.0,0.0,0.0 TRUE

SCISSOR TEST SCISSOR BOX STENCIL TEST STENCIL FUNC STENCIL VALUE MASK STENCIL REF STENCIL FAIL

Table 21.13: Pixel Operations

OpenGL ES 3.2 (May 5, 2022)

STENCIL BACK FUNC STENCIL BACK VALUE MASK STENCIL BACK REF STENCIL BACK FAIL STENCIL BACK PASS DEPTH FAIL STENCIL BACK PASS DEPTH PASS DEPTH TEST DEPTH FUNC

BLEND COLOR DITHER

Description Scissoring enabled Scissor box Stenciling enabled Front stencil function Front stencil mask Front stencil reference value Front stencil fail action Front stencil depth buffer fail action Front stencil depth buffer pass action Back stencil function Back stencil mask Back stencil reference value Back stencil fail action Back stencil depth buffer fail action Back stencil depth buffer pass action Depth test enabled Depth test function Blending enabled for draw buffer i Blending source RGB function for draw buffer i Blending source A function for draw buffer i Blending dest. RGB function for draw buffer i Blending dest. A function for draw buffer i RGB blending equation for draw buffer i Alpha blending equation for draw buffer i Constant blend color Dithering enabled

Sec. 13.8.2 13.8.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.2 15.1.3 15.1.3 15.1.5 15.1.5 15.1.5 15.1.5 15.1.5 15.1.5 15.1.5 15.1.5 15.1.7

459

STENCIL PASS DEPTH FAIL

Type B 4×Z B E Z+ Z+ E E

Get value

Type

Get Command

Initial Value

COLOR WRITEMASK

4 ∗ ×4 × B

GetBooleani v

(TRUE,TRUE,TRUE,TRUE)

DEPTH WRITEMASK

B

GetBooleanv

TRUE

STENCIL WRITEMASK

Z+

GetIntegerv

1’s

STENCIL BACK WRITEMASK

Z+

GetIntegerv

1’s

COLOR CLEAR VALUE

C

GetFloatv

0.0,0.0,0.0,0.0

DEPTH CLEAR VALUE

R+

GetFloatv

1

STENCIL CLEAR VALUE

Z+

GetIntegerv

0

DRAW FRAMEBUFFER BINDING

Z+

GetIntegerv

0

READ FRAMEBUFFER BINDING

Z+

GetIntegerv

0

RENDERBUFFER BINDING

Z

GetIntegerv

0

Get value

Table 21.14: Framebuffer Control

OpenGL ES 3.2 (May 5, 2022)

Sec. 15.2.2 15.2.2 15.2.2 15.2.2 15.2.3 15.2.3 15.2.3

9.2

9.2

9.2.4

460

Description Color write enables (R,G,B,A) for draw buffer i Depth buffer enabled for writing Front stencil buffer writemask Back stencil buffer writemask Color buffer clear value Depth buffer clear value Stencil clear value Framebuffer object bound to DRAW_FRAMEBUFFER Framebuffer object bound to READ_FRAMEBUFFER Renderbuffer object bound to RENDERBUFFER

OpenGL ES 3.2 (May 5, 2022)

† This state is queried from the currently bound read framebuffer.

Table 21.15: Framebuffer (state per framebuffer object)

Type

Get Command

Initial Value

DRAW BUFFERi

4 ∗ ×E

GetIntegerv

see 15.2.1

READ BUFFER

E

GetIntegerv

see 16.1.1

FRAMEBUFFER DEFAULT WIDTH

Z

+

GetFramebufferParameteriv

0

FRAMEBUFFER DEFAULT HEIGHT

Z+

GetFramebufferParameteriv

0

FRAMEBUFFER DEFAULT LAYERS

Z+

GetFramebufferParameteriv

0

FRAMEBUFFER DEFAULT SAMPLES

Z+

GetFramebufferParameteriv

0

FRAMEBUFFER DEFAULT FIXED SAMPLE LOCATIONS

B

GetFramebufferParameteriv

FALSE

Get value

Description Draw buffer selected for color output i Read source buffer † Default width of framebuffer w/o attachments Default height of framebuffer w/o attachments Default layer count of framebuffer w/o attachments Default sample count of framebuffer w/o attachments Default sample location pattern of framebuffer w/o attachments

Sec. 15.2.1 16.1.1 9.2 9.2 9.2.1 9.2 9.2

461

Get value

Type

Table 21.16: Framebuffer (state per attachment point)

OpenGL ES 3.2 (May 5, 2022)

FRAMEBUFFER ATTACHMENT OBJECT TYPE

E

FRAMEBUFFER ATTACHMENT OBJECT NAME

Z+

FRAMEBUFFER ATTACHMENT TEXTURE LEVEL

Z+

FRAMEBUFFER ATTACHMENT TEXTURE CUBE MAP FACE

Z+

FRAMEBUFFER ATTACHMENT TEXTURE LAYER

Z

FRAMEBUFFER ATTACHMENT LAYERED

B

FRAMEBUFFER ATTACHMENT COLOR ENCODING

E

FRAMEBUFFER ATTACHMENT COMPONENT TYPE

E

Get Command GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv

Z+

–

S

GetObjectLabel

NONE

0

0

NONE

0

Description Type of image attached to framebuffer attachment point Name of object attached to framebuffer attachment point Mipmap level of texture image attached, if object attached is texture Cubemap face of texture image attached, if object attached is cubemap texture Layer of texture image attached, if object attached is 3D texture

Sec. 9.2.2

9.2.2

9.2.8

9.2.8

9.2.8

FALSE

Framebuffer attachment is layered

9.8

-

Encoding of components in the attached image

9.2.3

-

Data type of components in the attached image

9.2.3

-

empty

Size in bits of attached image’s x component; x is RED, GREEN, BLUE, ALPHA, DEPTH, or STENCIL Debug label

9.2.3

18.9

462

FRAMEBUFFER ATTACHMENT x SIZE

GetFramebufferAttachmentParameteriv

Initial Value

Table 21.17: Renderbuffer (state per renderbuffer object)

OpenGL ES 3.2 (May 5, 2022)

RENDERBUFFER INTERNAL FORMAT

Type Z+ Z+ E

Get Command GetRenderbufferParameteriv GetRenderbufferParameteriv GetRenderbufferParameteriv

Initial Value 0 0 RGBA4

RENDERBUFFER RED SIZE

Z+

GetRenderbufferParameteriv

0

RENDERBUFFER GREEN SIZE

Z+

GetRenderbufferParameteriv

0

RENDERBUFFER BLUE SIZE

Z+

GetRenderbufferParameteriv

0

RENDERBUFFER ALPHA SIZE

Z+

GetRenderbufferParameteriv

0

RENDERBUFFER DEPTH SIZE

Z+

GetRenderbufferParameteriv

0

RENDERBUFFER STENCIL SIZE

Z+

GetRenderbufferParameteriv

0

+

GetRenderbufferParameteriv GetObjectLabel

0 empty

Get value RENDERBUFFER WIDTH RENDERBUFFER HEIGHT

RENDERBUFFER SAMPLES –

Z S

Description Width of renderbuffer Height of renderbuffer Internal format of renderbuffer Size in bits of renderbuffer image’s red component Size in bits of renderbuffer image’s green component Size in bits of renderbuffer image’s blue component Size in bits of renderbuffer image’s alpha component Size in bits of renderbuffer image’s depth component Size in bits of renderbuffer image’s stencil component No. of samples Debug label

Sec. 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 9.2.4 18.9

463

Table 21.18: Pixels

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

Initial Value

UNPACK IMAGE HEIGHT

Z+

GetIntegerv

0

+

GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv

0 0 0 0 4 0 0 0 4 0 0

UNPACK SKIP IMAGES UNPACK ROW LENGTH UNPACK SKIP ROWS UNPACK SKIP PIXELS UNPACK ALIGNMENT PACK ROW LENGTH PACK SKIP ROWS PACK SKIP PIXELS PACK ALIGNMENT PIXEL PACK BUFFER BINDING PIXEL UNPACK BUFFER BINDING

Z Z+ Z+ Z+ Z+ Z+ Z+ Z+ Z+ Z+ Z+

Description Value of UNPACK_IMAGE_HEIGHT Value of UNPACK_SKIP_IMAGES Value of UNPACK_ROW_LENGTH Value of UNPACK_SKIP_ROWS Value of UNPACK_SKIP_PIXELS Value of UNPACK_ALIGNMENT Value of PACK_ROW_LENGTH Value of PACK_SKIP_ROWS Value of PACK_SKIP_PIXELS Value of PACK_ALIGNMENT Pixel pack buffer binding Pixel unpack buffer binding

Sec. 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 16.1 16.1 16.1 16.1 16.1 6.6

464

Table 21.19: Shader Object State

OpenGL ES 3.2 (May 5, 2022)

Get value SHADER TYPE DELETE STATUS COMPILE STATUS

-– INFO LOG LENGTH

-– SHADER SOURCE LENGTH –

Type E B B S Z+ S Z+ S

Get Command GetShaderiv GetShaderiv GetShaderiv GetShaderInfoLog GetShaderiv GetShaderSource GetShaderiv GetObjectLabel

Initial Value – FALSE FALSE empty string 0 empty string 0 empty

Description Type of shader (see table 7.1) Shader flagged for deletion Last compile succeeded Info log for shader objects Length of info log Source code for a shader Length of source code Debug label

Sec. 7.1 7.1 7.1 7.12 7.12 7.1 7.12 18.9

465

Get value

Type

Get Command

Initial Value

Table 21.20: Program Pipeline Object State

OpenGL ES 3.2 (May 5, 2022)

ACTIVE PROGRAM

Z+

GetProgramPipelineiv

0

VERTEX SHADER

Z+

GetProgramPipelineiv

0

GEOMETRY SHADER

Z+

GetProgramPipelineiv

0

TESS CONTROL SHADER

Z+

GetProgramPipelineiv

0

TESS EVALUATION SHADER

Z+

GetProgramPipelineiv

0

FRAGMENT SHADER

Z+

GetProgramPipelineiv

0

COMPUTE SHADER

Z+

GetProgramPipelineiv

0

VALIDATE STATUS

B

GetProgramPipelineiv

FALSE

–

S

GetProgramPipelineInfoLog

empty

Z+ S

GetProgramPipelineiv GetObjectLabel

0 empty

INFO LOG LENGTH –

Description The program object that Uniform* commands update when PPO bound Name of current vertex shader program object Name of current geometry shader program object Name of current TCS program object Name of current TES program object Name of current fragment shader program object Name of current compute shader program object Validate status of program pipeline object Info log for program pipeline object Length of info log Debug label

Sec. 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.12 7.12 18.9

466

Get value

Type

Get Command

Initial Value

Table 21.21: Program Object State

OpenGL ES 3.2 (May 5, 2022)

CURRENT PROGRAM

Z+

GetIntegerv

0

PROGRAM PIPELINE BINDING

Z+

GetIntegerv

0

PROGRAM SEPARABLE

B

GetProgramiv

FALSE

DELETE STATUS

B

GetProgramiv

FALSE

LINK STATUS

B

GetProgramiv

FALSE

VALIDATE STATUS

B

GetProgramiv

FALSE

ATTACHED SHADERS

Z+

GetProgramiv

0

-–

0 ∗ ×Z +

GetAttachedShaders

empty

-–

S

GetProgramInfoLog

empty

INFO LOG LENGTH

Z+

GetProgramiv

0

PROGRAM BINARY LENGTH

Z+

GetProgramiv

0

PROGRAM BINARY RETRIEVABLE HINT

B

GetProgramiv

FALSE

-– COMPUTE WORK GROUP SIZE

GetProgramBinary

–

3 × Z+

GetProgramiv

{0, . . . }

S

GetObjectLabel

empty

Sec. 7.3 7.4 7.4 7.3 7.3 7.3 7.12 7.12 7.12 7.6 7.5 7.5 7.5 17 18.9

467

–

0 ∗ ×BM U

Description Name of current program object Current program pipeline object binding Program object can be bound for separate pipeline stages Program object deleted Last link attempt succeeded Last validate attempt succeeded No. of attached shader objects Shader objects attached Info log for program object Length of info log Length of program binary Retrievable binary hint enabled Binary representation of program Local work size of a linked compute program Debug label

Table 21.22: Program Object State (cont.)

OpenGL ES 3.2 (May 5, 2022)

Get value ACTIVE UNIFORMS

Get Command GetProgramiv

Type Z+

-–

0 ∗ ×Z

-– -– -–

+

ACTIVE UNIFORM MAX LENGTH

-– ACTIVE ATTRIBUTES

0 ∗ ×Z 0 ∗ ×Z + 0 ∗ ×char

Initial Value 0

GetUniformLocation

–

GetActiveUniform GetActiveUniform GetActiveUniform

– – empty

Z+

GetProgramiv

0

− Z+

GetUniform GetProgramiv

0 0

Description No. of active uniforms Location of active uniforms Size of active uniform Type of active uniform Name of active uniform Max. active uniform name length Uniform value No. of active attributes

Sec. 7.6 7.12 7.6 7.6 7.6 7.12 7.6 11.1.1

468

Get Command

-–

0 ∗ ×Z

GetAttribLocation

–

-– -– -–

0 ∗ ×Z + 0 ∗ ×Z + 0 ∗ ×char

GetActiveAttrib GetActiveAttrib GetActiveAttrib

– – empty

Z+

GetProgramiv

0

+

Z E E

GetProgramiv GetProgramiv GetProgramiv

0 TRIANGLES TRIANGLE_STRIP

Z+

GetProgramiv

1

E

GetProgramiv

INTERLEAVED_ATTRIBS

Z+

GetProgramiv

0

Z+

GetProgramiv

0

-–

Z+

-–

Z+

-–

0+ × char

GetTransformFeedbackVarying GetTransformFeedbackVarying GetTransformFeedbackVarying

OpenGL ES 3.2 (May 5, 2022)

Table 21.23: Program Object State (cont.)

ACTIVE ATTRIBUTE MAX LENGTH GEOMETRY VERTICES OUT GEOMETRY INPUT TYPE GEOMETRY OUTPUT TYPE GEOMETRY SHADER INVOCATIONS TRANSFORM FEEDBACK BUFFER MODE TRANSFORM FEEDBACK VARYINGS TRANSFORM FEEDBACK VARYING MAX LENGTH

Initial Value

– – –

Description Location of active generic attribute Size of active attribute Type of active attribute Name of active attribute Max. active attribute name length Max. no. of output vertices Primitive input type Primitive output type No. of times a geom. shader should be executed for each input primitive Transform feedback mode for the program No. of outputs to stream to buffer object(s) Max. transform feedback output variable name length Size of each transform feedback output variable Type of each transform feedback output variable Name of each transform feedback output variable

Sec. 11.1.1 11.1.1 11.1.1 11.1.1 7.12 11.3.4 11.3.1 11.3.2 11.3.4.2 7.12 7.12 7.12 11.1.2.1 11.1.2.1 11.1.2.1

469

Type

Get value

Table 21.24: Program Object State (cont.)

OpenGL ES 3.2 (May 5, 2022)

Get value ACTIVE UNIFORM BLOCKS ACTIVE UNIFORM BLOCK MAX NAME LENGTH UNIFORM TYPE UNIFORM SIZE UNIFORM NAME LENGTH UNIFORM BLOCK INDEX UNIFORM OFFSET

Type

Get Command

Z+

GetProgramiv

0

Z+

GetProgramiv

0

GetActiveUniformsiv GetActiveUniformsiv GetActiveUniformsiv GetActiveUniformsiv GetActiveUniformsiv

– – – – –

0 ∗ ×E 0 ∗ ×Z + 0 ∗ ×Z + 0 ∗ ×Z 0 ∗ ×Z

Initial Value

Description No. of active uniform blocks in a program Length of longest active uniform block name Type of active uniform Size of active uniform Uniform name length Uniform block index Uniform buffer offset

Sec. 7.6.2 7.6.2 7.6.2 7.6.2 7.6.2 7.6.2 7.6.2

470

Get Command

Initial Value

Table 21.25: Program Object State (cont.)

OpenGL ES 3.2 (May 5, 2022)

Type

UNIFORM ARRAY STRIDE

0 ∗ ×Z

GetActiveUniformsiv –

UNIFORM MATRIX STRIDE

0 ∗ ×Z

GetActiveUniformsiv –

UNIFORM IS ROW MAJOR

0 ∗ ×B

GetActiveUniformsiv –

UNIFORM BLOCK BINDING

Z+

GetActiveUniformBlockiv

0

UNIFORM BLOCK DATA SIZE

Z+

GetActiveUniformBlockiv

–

UNIFORM BLOCK NAME LENGTH

Z+

GetActiveUniformBlockiv

–

Z+

GetActiveUniformBlockiv

–

n × Z+

GetActiveUniformBlockiv

–

B

GetActiveUniformBlockiv

0

B

GetActiveUniformBlockiv

0

UNIFORM BLOCK ACTIVE UNIFORMS

UNIFORM BLOCK ACTIVE UNIFORM INDICES

UNIFORM BLOCK REFERENCED BY VERTEX SHADER

UNIFORM BLOCK REFERENCED BY FRAGMENT SHADER

Description Uniform buffer array stride Uniform buffer intramatrix stride Whether uniform is a row-major matrix Uniform buffer binding points associated with the specified uniform block Size of the storage needed to hold this uniform block’s data Uniform block name length Count of active uniforms in the specified uniform block Array of active uniform indices of the specified uniform block True if uniform block is actively referenced by the vertex stage True if uniform block is actively referenced by the fragment stage

Sec. 7.6.2 7.6.2 7.6.2

7.6.2

7.6.2 7.6.2 7.6.2

7.6.2

7.6.2

7.6.2

471

Get value

OpenGL ES 3.2 (May 5, 2022)

Table 21.26: Program Object State (cont.)

Get value

Type

Get Command

Initial Value

TESS CONTROL OUTPUT VERTICES

Z

+

GetProgramiv

0

TESS GEN MODE

E

GetProgramiv

QUADS

TESS GEN SPACING

E

GetProgramiv

EQUAL

TESS GEN VERTEX ORDER

E

GetProgramiv

CCW

TESS GEN POINT MODE

B

GetProgramiv

FALSE

ACTIVE ATOMIC COUNTER -

Z+

GetProgramiv

0

BUFFERS

Description Output patch size for tess. control shader Base primitive type for tess. prim. generator Spacing of tess. prim. generator edge subdivision Order of vertices in primitives generated by tess. primitive generator Tess. prim. generator emits points? No. of active atomic counter buffers (AACBs) used by a program

Sec. 11.2.1 11.2.2 11.2.2

11.2.2 11.2.2 7.7

472

Table 21.27: Program Interface State

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

ACTIVE RESOURCES

n × Z+

MAX NAME LENGTH

n × Z+

MAX NUM ACTIVE VARIABLES

n × Z+

Get Command GetProgramInterfaceiv GetProgramInterfaceiv GetProgramInterfaceiv

Initial Value 0 0 0

Description No. of active resources on an interface Max. name length for active resources Max. no. of active variables for active resources

Sec. 7.3.1 7.3.1 7.3.1

473

Get value

Type

Table 21.28: Program Object Resource State

OpenGL ES 3.2 (May 5, 2022)

ACTIVE VARIABLES

Z+

ARRAY SIZE

Z+

ARRAY STRIDE

Z+

ATOMIC COUNTER BUFFER INDEX

Z+

BLOCK INDEX

Z+

BUFFER BINDING

Z+

BUFFER DATA SIZE

Z+

IS ROW MAJOR

Z+

LOCATION

Z+

MATRIX STRIDE

Z+

Get Command GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv

Initial Value -

Description List of active variables owned by active resource Active resource array size Active resource array stride in memory Index of atomic counter buffer owning resource Index of interface block owning resource Buffer binding assigned to active resource Min. buffer data size required for resource Active resource stored as a row major matrix? Location assigned to active resource Active resource matrix stride in memory

Sec. 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1

474

Get value

Type

Table 21.29: Program Object Resource State (cont.)

OpenGL ES 3.2 (May 5, 2022)

NAME LENGTH

Z+

NUM ACTIVE VARIABLES

Z+

OFFSET

Z+

REFERENCED BY VERTEX SHADER

Z+

REFERENCED BY TESS CONTROL SHADER REFERENCED BY TESS EVALUATION SHADER REFERENCED BY GEOMETRY SHADER REFERENCED BY FRAGMENT SHADER REFERENCED BY COMPUTE SHADER

Z+ Z+ Z+ Z+ Z+

Get Command GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv

Initial Value -

Z+

GetProgramResourceiv

-

TOP LEVEL ARRAY STRIDE

Z+

GetProgramResourceiv

-

TYPE

Z+

GetProgramResourceiv

-

Active resource data type

Sec. 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1 7.3.1

7.3.1 7.3.1

475

TOP LEVEL ARRAY SIZE

Description Length of active resource name No. of active variables owned by active resource Active resource offset in memory Active resource used by vertex shader? Active resource used by tess. control shader? Active resource used by tess. evaluation shader? Active resource used by geometry shader? Active resource used by fragment shader? Active resource used by compute shader? Array size of top level shd. storage block member Array stride of top level shd. storage block member

Table 21.30: Vertex Shader State (not part of program objects)

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

Initial Value

CURRENT VERTEX ATTRIB

16 ∗ ×R4

GetVertexAttribfv

0.0,0.0,0.0,1.0

Description Current generic vertex attribute values

Sec. 10.2

476

Table 21.31: Query Object State

OpenGL ES 3.2 (May 5, 2022)

Get value QUERY RESULT QUERY RESULT AVAILABLE –

Type Z+ B S

Get Command GetQueryObjectuiv GetQueryObjectuiv GetObjectLabel

Initial Value 0 or FALSE TRUE empty

Description Query object result Is the query object result available? Debug label

Sec. 4.2.1 4.2.1 18.9

477

Table 21.32: Atomic Counter Buffer Binding State

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

ATOMIC COUNTER BUFFER BINDING

Z+

GetIntegerv

0

ATOMIC COUNTER BUFFER BINDING

n × Z+

GetIntegeri v

0

ATOMIC COUNTER BUFFER START

n × Z+

GetInteger64i v

0

ATOMIC COUNTER BUFFER SIZE

n × Z+

GetInteger64i v

0

Initial Value

Description Current value of generic atomic counter buffer buffer Buffer object bound to each atomic counter buffer binding point Start offset of binding range for each atomic counter buffer Size of binding range for each atomic counter buffer

Sec. 7.7

7.7

7.7 7.7

478

Table 21.33: Image State (state per image unit)

OpenGL ES 3.2 (May 5, 2022)

IMAGE BINDING LEVEL

Type 8 ∗ ×Z + 8 ∗ ×Z +

Get Command GetIntegeri v GetIntegeri v

Initial Value 0 0

IMAGE BINDING LAYERED

8 ∗ ×B

GetBooleani v

FALSE

IMAGE BINDING LAYER

8 ∗ ×Z +

GetIntegeri v

0

IMAGE BINDING ACCESS

8 ∗ ×E

GetIntegeri v

READ_ONLY

IMAGE BINDING FORMAT

8 ∗ ×Z +

GetIntegeri v

R32UI

Get value IMAGE BINDING NAME

Description Name of bound texture object Level of bound texture object Texture object bound with multiple layers Layer of bound texture, if not layered Read and/or write access for bound texture Format used for accesses to bound texture

Sec. 8.23 8.23 8.23 8.23 8.23 8.23

479

Table 21.34: Shader Storage Buffer Binding State

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

SHADER STORAGE BUFFER BINDING

Z+

GetIntegerv

0

SHADER STORAGE BUFFER BINDING

n × Z+

GetIntegeri v

0

SHADER STORAGE BUFFER START

n × Z+

GetInteger64i v

0

SHADER STORAGE BUFFER SIZE

n × Z+

GetInteger64i v

0

Initial Value

Description Current value of generic shader storage buffer binding Buffer object bound to each shader storage buffer binding point Start offset of binding range for each shader storage buffer Size of binding range for each shader storage buffer

Sec. 7.8

7.8

7.8 7.8

480

Table 21.35: Transform Feedback State

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

Initial Value

TRANSFORM FEEDBACK BUFFER BINDING

n × Z+

GetIntegeri v

0

TRANSFORM FEEDBACK BUFFER START

n × Z+

GetInteger64i v

0

TRANSFORM FEEDBACK BUFFER SIZE

n × Z+

GetInteger64i v

0

TRANSFORM FEEDBACK PAUSED

B

GetBooleanv

FALSE

TRANSFORM FEEDBACK ACTIVE

B

GetBooleanv

FALSE

–

S

GetObjectLabel

empty

Description Buffer object bound to each transform feedback attribute stream Start offset of binding range for each transform feedback attrib. stream Size of binding range for each transform feedback attrib. stream Is transform feedback paused on this object? Is transform feedback active on this object? Debug label

Sec. 6.6

6.6

6.6 6.6 6.6 18.9

481

Table 21.36: Uniform Buffer Binding State

OpenGL ES 3.2 (May 5, 2022)

Get value

Type

Get Command

Initial Value

UNIFORM BUFFER BINDING

Z+

GetIntegerv

0

UNIFORM BUFFER BINDING

n × Z+

GetIntegeri v

0

UNIFORM BUFFER START

n × Z+

GetInteger64i v

0

UNIFORM BUFFER SIZE

n × Z+

GetInteger64i v

0

Description Uniform buffer object bound to the context for buffer object manipulation Uniform buffer object bound to the specified context binding point Start of bound uniform buffer region Size of bound uniform buffer region

Sec. 7.6.2

7.6.2 6.6 6.6

482

Table 21.37: Sync (state per sync object)

OpenGL ES 3.2 (May 5, 2022)

Get value OBJECT TYPE SYNC STATUS SYNC CONDITION SYNC FLAGS –

Type E E E Z S

Get Command GetSynciv GetSynciv GetSynciv GetSynciv GetObjectPtrLabel

Initial Value SYNC_FENCE UNSIGNALED SYNC_GPU_COMMANDS_COMPLETE 0 empty

Description Type of sync object Sync object status Sync object condition Sync object flags Debug label

Sec. 4.1 4.1 4.1 4.1 18.9

483

Table 21.38: Hints

OpenGL ES 3.2 (May 5, 2022)

Get value GENERATE MIPMAP HINT FRAGMENT SHADER DERIVATIVE HINT

Type E

Get Command GetIntegerv

Initial Value DONT_CARE

E

GetIntegerv

DONT_CARE

Description Mipmap generation hint Fragment shader derivative accuracy hint

Sec. 19.1 19.1

484

Table 21.39: Compute Dispatch State

OpenGL ES 3.2 (May 5, 2022)

Get value DISPATCH INDIRECT BUFFER BINDING

Type Z+

Get Command GetIntegerv

Initial Value 0

Description Indirect dispatch buffer binding

Sec. 17

485

Get Command

Minimum Value

Z+

GetIntegerv

4

MAX ARRAY TEXTURE LAYERS

Z+ Z+ Z+ Z+

GetInteger64v GetIntegerv GetIntegerv GetIntegerv

224 − 1 256 2048 256

MAX TEXTURE LOD BIAS

R+

GetFloatv

2.0

MAX CUBE MAP TEXTURE SIZE

Z+

GetIntegerv

2048

MAX RENDERBUFFER SIZE

Z+

SUBPIXEL BITS MAX ELEMENT INDEX

Table 21.40: Implementation Dependent Values

MAX 3D TEXTURE SIZE MAX TEXTURE SIZE

GetIntegerv

2048

+

ALIASED LINE WIDTH RANGE

2×R 2 × R+

GetFloatv GetFloatv

1,1 1,1

MULTISAMPLE LINE WIDTH RANGE

2 × R+

GetFloatv

1,1

MAX FRAMEBUFFER HEIGHT

R+ Z+ Z+ Z+

GetFloatv GetIntegerv GetIntegerv GetIntegerv

– 4 2048 † 2048 †

MAX FRAMEBUFFER LAYERS

Z+

GetIntegerv

256

MAX FRAMEBUFFER SAMPLES

Z+

GetIntegerv

4†

MAX COLOR ATTACHMENTS

Z+

GetIntegerv

4

ALIASED POINT SIZE RANGE

MULTISAMPLE LINE WIDTH GRANULARITY MAX DRAW BUFFERS MAX FRAMEBUFFER WIDTH

Description No. of bits of subpixel precision in screen xw and yw Max. element index Max. 3D texture image dimension Max. 2D texture image dimension Max. no. of layers for texture arrays Max. absolute texture level of detail bias Max. cube map texture image dimension Max. width and height of renderbuffers Range (lo to hi) of point sizes Range (lo to hi) of line widths Range (lo to hi) of multisampled line widths Multisampled line width granularity Max. no. of active draw buffers Max. width for framebuffer object Max. height for framebuffer object Max. layer count for layered framebuffer object Max. sample count for framebuffer object Max. no. of FBO attachment points for color buffers

Sec. 13 10.5 8.5 8.5 8.5 8.14 8.5 9.2.4 13.5 13.6 13.6.4 13.6.4 15.2.1 9.2 9.2 9.2.1 9.2 9.2.7

486

† These limits are tied to the values of MAX_TEXTURE_SIZE (for width/height) and MAX_SAMPLES (for samples) respectively.

OpenGL ES 3.2 (May 5, 2022)

Type

Get value

Table 21.41: Implementation Dependent Values (cont.)

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

MIN FRAGMENT INTERPOLATION OFFSET

R

GetFloatv

-0.5

MAX FRAGMENT INTERPOLATION OFFSET

R

GetFloatv

+0.5 - 1 ULP‡

FRAGMENT INTERPOLATION OFFSET BITS

Z+

GetIntegerv

4

2 × Z+

GetIntegerv

see 12.6.1

MAX SAMPLES

Z+

GetIntegerv

4

MAX SAMPLE MASK WORDS

Z+

GetIntegerv

1

MAX COLOR TEXTURE SAMPLES

Z+

GetIntegerv

1

MAX DEPTH TEXTURE SAMPLES

Z+

GetIntegerv

1

MAX INTEGER SAMPLES

Z+

GetIntegerv

1

MAX SERVER WAIT TIMEOUT

Z+

GetInteger64v

0

LAYER PROVOKING VERTEX

E

GetIntegerv

See sec. 11.3.4

PRIMITIVE RESTART FOR PATCHES SUPPORTED

B

GetBooleanv

–

MAX VIEWPORT DIMS

Description Furthest negative offset for interpolateAtOffset Furthest positive offset for interpolateAtOffset Subpixel bits for interpolateAtOffset Max. viewport dimensions Max. no. of samples supported for multisampling for all internal formats required to support multisampled rendering. Max. no. of sample mask words Max. no. of samples in a color multisample texture† Max. no. of samples in a depth/stencil multisample texture† Max. no. of samples in integer format multisample buffers† Max. WaitSync timeout interval Vertex convention followed by gl_Layer Primitive restart support for PATCHES

Sec. 14.1

14.1

14.1 12.6.1

9.2.4

13.8.3 13.8.3

13.8.3

9.2.4 4.1.1 11.3.4 10.3.4

487

† These queries return the maximum no. of samples for all internal formats required to support multisampled rendering. ‡“ULP” means “Units in the Last Place” or “Unit of Least Precision”.

Get value

Table 21.42: Implementation Dependent Values (cont.)

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

MAX VERTEX ATTRIB RELATIVE OFFSET

Z

GetIntegerv

2047

MAX VERTEX ATTRIB BINDINGS

Z

GetIntegerv

16

MAX VERTEX ATTRIB STRIDE

Z

GetIntegerv

2048

MAX ELEMENTS INDICES

Z+

GetIntegerv

–

MAX ELEMENTS VERTICES

Z+

GetIntegerv

–

MAX TEXTURE BUFFER SIZE

Z+

GetIntegerv

65536

NUM COMPRESSED TEXTURE FORMATS

Z+

GetIntegerv

10

COMPRESSED TEXTURE FORMATS

10 ∗ ×Z +

GetIntegerv

–

NUM PROGRAM BINARY FORMATS

Z+

GetIntegerv

0

0 ∗ ×Z +

GetIntegerv

–

Z+

GetIntegerv

0

0 ∗ ×Z +

GetIntegerv

–

SHADER COMPILER

B

GetBooleanv

–

TEXTURE BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

PROGRAM BINARY FORMATS

NUM SHADER BINARY FORMATS

SHADER BINARY FORMATS

-–

2 × 6 × 2 × Z+

-–

2 × 6 × Z+

GetShaderPrecisionFormat GetShaderPrecisionFormat

Description Max. offset added to vertex buffer binding offset Max. no. of vertex buffers Max. vertex attribute stride Recommended max. no. of DrawRangeElements indices Recommended max. no. of DrawRangeElements vertices No. of addressable texels for buffer textures No. of compressed texture formats Enumerated compressed texture formats No. of program binary formats Enumerated program binary formats No. of shader binary formats Enumerated shader binary formats Shader compiler supported, always TRUE Min. required alignment for texture buffer offsets

Sec. 10.3 10.3 10.3 10.3

10.3 8.9 8.7 8.7 7.5 7.5 7.2 7.2 11.1 8.9

–

Shader data type ranges

7.12

–

Shader data type precisions

7.12

488

† The value of TEXTURE_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum.

Get value

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

0 ∗ ×S

GetStringi

–

NUM EXTENSIONS

Z+

GetIntegerv

–

MAJOR VERSION

Z+

GetIntegerv

3

MINOR VERSION

Z+

GetIntegerv

–

RENDERER

Z+ S

GetIntegerv GetString

– –

SHADING LANGUAGE VERSION

S

GetString

–

VENDOR

S

GetString

–

VERSION

S

GetString

–

EXTENSIONS

CONTEXT FLAGS

Description Supported individual extension names No. of individual extension names Major version no. supported Minor version no. supported Context flags Renderer string Shading Language version supported Vendor string OpenGL ES version supported

Sec. 20.2 20.2 20.2 20.2 20.2 20.2 20.2 20.2 20.2

489

Table 21.43: Implementation Dependent Version and Extension Support

Get value

OpenGL ES 3.2 (May 5, 2022)

Table 21.44: Implementation Dependent Vertex Shader Limits

Get value

Type

Get Command

Minimum Value

MAX VERTEX ATTRIBS

Z+

GetIntegerv

16

MAX VERTEX UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX VERTEX UNIFORM VECTORS

Z+

GetIntegerv

256

MAX VERTEX UNIFORM BLOCKS

Z+

GetIntegerv

12

MAX VERTEX OUTPUT COMPONENTS

Z+

GetIntegerv

64

MAX VERTEX TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX VERTEX ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

0

MAX VERTEX ATOMIC COUNTERS

Z+

GetIntegerv

0

MAX VERTEX SHADER STORAGE BLOCKS

Z+

GetIntegerv

0

Description No. of active vertex attributes No. of components for vertex shader uniform variables No. of vectors for vertex shader uniform variables Max. no. of vertex uniform buffers per program Max. no. of components of outputs written by a vertex shader No. of texture image units accessible by a vertex shader No. of atomic counter buffers accessed by a vertex shader No. of atomic counters accessed by a vertex shader No. of shader storage blocks accessed by a vertex shader

Sec. 10.2 7.6 7.6 7.6.2 11.1.2.1

11.1.3.5

7.7

7.7

7.8

490

OpenGL ES 3.2 (May 5, 2022)

Table 21.45: Implementation Dependent Tessellation Shader Limits

Get value

Type

Get Command

Minimum Value

MAX TESS GEN LEVEL

Z

+

GetIntegerv

64

MAX PATCH VERTICES

Z+

GetIntegerv

32

MAX TESS CONTROL UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX TESS CONTROL TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX TESS CONTROL OUTPUT COMPONENTS

Z+

GetIntegerv

64

MAX TESS PATCH COMPONENTS

Z+

GetIntegerv

120

Z+

GetIntegerv

2048

MAX TESS CONTROL INPUT COMPONENTS

Z+

GetIntegerv

64

MAX TESS CONTROL UNIFORM BLOCKS

Z+

GetIntegerv

12∗

MAX TESS CONTROL ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

0

MAX TESS CONTROL ATOMIC COUNTERS

Z+

GetIntegerv

0

MAX TESS CONTROL SHADER STORAGE BLOCKS

Z+

GetIntegerv

0

MAX TESS CONTROL TOTAL OUTPUT COMPONENTS

Sec. 11.2.2 10.1 11.2.1.1 11.1.3 11.2.1.2 11.2.1.2 11.2.1.2 11.2.1.2 7.6.2 7.7 7.7 7.8

491

Description Max. level supported by tess. primitive generator Max. patch size No. of words for tess. control shader (TCS) uniforms No. of tex. image units for TCS No. components for TCS per-vertex outputs No. components for TCS per-patch outputs Total no. components for TCS per-patch outputs No. components for TCS per-vertex inputs No. of supported uniform blocks for TCS No. of atomic counter (AC) buffers accessed by a TCS No. of ACs accessed by a TCS No. of shader storage blocks accessed by a tess. control shader

OpenGL ES 3.2 (May 5, 2022)

Minimum Value

Type

MAX TESS EVALUATION UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX TESS EVALUATION TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX TESS EVALUATION OUTPUT COMPONENTS

Z+

GetIntegerv

64

MAX TESS EVALUATION INPUT COMPONENTS

Z+

GetIntegerv

64

MAX TESS EVALUATION UNIFORM BLOCKS

Z+

GetIntegerv

12∗

Z+

GetIntegerv

0

Z+

GetIntegerv

0

Z+

GetIntegerv

0

MAX TESS EVALUATION ATOMIC COUNTER BUFFERS MAX TESS EVALUATION ATOMIC COUNTERS

MAX TESS EVALUATION SHADER STORAGE BLOCKS

Description No. of words for tess. evaluation shader (TES) uniforms No. of tex. image units for TES No. components for TES per-vertex outputs No. components for TES per-vertex inputs No. of supported uniform blocks for TES No. of AC buffers accessed by a TES No. of ACs accessed by a TES No. of shader storage blocks accessed by a tess. evaluation shader

Sec. 11.2.3.1 11.1.3 11.2.3.2 11.2.3.2 7.6.2 11.1.3.6 11.1.3.6 7.8

492

Table 21.46: Implementation Dependent Tessellation Shader Limits (cont.)

Get Command

Get value

OpenGL ES 3.2 (May 5, 2022)

Table 21.47: Implementation Dependent Geometry Shader Limits

Get value

Type

Get Command

Minimum Value

MAX GEOMETRY UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX GEOMETRY UNIFORM BLOCKS

Z+

GetIntegerv

12∗

MAX GEOMETRY INPUT COMPONENTS

Z+

GetIntegerv

64

MAX GEOMETRY OUTPUT COMPONENTS

Z+

GetIntegerv

64

MAX GEOMETRY OUTPUT VERTICES

Z+

GetIntegerv

256

MAX GEOMETRY TOTAL OUTPUT COMPONENTS

Z+

GetIntegerv

1024

MAX GEOMETRY TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX GEOMETRY SHADER INVOCATIONS

Z+

GetIntegerv

32

MAX GEOMETRY ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

0

MAX GEOMETRY ATOMIC COUNTERS

Z+

GetIntegerv

0

MAX GEOMETRY SHADER STORAGE BLOCKS

Z+

GetIntegerv

0

Sec. 11.3.3 7.6.2 11.3.4.3 11.3.4.4 11.3.4

11.3.4

11.3.4 11.3.4.2 7.7 11.1.3.6 7.8

493

Description No. of components for geometry shader (GS) uniform variables Max. no. of GS uniform buffers per program Max. no. of components of inputs read by a GS Max. no. of components of outputs written by a GS Max. no. of vertices that any GS can emit Max. no. of total components (all vertices) of active outputs that a GS can emit No. of texture image units accessible by a GS Max. supported GS invocation count No. of atomic counter buffers accessed by a GS No. of atomic counters accessed by a GS No. of shader storage blocks accessed by a GS

Get value

Type

Get Command

Minimum Value

Table 21.48: Implementation Dependent Fragment Shader Limits

OpenGL ES 3.2 (May 5, 2022)

Z+

GetIntegerv

1024

MAX FRAGMENT UNIFORM VECTORS

Z+

GetIntegerv

256

MAX FRAGMENT UNIFORM BLOCKS

Z+

GetIntegerv

12

MAX FRAGMENT INPUT COMPONENTS

Z+

GetIntegerv

60

MAX TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX FRAGMENT ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

1

MAX FRAGMENT ATOMIC COUNTERS

Z+

GetIntegerv

8

MAX FRAGMENT SHADER STORAGE BLOCKS

Z+

GetIntegerv

4

MIN PROGRAM TEXTURE GATHER OFFSET

Z

GetIntegerv

–

MAX PROGRAM TEXTURE GATHER OFFSET

Z+

GetIntegerv

–

MIN PROGRAM TEXEL OFFSET

Z

GetIntegerv

-8

MAX PROGRAM TEXEL OFFSET

Z

GetIntegerv

7

Sec. 14.1

14.1

7.6.2

14.2.2

11.1.3.5

7.7

7.7

7.8 8.14 8.14 11.1.3.5 11.1.3.5

494

MAX FRAGMENT UNIFORM COMPONENTS

Description No. of components for fragment shader uniform variables No. of vectors for fragment shader uniform variables Max. no. of fragment uniform buffers per program Max. no. of components of inputs read by a fragment shader No. of texture image units accessible by a fragment shader No. of atomic counter buffers accessed by a fragment shader No. of atomic counters accessed by a fragment shader No. of shader storage blocks accessed by a fragment shader Min. texel offset for textureGather Max. texel offset for textureGather Min. texel offset allowed in lookup Max. texel offset allowed in lookup

Table 21.49: Implementation Dependent Compute Shader Limits

Get Command

MAX COMPUTE WORK GROUP COUNT

3 × Z+

GetIntegeri v

65535

MAX COMPUTE WORK GROUP SIZE

3 × Z+

GetIntegeri v

128 (x, y), 64 (z)

MAX COMPUTE WORK GROUP INVOCATIONS

Z+

GetIntegerv

128

MAX COMPUTE UNIFORM BLOCKS

Z+

GetIntegerv

12

MAX COMPUTE TEXTURE IMAGE UNITS

Z+

GetIntegerv

16

MAX COMPUTE SHARED MEMORY SIZE

Z+

GetIntegerv

16384

MAX COMPUTE UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX COMPUTE ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

1

MAX COMPUTE ATOMIC COUNTERS

Z+

GetIntegerv

8

Z+

GetIntegerv

†

Z+

GetIntegerv

4

MAX COMBINED COMPUTE UNIFORM COMPONENTS

MAX COMPUTE SHADER STORAGE BLOCKS

Minimum Value

Description Max. no. of workgroups that may be dispatched by a single dispatch command (per dimension) Max. local size of a compute workgroup (per dimension) Max. total compute shader (CS) invocations in a single local workgroup Max. no. of uniform blocks per compute program Max. no. of texture image units accessible by a CS Max. total storage size of all variables declared as shared in all CSs linked into a single program object No. of components for CS uniform variables No. of atomic counter buffers accessed by a CS No. of atomic counters accessed by a CS No. of words for CS uniform variables in all uniform blocks, including the default No. of shader storage blocks accessed by a compute shader

Sec. 17

17

17

11.1.3

11.1.3

17.1

17.1 7.7 11.1.3

17.1

7.8

495

† The minimum value is MAX_COMPUTE_UNIFORM_BLOCKS × MAX_UNIFORM_BLOCK_SIZE / 4 + MAX_COMPUTE_UNIFORM_COMPONENTS

OpenGL ES 3.2 (May 5, 2022)

Type

Get value

Table 21.50: Implementation Dependent Aggregate Shader Limits

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

MAX UNIFORM BUFFER BINDINGS

Z+

GetIntegerv

72

MAX UNIFORM BLOCK SIZE

Z+

GetInteger64v

16384

UNIFORM BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

MAX COMBINED UNIFORM BLOCKS

Z+

GetIntegerv

60∗

Z+

GetInteger64v

‡

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Z+

GetInteger64v

‡

MAX COMBINED VERTEX UNIFORM COMPONENTS

MAX COMBINED TESS CONTROL UNIFORM COMPONENTS

MAX COMBINED TESS EVALUATION UNIFORM COMPONENTS

MAX COMBINED GEOMETRY UNIFORM COMPONENTS

MAX COMBINED FRAGMENT UNIFORM COMPONENTS

Description Max. no. of uniform buffer binding points on the context Max. size in basic machine units of a uniform block Min. required alignment for uniform buffer sizes and offsets Max. no. of uniform buffers per program No. of words for vertex shader uniform var. in all uniform blocks (incl. default) No. of words for TCS uniform var. in all uniform blocks (incl. default) No. of words for TES uniform var. in all uniform blocks (incl. default) No. of words for geometry shader uniform var. in all uniform blocks (incl. default) No. of words for fragment shader uniform var. in all uniform blocks (incl. default)

Sec. 7.6.2

7.6.2

7.6.2 7.6.2

7.6.2

11.2.1.1

11.2.3.1

7.6.2

496

† The value of UNIFORM_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum. ‡The minimum value for each stage is MAX_stage_UNIFORM_BLOCKS × MAX_UNIFORM_BLOCK_SIZE / 4 + MAX_stage_UNIFORM_COMPONENTS. The limit is totalled for all uniform variables in all uniform blocks, including the default.

Get value

7.6.2

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

MAX VARYING COMPONENTS

Z+

GetIntegerv

60

MAX VARYING VECTORS

Z+

GetIntegerv

15

MAX COMBINED TEXTURE IMAGE UNITS

Z+

GetIntegerv

96

MAX COMBINED SHADER OUTPUT RESOURCES

Z+

GetIntegerv

4

MAX UNIFORM LOCATIONS

Z+

GetIntegerv

1024

MAX ATOMIC COUNTER BUFFER BINDINGS

Z+

GetIntegerv

1

MAX ATOMIC COUNTER BUFFER SIZE

Z+

GetIntegerv

32

MAX COMBINED ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

1

MAX COMBINED ATOMIC COUNTERS

Z+

GetIntegerv

8

Description No. of components for output variables No. of vectors for output variables Total no. of texture units accessible by the GL Limit on active image units, shader storage blocks, and frag. outputs Max. no. of userassignable uniform locations Max. no. of atomic counter buffer bindings Max. size in basic machine units of an atomic counter buffer Max. no. of atomic counter buffers per program Max. no. of atomic counter uniforms per program

Sec. 11.1.2.1 11.1.2.1 11.1.3.5 8.23

7.6 7.7 7.7

7.7

7.7

497

Table 21.51: Implementation Dependent Aggregate Shader Limits (cont.)

Get value

Minimum Value

MAX IMAGE UNITS

Z

+

GetIntegerv

4

MAX VERTEX IMAGE UNIFORMS

Z+

GetIntegerv

0

MAX TESS CONTROL IMAGE UNIFORMS

Z+

GetIntegerv

0

MAX TESS EVALUATION IMAGE UNIFORMS

Z+

GetIntegerv

0

MAX GEOMETRY IMAGE UNIFORMS

Z+

GetIntegerv

0

MAX FRAGMENT IMAGE UNIFORMS

Z+

GetIntegerv

4

+

MAX COMBINED IMAGE UNIFORMS

Z Z+

GetIntegerv GetIntegerv

4 4

MAX SHADER STORAGE BUFFER BINDINGS

Z+

GetIntegerv

4

MAX SHADER STORAGE BLOCK SIZE

Z+

GetInteger64v

227

MAX COMBINED SHADER STORAGE BLOCKS

Z+

GetIntegerv

4

SHADER STORAGE BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

MAX COMPUTE IMAGE UNIFORMS

Type

Description No. of units for image load/store/atomics No. of image variables in vertex shaders No. of image variables in tess. control shaders No. of image variables in tess. eval. shaders No. of image variables in geometry shaders No. of image variables in fragment shaders No. of image variables in CSs No. of image variables in all shaders Max. no. of shader storage buffer bindings in the context Max. size in basic machine units of a shader storage block No. of shader storage blocks accessed by a program Min. required alignment for shader storage buffer binding offsets

Sec. 8.23 11.1.3 11.1.3.7 11.1.3.7 11.1.3.7 11.1.3 11.1.3 11.1.3 7.8 7.8 7.8 7.8

498

Table 21.52: Implementation Dependent Aggregate Shader Limits (cont.)

† The value of SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum.

OpenGL ES 3.2 (May 5, 2022)

Get Command

Get value

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Initial Value

DEBUG CALLBACK FUNCTION

Y

GetPointerv

NULL

DEBUG CALLBACK USER PARAM

Y

GetPointerv

NULL

DEBUG LOGGED MESSAGES

Z+

GetIntegerv

0

Z+

GetIntegerv

0

DEBUG OUTPUT SYNCHRONOUS

B

IsEnabled

FALSE

DEBUG GROUP STACK DEPTH

Z+

GetIntegerv

1

DEBUG OUTPUT

B

IsEnabled

Depends on the context†

DEBUG NEXT LOGGED MESSAGE LENGTH

Description The current debug output callback function pointer The current debug output callback user parameter The no. of messages currently in the debug message log The string length of the oldest debug message in the debug message log The enabled state for synchronous debug message callbacks Debug group stack pointer The enabled state for debug output functionality

Sec. 18.2 18.2 18.3

18.3

18.8 18.6 18

499

Table 21.53: Debug Output State † The initial value of DEBUG_OUTPUT is TRUE in a debug context and FALSE in a non-debug context.

Get value

Table 21.54: Implementation Dependent Debug Output State

OpenGL ES 3.2 (May 5, 2022)

Get Command

Minimum Value

Get value

Type

MAX DEBUG MESSAGE LENGTH

Z+

GetIntegerv

1

MAX DEBUG LOGGED MESSAGES

Z+

GetIntegerv

1

MAX DEBUG GROUP STACK DEPTH

Z+

GetIntegerv

64

MAX LABEL LENGTH

Z+

GetIntegerv

256

Description The max length of a debug message string, including its null terminator The max no. of messages stored in the debug message log Max. group stack depth Max. length of a label string

Sec. 18.1

18.3 18.6 18.7

500

MAX TRANSFORM FEEDBACK INTERLEAVED COMPONENTS

MAX TRANSFORM FEEDBACK SEPARATE ATTRIBS

MAX TRANSFORM FEEDBACK SEPARATE COMPONENTS

Type

Get Command

Minimum Value

Z+

GetIntegerv

64

Z+

GetIntegerv

4

Z+

GetIntegerv

4

Description Max. no. of components to write to a single buffer in interleaved mode Max. no. of separate attributes or outputs that can be captured in transform feedback Max. no. of components per attribute or output in separate mode

Sec. 12.2

12.2

12.2

501

Table 21.55: Implementation Dependent Transform Feedback Limits

OpenGL ES 3.2 (May 5, 2022)

Get value

Table 21.56: Framebuffer Dependent Values

OpenGL ES 3.2 (May 5, 2022)

Type

Get Command

Minimum Value

SAMPLE BUFFERS

Z2

GetIntegerv

0

SAMPLES

Z+

GetIntegerv

0

x BITS

Z+

GetIntegerv

–

DEPTH BITS

Z+

GetIntegerv

–

STENCIL BITS

Z+

GetIntegerv

–

IMPLEMENTATION COLOR READ TYPE

E

GetIntegerv

–

IMPLEMENTATION COLOR READ FORMAT

E

GetIntegerv

–

SAMPLE POSITION

n × 2 × R[0,1]

GetMultisamplefv

impl-dependent

Description No. of multisample buffers Coverage mask size No. of bits in x color buffer component. x is one of RED, GREEN, BLUE, ALPHA No. of depth buffer planes No. of stencil planes Implementation preferred pixel type † Implementation preferred pixel format † Explicit sample positions

Sec. 13.4 13.4 9

9 9 16.1 16.1 13.4

502

† This state is queried from the currently bound read framebuffer, rather than the draw framebuffer. n is the value of SAMPLES.

Get value

Table 21.57: Miscellaneous

OpenGL ES 3.2 (May 5, 2022)

Type n×E n×B 3 × Z+

Get Command GetError – GetQueryiv

Initial Value 0 FALSE 0

COPY READ BUFFER BINDING

Z+

GetIntegerv

0

COPY WRITE BUFFER BINDING

Z+

GetIntegerv

0

RESET NOTIFICATION STRATEGY

Z2

GetIntegerv

See sec. 2.3.2

+

GetIntegerv

0

Get value -– -– CURRENT QUERY

TEXTURE BUFFER BINDING

Z

Description Current error code(s) True if there is a corresponding error Active query object names Buffer object bound to copy buffer “read” bind point Buffer object bound to copy buffer “write” bind point Reset notification behavior Buffer object bound to generic texture buffer bind point

Sec. 2.3.1 2.3.1 4.2.1 6.5 6.5 2.3.2 8.1

503

Appendix A

Invariance The OpenGL ES specification is not pixel exact. It therefore does not guarantee an exact match between images produced by different GL implementations. However, the specification does specify exact matches, in some cases, for images produced by the same implementation. The purpose of this appendix is to identify and provide justification for those cases that require exact matches.

A.1

Repeatability

The obvious and most fundamental case is repeated issuance of a series of GL commands. For any given GL and framebuffer state vector, and for any GL command, the resulting GL and framebuffer state must be identical whenever the command is executed on that initial GL and framebuffer state. This repeatability requirement doesn’t apply when using shaders containing side effects (image stores, image atomic operations, atomic counter operations, buffer variable stores, buffer variable atomic operations), because these memory operations are not guaranteed to be processed in a defined order. One purpose of repeatability is avoidance of visual artifacts when a doublebuffered scene is redrawn. If rendering is not repeatable, swapping between two buffers rendered with the same command sequence may result in visible changes in the image. Such false motion is distracting to the viewer. Another reason for repeatability is testability. Repeatability, while important, is a weak requirement. Given only repeatability as a requirement, two scenes rendered with one (small) polygon changed in position might differ at every pixel. Such a difference, while within the law of repeatability, is certainly not within its spirit. Additional invariance rules are desirable to ensure useful operation.

504

A.2. MULTI-PASS ALGORITHMS

A.2

505

Multi-pass Algorithms

Invariance is necessary for a whole set of useful multi-pass algorithms. Such algorithms render multiple times, each time with a different GL mode vector, to eventually produce a result in the framebuffer. Examples of these algorithms include: • “Erasing” a primitive from the framebuffer by redrawing it in a different color. • Using stencil operations to compute capping planes for stencil shadow volumes. On the other hand, invariance rules can greatly increase the complexity of highperformance implementations of the GL. Even the weak repeatability requirement significantly constrains a parallel implementation of the GL. Because GL implementations are required to implement ALL GL capabilities, not just a convenient subset, those that utilize hardware acceleration are expected to alternate between hardware and software modules based on the current GL mode vector. A strong invariance requirement forces the behavior of the hardware and software modules to be identical, something that may be very difficult to achieve (for example, if the hardware does floating-point operations with different precision than the software). What is desired is a compromise that results in many compliant, highperformance implementations, and in many software vendors choosing to port to OpenGL ES.

A.3

Invariance Rules

For a given instantiation of an OpenGL ES rendering context: Rule 1 For any given GL and framebuffer state vector, and for any given GL command, the resulting GL and framebuffer state must be identical each time the command is executed on that initial GL and framebuffer state. Rule 2 Changes to the following state values have no side effects (the use of any other state value is not affected by the change): Required: • Framebuffer contents (all bitplanes) • The color buffers enabled for writing

OpenGL ES 3.2 (May 5, 2022)

A.3. INVARIANCE RULES

506

• Scissor parameters (other than enable) • Writemasks (color, depth, stencil) • Clear values (color, depth, stencil) Strongly suggested: • Stencil parameters (other than enable) • Depth test parameters (other than enable) • Blend parameters (other than enable) • Pixel storage state • Polygon offset parameters (other than enables, and except as they affect the depth values of fragments) Corollary 1 Fragment generation is invariant with respect to the state values marked with • in rule 2. Rule 3 The arithmetic of each per-fragment operation is invariant except with respect to parameters that directly control it. Corollary 2 Images rendered into different color buffers sharing the same framebuffer, either simultaneously or separately using the same command sequence, are pixel identical. Rule 4 The same vertex or fragment shader will produce the same result when run multiple times with the same input. The wording ‘the same shader’ means a program object that is populated with the same source strings, which are compiled and then linked, possibly multiple times, and which program object is then executed using the same GL state vector. Invariance is relaxed for shaders with side effects, such as image stores, image atomic operations, or accessing atomic counters (see section A.5). Rule 5 All fragment shaders that either conditionally or unconditionally assign gl_FragCoord.z to gl_FragDepth are depth-invariant with respect to each other, for those fragments where the assignment to gl_FragDepth actually is done. If a sequence of GL commands specifies primitives to be rendered with shaders containing side effects (image stores, image atomic operations, atomic counter operations, buffer variable stores, buffer variable atomic operations), invariance rules

OpenGL ES 3.2 (May 5, 2022)

A.4. TESSELLATION INVARIANCE

507

are relaxed. In particular, rule 1, corollary 2, and rule 4 do not apply in the presence of shader side effects. The following weaker versions of rule 1 and rule 4 apply to GL commands involving shader side effects: Rule 6 For any given GL and framebuffer state vector, and for any given GL command, the contents of any framebuffer state not directly or indirectly affected by results of shader image stores, image atomic operations, or atomic counter operations must be identical each time the command is executed on that initial GL and framebuffer state. Rule 7 The same vertex or fragment shader will produce the same result when run multiple times with the same input as long as: • shader invocations do not use image atomic operations or atomic counters; • no framebuffer memory is written to more than once by image stores, unless all such stores write the same value; and • no shader invocation, or other operation performed to process the sequence of commands, reads memory written to by an image store. When any sequence of GL commands triggers shader invocations that perform image stores, image atomic operations, atomic counter operations, buffer variable stores, or buffer variable atomic operations), and subsequent GL commands read the memory written by those shader invocations, these operations must be explicitly synchronized. For more details, see section 7.11.

A.4

Tessellation Invariance

When using a program containing tessellation evaluation shaders, the fixedfunction tessellation primitive generator consumes the input patch specified by an application and emits a new set of primitives. The following invariance rules are intended to provide repeatability guarantees. Additionally, they are intended to allow an application with a carefully crafted tessellation evaluation shader to ensure that the sets of triangles generated for two adjacent patches have identical vertices along shared patch edges, avoiding “cracks” caused by minor differences in the positions of vertices along shared edges. Rule 1 When processing two patches with identical outer and inner tessellation levels, the tessellation primitive generator will emit an identical set of point, line,

OpenGL ES 3.2 (May 5, 2022)

A.4. TESSELLATION INVARIANCE

508

or triangle primitives as long as the active program used to process the patch primitives has tessellation evaluation shaders specifying the same tessellation mode, spacing, vertex order, and point mode input layout qualifiers. Two sets of primitives are considered identical if and only if they contain the same number and type of primitives and the generated tessellation coordinates for the vertex numbered m of the primitive numbered n are identical for all values of m and n. Rule 2 The set of vertices generated along the outer edge of the subdivided primitive in triangle and quad tessellation, and the tessellation coordinates of each, depends only on the corresponding outer tessellation level and the spacing input layout qualifier in the tessellation evaluation shader of the active program. Rule 3 The set of vertices generated when subdividing any outer primitive edge is always symmetric. For triangle tessellation, if the subdivision generates a vertex with tessellation coordinates of the form (0, x, 1 − x), (x, 0, 1 − x), or (x, 1 − x, 0), it will also generate a vertex with coordinates of exactly (0, 1 − x, x), (1 − x, 0, x), or (1 − x, x, 0), respectively. For quad tessellation, if the subdivision generates a vertex with coordinates of (x, 0) or (0, x), it will also generate a vertex with coordinates of exactly (1 − x, 0) or (0, 1 − x), respectively. For isoline tessellation, if it generates vertices at (0, x) and (1, x) where x is not zero, it will also generate vertices at exactly (0, 1 − x) and (1, 1 − x), respectively. Rule 4 The set of vertices generated when subdividing outer edges in triangular and quad tessellation must be independent of the specific edge subdivided, given identical outer tessellation levels and spacing. For example, if vertices at (x, 1 − x, 0) and (1−x, x, 0) are generated when subdividing the w = 0 edge in triangular tessellation, vertices must be generated at (x, 0, 1 − x) and (1 − x, 0, x) when subdividing an otherwise identical v = 0 edge. For quad tessellation, if vertices at (x, 0) and (1 − x, 0) are generated when subdividing the v = 0 edge, vertices must be generated at (0, x) and (0, 1 − x) when subdividing an otherwise identical u = 0 edge. Rule 5 When processing two patches that are identical in all respects enumerated in rule 1 except for vertex order, the set of triangles generated for triangle and quad tessellation must be identical except for vertex and triangle order. For each triangle n1 produced by processing the first patch, there must be a triangle n2 produced when processing the second patch each of whose vertices has the same tessellation coordinates as one of the vertices in n1 . Rule 6 When processing two patches that are identical in all respects enumerated in rule 1 other than matching outer tessellation levels and/or vertex order, the set

OpenGL ES 3.2 (May 5, 2022)

A.5. ATOMIC COUNTER INVARIANCE

509

of interior triangles generated for triangle and quad tessellation must be identical in all respects except for vertex and triangle order. For each interior triangle n1 produced by processing the first patch, there must be a triangle n2 produced when processing the second patch each of whose vertices has the same tessellation coordinates as one of the vertices in n1 . A triangle produced by the tessellator is considered an interior triangle if none of its vertices lie on an outer edge of the subdivided primitive. Rule 7 For quad and triangle tessellation, the set of triangles connecting an inner and outer edge depends only on the inner and outer tessellation levels corresponding to that edge and the spacing input layout qualifier. Rule 8 The value of all defined components of gl_TessCoord will be in the range [0, 1]. Additionally, for any defined component x of gl_TessCoord, the results of computing 1.0 − x in a tessellation evaluation shader will be exact. Some floatingpoint values in the range [0, 1] may fail to satisfy this property, but such values may never be used as tessellation coordinate components.

A.5

Atomic Counter Invariance

When using a program containing atomic counters, the following invariance rules are intended to provide repeatability guarantees but within certain constraints. Rule 1 When a single shader type within a program accesses an atomic counter with only atomicCounterIncrement, any individual shader invocation is guaranteed to get a unique value returned. Corollary 1 Also holds true with atomicCounterDecrement. Corollary 2 This does not hold true for atomicCounter. Corollary 3 Repeatability is relaxed. While a unique value is returned to the shader, even given the same initial state vector and buffer contents, it is not guaranteed that the same unique value will be returned for each individual invocation of a shader (For example, on any single vertex, or any single fragment). It is wholly the shader writer’s responsibility to respect this constraint. Rule 2 When two or more shader types within a program access an atomic counter with only atomicCounterIncrement, there is no repeatability of the ordering of operations between stages. For example, some number of vertices may be processed, then some number of fragments may be processed. Corollary 4 This also holds true with atomicCounterDecrement and atomicCounter.

OpenGL ES 3.2 (May 5, 2022)

A.6. WHAT ALL THIS MEANS

A.6

510

What All This Means

Hardware accelerated GL implementations are expected to default to software operation when some GL state vectors are encountered. Even the weak repeatability requirement means, for example, that OpenGL ES implementations cannot apply hysteresis to this swap, but must instead guarantee that a given mode vector implies that a subsequent command always is executed in either the hardware or the software machine. The stronger invariance rules constrain when the switch from hardware to software rendering can occur, given that the software and hardware renderers are not pixel identical. For example, the switch can be made when blending is enabled or disabled, but it should not be made when a change is made to the blending parameters. Because floating point values may be represented using different formats in different renderers (hardware and software), many OpenGL ES state values may change subtly when renderers are swapped. This is the type of state value change that rule 1 in section A.3 seeks to avoid.

OpenGL ES 3.2 (May 5, 2022)

Appendix B

Corollaries The following observations are derived from the body and the other appendixes of the specification. Absence of an observation from this list in no way impugns its veracity. 1. The error semantics of upward compatible OpenGL ES revisions may change. Otherwise, only additions can be made to upward compatible revisions. 2. GL query commands are not required to satisfy the semantics of the Flush or the Finish commands. All that is required is that the queried state be consistent with complete execution of all previously executed GL commands. 3. Application specified line width must be returned as specified when queried. Implementation-dependent clamping affects the values only while they are in use. 4. The mask specified as the third argument to StencilFunc affects the operands of the stencil comparison function, but has no direct effect on the update of the stencil buffer. The mask specified by StencilMask has no effect on the stencil comparison function; it limits the effect of the update of the stencil buffer. 5. There is no atomicity requirement for OpenGL ES rendering commands, even at the fragment level. 6. Because rasterization of polygons is point sampled, polygons that have no area generate no fragments when they are rasterized, and the fragments generated by the rasterization of “narrow” polygons may not form a continuous array.

511

512 7. OpenGL ES does not force left- or right-handedness on any of its coordinates systems. 8. (No pixel dropouts or duplicates.) Let two polygons share an identical edge. That is, there exist vertices A and B of an edge of one polygon, and vertices C and D of an edge of the other polygon; the positions of vertex A and C are identical; and the positions of vertex B and D are identical. Vertex positions are identical if the gl_Position values output by the vertex shader are identical. Then, when the fragments produced by rasterization of both polygons are taken together, each fragment intersecting the interior of the shared edge is produced exactly once. 9. Dithering algorithms may be different for different components. In particular, alpha may be dithered differently from red, green, or blue, and an implementation may choose to not dither alpha at all.

OpenGL ES 3.2 (May 5, 2022)

Appendix C

Compressed Texture Image Formats The compressed texture formats used by OpenGL ES are described in the specifically identified sections of the Khronos Data Format Specification, version 1.1, available at URL https://www.khronos.org/registry/dataformat/specs/1.1/dataformat.1.1.html Unless otherwise described, the quantities encoded in these compressed formats are treated as normalized, unsigned values. Those formats listed as sRGB-encoded have in-memory representations of R, G and B components which are nonlinearly-encoded as R0 , G0 , and B 0 ; any alpha component is unchanged. As part of filtering, the nonlinear R0 , G0 , and B 0 values are converted to linear R, G, and B components; any alpha component is unchanged. The conversion between linear and nonlinear encoding is performed as described in the KHR_DF_TRANSFER_SRGB section of the Khronos Data Format Specification.

C.1

ASTC Compressed Texture Image Formats

ASTC formats are described in the “ASTC Compressed Texture Image Formats” chapter of the Khronos Data Format Specification. The mapping between OpenGL ES ASTC formats and that specification is shown in table C.1. Only the ASTC LDR Profile modes are supported by OpenGL ES.

513

C.1. ASTC COMPRESSED TEXTURE IMAGE FORMATS

514

OpenGL ES format

sRGB-encoded

COMPRESSED_RGBA_ASTC_4x4 COMPRESSED_RGBA_ASTC_5x4 COMPRESSED_RGBA_ASTC_5x5 COMPRESSED_RGBA_ASTC_6x5 COMPRESSED_RGBA_ASTC_6x6 COMPRESSED_RGBA_ASTC_8x5 COMPRESSED_RGBA_ASTC_8x6 COMPRESSED_RGBA_ASTC_8x8 COMPRESSED_RGBA_ASTC_10x5 COMPRESSED_RGBA_ASTC_10x6 COMPRESSED_RGBA_ASTC_10x8 COMPRESSED_RGBA_ASTC_10x10 COMPRESSED_RGBA_ASTC_12x10 COMPRESSED_RGBA_ASTC_12x12 COMPRESSED_SRGB8_ALPHA8_ASTC_4x4 COMPRESSED_SRGB8_ALPHA8_ASTC_5x4 COMPRESSED_SRGB8_ALPHA8_ASTC_5x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x5 COMPRESSED_SRGB8_ALPHA8_ASTC_6x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x5 COMPRESSED_SRGB8_ALPHA8_ASTC_8x6 COMPRESSED_SRGB8_ALPHA8_ASTC_8x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x5 COMPRESSED_SRGB8_ALPHA8_ASTC_10x6 COMPRESSED_SRGB8_ALPHA8_ASTC_10x8 COMPRESSED_SRGB8_ALPHA8_ASTC_10x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x10 COMPRESSED_SRGB8_ALPHA8_ASTC_12x12

Compressed texel block dimensions 4×4 5×4 5×5 6×5 6×6 8×5 8×6 8×8 10 × 5 10 × 6 10 × 8 10 × 10 12 × 10 12 × 12 4×4 5×4 5×5 6×5 6×6 8×5 8×6 8×8 10 × 5 10 × 6 10 × 8 10 × 10 12 × 10 12 × 12

Table C.1: Mapping of OpenGL ES ASTC formats to descriptions.

OpenGL ES 3.2 (May 5, 2022)

             

C.2. ETC COMPRESSED TEXTURE IMAGE FORMATS OpenGL ES format COMPRESSED_R11_EAC COMPRESSED_SIGNED_R11_EAC COMPRESSED_RG11_EAC COMPRESSED_SIGNED_RG11_EAC COMPRESSED_RGB8_ETC2 COMPRESSED_SRGB8_ETC2 COMPRESSED_RGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_RGBA8_ETC2_EAC COMPRESSED_SRGB8_ALPHA8_ETC2_EAC

515 Data Format Specification description Unsigned R11 EAC Signed R11 EAC Unsigned RG11 EAC Signed RG11 EAC RGB ETC2 RGB ETC2 with sRGB encoding RGB ETC2 with punchthrough alpha RGB ETC2 with punchthrough alpha and sRGB encoding RGBA ETC2 RGBA ETC2 with sRGB encoding

Table C.2: Mapping of OpenGL ES ETC formats to descriptions.

C.2

ETC Compressed Texture Image Formats

ETC formats are described in the “ETC2 Compressed Texture Image Formats” chapter of the Khronos Data Format Specification. The mapping between OpenGL ES ETC formats and that specification is shown in table C.2.

OpenGL ES 3.2 (May 5, 2022)

Appendix D

Version 3.0 and Before OpenGL ES version 3.0, released on August 6, 2012, is the third revision since the original version 1.0. OpenGL ES 3.0 is upward compatible with OpenGL ES version 2.0, meaning that any program that runs with an OpenGL ES 2.0 implementation will also run unchanged with an OpenGL ES 3.0 implementation. Note the subtle changes in runtime behavior between versions 2.0 and 3.0, documented in Appendix G.2. Following are brief descriptions of changes and additions to OpenGL ES 3.0.

D.1

New Features

New features in OpenGL ES 3.0 include: • OpenGL Shading Language ES 3.00 • transform feedback 1 and 2 (with restrictions) • uniform buffer objects including block arrays • vertex array objects • sampler objects • sync objects and fences • pixel buffer objects • buffer subrange mapping • buffer object to buffer object copies

516

D.1. NEW FEATURES

517

• boolean occlusion queries, including conservative mode • instanced rendering, via shader variable and/or vertex attribute divisor • multiple render targets • 2D array and 3D textures • simplified texture storage specification • R and RG textures • texture swizzles • seamless cube maps • non-power-of-two textures with full wrap mode support and mipmapping • texture LOD clamps and mipmap level base offset and max clamp • at least 32 textures, at least 16 each for fragment and vertex shaders • 16-bit (with filtering) and 32-bit (without filtering) floating-point textures • 32-bit, 16-bit, and 8-bit signed and unsigned integer renderbuffers, textures, and vertex attributes • 8-bit sRGB textures and framebuffers (without mixed RGB/sRGB rendering) • 11/11/10 floating-point RGB textures • shared exponent RGB 9/9/9/5 textures • 10/10/10/2 unsigned normalized and unnormalized integer textures • 10/10/10/2 signed and unsigned normalized vertex attributes • 16-bit floating-point vertex attributes • 8-bit-per-component signed normalized textures • ETC2/EAC texture compression formats • sized internal texture formats with minimum precision guarantees • multisample renderbuffers

OpenGL ES 3.2 (May 5, 2022)

D.2. CHANGE LOG FOR 3.0.3

518

• 8-bit unsigned normalized renderbuffers • depth textures and shadow comparison • 24-bit depth renderbuffers and textures • 24/8 depth/stencil renderbuffers and textures • 32-bit depth and 32F/8 depth/stencil renderbuffers and textures • stretch blits (with restrictions) • framebuffer invalidation hints • primitive restart with fixed index • unsigned integer element indices with at least 24 usable bits • draw command allowing specification of range of accessed elements • ability to attach any mipmap level to a framebuffer object • minimum/maximum blend equations • program binaries, including querying binaries from linked GLSL programs • mandatory online compiler • non-square and transposable uniform matrices • additional pixel store state • indexed extension string queries

D.2

Change Log for 3.0.3

Changes since the 3.0.2 specification: • Remove ”non-64-bit” from first sentence of section 6.1.2 (Bug 7895). • Remove redundant reference to setting TEXTURE_IMMUTABLE_FORMAT and TEXTURE_IMMUTABLE_LEVELS from the end of section 3.8.4 (Bug 9342). • Clarify framebuffer attachment completeness rules with respect to the FRAMEBUFFER_ATTACHMENT_TEXTURE_LEVEL and mipmap completeness (Bug 9689).

OpenGL ES 3.2 (May 5, 2022)

D.2. CHANGE LOG FOR 3.0.3

519

• Clarify active uniform enumeration rules (Bug 9797). • Clarify behavior of mipmap completeness with unsized base internal formats (Bug 9807). • Introduce INVALID_VALUE error when BindBufferRange is called with a negative offset (Bug 9873). • Clarify that when DrawBuffers is called with 0 as the value of n, in the default framebuffer case INVALID_OPERATION is generated, and in the framebuffer object case, NONE is assigned to all draw buffers (Bug 10059). • Allow alternate formulation of equation 3.21’s mipmap array selection (Bug 10119). • Untangle ReadBuffer from ReadPixels and put it into its own section, while clarifying the error conditions (Bug 10172). • Specify that std140 and shared layout uniform blocks and their members are always active (Bug 10182). • Introduce missing INVALID_OPERATION error when BindAttribLocation is called with a name that starts with the reserved "gl_" prefix (Bug 10271). • Clarify return values from GetFramebufferAttachmentParameteriv of NONE and LINEAR for FRAMEBUFFER_ATTACHMENT_COMPONENT_TYPE and FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING, respectively, when the attachment has not been initialized (Bug 10357). • Fix description of fragment shader outputs to only require explicit output variable bindings to fragment colors when there are more than one output variable (Bug 10363). • Clarify that ValidateProgram is only required to check for the errors described in the Validation section, not all INVALID_OPERATION errors that can be generated by rendering commands (Bug 10650). • Clarify behavior of commands that don’t specify whether an error is generated when accessing a mapped buffer object (Bug 10684). • Clarify that SAMPLE_BUFFERS and SAMPLES are framebuffer-dependent state, and that SAMPLE_BUFFERS can only assume the values zero or one (Bug 10689).

OpenGL ES 3.2 (May 5, 2022)

D.3. CHANGE LOG FOR 3.0.2

520

• Simplify description of multisample rasterization to specify it is in effect when SAMPLE_BUFFERS is one, eliminating extraneous language about GL contexts, EGL, etc. (Bug 10690). • Clarify the type of stencil bits in Table 8.11 (Bug 10748). • Clarify that writing different color values to the same image attached multiple times is undefined (Bug 10983). • Clean up description of FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER query (Bug 11199). • Clarify that samplers behave the same as textures, renderbuffers, and buffers with respective to object name lifetimes (Bug 11374).

D.3

Change Log for 3.0.2

Changes since the 3.0.1 specification: • Clarify BlitFramebuffer downsampling behavior for different types of samples (Bug 9690). • Clarify that program object state queries return the state presently in effect, which may be different than most recently set state (Bug 9702). • Clarify that current vertex attributes are not program object state (Bug 9781). • Clarify that integer state is undefined when set with out-of-range floatingpoint values (Bug 9846). • Clarify that Draw* commands are silently ignored when there is no current program object, rather than it being an error condition (Bug 9879). • Clarify that texel fetches are undefined when texel coordinates fall outside the computed level of detail, not the specified level of detail (Bug 9891). • Clarify which pixels are read and written by BlitFramebuffer (Bug 9946). • Clarify that either truncation or rounding are acceptable when converting from floating-point to normalized fixed-point (Bug 9976). • Make the minification vs. magnification switch-over point always zero (Bug 9997).

OpenGL ES 3.2 (May 5, 2022)

D.4. CHANGE LOG FOR 3.0.1

521

• Clarify that DrawArrays transfers no elements when count is zero (Bug 10015). • Tweak the language covering the conditions that can affect framebuffer completeness (Bug 10047). • Remove language in Appendix D that preserves binding-related state after an object is deleted and automatically unbound (Bug 10076). • Remove language in Appendix D that implies that active transform feedback objects can be deleted (Bug 10079).

D.4

Change Log for 3.0.1

Changes since the 3.0.0 specification: • Remove the clamp on reference value for shadow maps with floating-point depth formats (Bug 7975). • Clarify GetFramebufferAttachmentParameteriv behavior for a few different cases (Bug 9170). • Move description of level base and level max clamping for immutable textures to Mipmapping section (Bug 9342). • Remove references to floating-point formats when describing BlitFramebuffer (Bug 9388). • Remove PACK_IMAGE_HEIGHT and PACK_SKIP_IMAGES which have no effect (Bug 9414). • Require that Invalidate[Sub]Framebuffer accept DRAW_FRAMEBUFFER and READ_FRAMEBUFFER (Bug 9421). • Fix initial value of read buffer to be NONE if there is no default framebuffer associated with the context (Bug 9473). • Require that Invalidate[Sub]Framebuffer accept DEPTH_STENCIL_ATTACHMENT (Bug 9480). • Require that GenerateMipmap throw INVALID_OPERATION for depth textures (Bug 9481).

OpenGL ES 3.2 (May 5, 2022)

D.4. CHANGE LOG FOR 3.0.1

522

• Clarify that a texture is incomplete if it has a depth component, no shadow comparison, and linear filtering (also Bug 9481). • Minor tweaks to description of RGB9_E5 (Bug 9486). • Clarify behavior when drawing to an FBO with both NULL and non-NULL attachments (Bug 9494). • Clarify behavior of BindBufferBase (Bug 9513). • Return to a clamp-on-specification behavior for ClearDepth and DepthRange (Bug 9517). • Eliminate references to programs without fragment shaders (Bug 9543). • Move some uniform buffer state out of program object state tables (Bug 9566). • Clarify that gl_VertexID is undefined if any client-side vertex arrays are enabled (Bug 9603). • Clarify that vertex attribute aliasing is not permitted in conjunction with GLSL-ES 3.00 shaders (Bug 9609). • Fix description of LINK_STATUS which was incorrectly specified to return the compilation status (Bug 9698). • Clarifications and clean up in query object language (Bug 9766). • Clarify that mask may be zero for BlitFramebuffer indicating no action be taken (Bug 9748). • Clarify that arguments to TexSubImage* need not exactly match the values passed to TexImage* (Bug 9750). • Clarify that BindBufferRange only performs error checking of size and offset if buffer is not zero (Bug 9765). • Fix minor typos and other minor tweaks to transform feedback description (Bug 9842). • Clarify that primitives collected with transform feedback must match (not merely be compatible with) the transform feedback primitiveMode.

OpenGL ES 3.2 (May 5, 2022)

D.5. CREDITS AND ACKNOWLEDGEMENTS

523

• Clarify that only the specified portion(s) (depth and/or stencil) of depth/stencil attachment may be invalidated by Invalidate[Sub]Framebuffer. • Remove references to FLOAT in table 3.14. • Cleaned up index entries for state tables 6.13 and 6.35 which were overly verbose. • Added individual bookmarks to each state table in the PDF.

D.5

Credits and Acknowledgements

OpenGL ES 3.0 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ES Working Group during the development of OpenGL ES 3.0, including the company that they represented at the time of their contributions, follow. In addition, many people participated in developing desktop OpenGL specifications and extensions on which the OpenGL ES 3.0 functionality is based in large part; those individuals are listed in the respective specifications in the OpenGL Registry. Acorn Pooley, NVIDIA Alberto Moreira, Qualcomm Aleksandra Krstic, Qualcomm Alex Eddy, Apple Alon Or-Bach, Nokia Andrzej Kacprowski, Intel Arzhange Safdarzadeh, Intel Aske Simon Christensen, ARM Avi Shapira, Graphic Remedy Barthold Lichtenbelt, NVIDIA Ben Bowman, Imagination Technologies Ben Brierton, Broadcom Benj Lipchak, Apple Benson Tao, Vivante Bill Licea-Kane, AMD Brent Insko, Intel Brian Murray, Freescale Bruce Merry, ARM Carlos Santa, TI Cass Everitt, Epic Games & NVIDIA Cemil Azizoglu, TI Chang-Hyo Yu, Samsung Chris Dodd, NVIDIA Chris Knox, NVIDIA

Chris Tserng, TI Clay Montgomery, TI Cliff Gibson, Imagination Technologies Daniel Kartch, NVIDIA Daniel Koch, Transgaming Daoxiang Gong, Imagination Technologies Dave Shreiner, ARM David Garcia, AMD David Jarmon, Vivante Derek Cornish, Epic Games Dominik Witczak, ARM & Mobica Eben Upton, Broadcom Ed Plowman, Intel & ARM Eisaku Ohbuchi, DMP Elan Lennard, ARM Erik Faye-Lund, ARM Georg Kolling, Imagination Technologies Graeme Leese, Broadcom Graham Connor, Imagination Technologies Graham Sellers, AMD Greg Roth, NVIDIA Guillaume Portier, Hi Guofang Jiao, Qualcomm Hans-Martin Will, Vincent

OpenGL ES 3.2 (May 5, 2022)

D.5. CREDITS AND ACKNOWLEDGEMENTS

Hwanyong Lee, Huone I-Gene Leong, NVIDIA Ian Romanick, Intel Ian South-Dickinson, NVIDIA Ilan Aelion-Exch, Samsung Inkyun Lee, Huone Jacob Ström, Ericsson James Adams, Broadcom James Jones, Imagination Technologies James McCombe, Imagination Technologies Jamie Gennis, Google Jan-Harald Fredriksen, ARM Jani Vaisanen, Nokia Jarkko Kemppainen, Symbio Jauko Kylmaoja, Symbio Jeff Bolz, NVIDIA Jeff Leger, Qualcomm Jeff Vigil, Qualcomm Jeremy Sandmel, Apple Jeremy Thorne, Broadcom Jim Hauxwell, Broadcom Jinsung Kim, Huone Jiyoung Yoon, Huone Jon Kennedy, 3DLabs Jon Leech, Khronos Jonathan Putsman, Imagination Technologies Jørn Nystad, ARM Jussi Rasanen, NVIDIA Kalle Raita, drawElements Kari Pulli, Nokia Keith Whitwell, VMware Kent Miller, Netlogic Microsystems Kimmo Nikkanen, Nokia Konsta Karsisto, Nokia Krzysztof Kaminski, Intel Kyle Haughey, Apple Larry Seiler, Intel Lars Remes, Symbio Lee Thomason, Adobe Lefan Zhong, Vivante Luc Semeria, Apple Marcus Lorentzon, Ericsson Mark Butler, Imagination Technologies Mark Callow, Hi Mark Cresswell, Broadcom Mark Snyder, Alt Software Mark Young, AMD Mathieu Robart, STM Matt Russo, Matrox

524

Matthew Netsch, Qualcomm Maurice Ribble, AMD & Qualcomm Max Kazakov, DMP Mika Pesonen, Nokia Mike Cai, Vivante Mike Weiblen, Zebra Imaging Mila Smith, AMD Nakhoon Baek, Kyungpook Univeristy Nate Huang, NVIDIA Neil Trevett, NVIDIA Nelson Kidd, Intel Nick Haemel, AMD & NVIDIA Nick Penwarden, Epic Games Niklas Smedberg, Epic Games Nizar Romdan, ARM Oliver Wohlmuth, Fujitsu Pat Brown, NVIDIA Paul Ruggieri, Qualcomm Paul Wilkinson, Broadcom Per Wennersten, Ericsson Petri Talalla, Symbio Phil Huxley, ZiiLabs Philip Hatcher, Freescale Piers Daniell, NVIDIA Piotr Tomaszewski, Ericsson Piotr Uminski, Intel Rami Mayer, Samsung Rauli Laatikainen, RightWare Richard Schreyer, Apple Rob Barris, NVIDIA Rob Simpson, Qualcomm Robert Simpson, AMD Roj Langhi, Vivante Rune Holm, ARM Sami Kyostila, Nokia Scott Bassett, Apple Sean Ellis, ARM Shereef Shehata, TI Sila Kayo, Nokia Slawomir Grajewski, Intel Steve Hill, STM & Broadcom Steven Olney, DMP Suman Sharma, Intel Tapani Palli, Nokia Teemu Laakso, Symbio Tero Karras, NVIDIA Timo Suoranta, Imagination Technologies Tom Cooksey, ARM Tom McReynolds, NVIDIA

OpenGL ES 3.2 (May 5, 2022)

D.5. CREDITS AND ACKNOWLEDGEMENTS

Tom Olson, TI & ARM Tomi Aarnio, Nokia Tommy Asano, Takumi

525

Wes Bang, Nokia Yanjun Zhang, Vivante Yuan Wang, Imagination Technologies

The OpenGL ES Working Group gratefully acknowledges administrative support by the members of Gold Standard Group, including Andrew Riegel, Elizabeth Riegel, Glenn Fredericks, and Michelle Clark, and technical support from James Riordon, webmaster of Khronos.org and OpenGL.org.

OpenGL ES 3.2 (May 5, 2022)

Appendix E

Version 3.1 OpenGL ES version 3.1, released on March 17, 2014, is the fourth revision since the original version 1.0. OpenGL ES 3.1 is upward compatible with OpenGL ES version 3.0, meaning that any program that runs with an OpenGL ES 3.0 implementation will also run unchanged with an OpenGL ES 3.1 implementation. Following are brief descriptions of changes and additions to OpenGL ES 3.1.

E.1

New Features

New features in OpenGL ES 3.1 include: • Arrays of arrays (shading language only) • Compute shaders • Indirect draw commands (with draw parameters in buffer storage) • Explicit uniform location • Support for framebuffers with no attachments • Program interface queries • Atomic counters • Shader bitfield operations (shading language only) • Shader helper invocation (shading language only) • Shader image load/store operations

526

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

527

• Shader layout binding (shading language only) • Shader storage buffer objects • Separate shader objects • Stencil texturing • Texture gather operations • Multisample formats for immutable textures • Vertex attribute binding

E.2

Change Log for Released Specifications

Changes in the released Specification update of January 29, 2015 • Clean up error language around reuse of query objects in section 4.2 to make clear that occlusion query objects may be specified and reused with any mix of the two valid occlusion query targets (Bug 13342). • Add columns to table 7.3 marking which types can be declared as vertex attributes or returned by transform feedback (Bug 11553). • Cosmetic edit to match GL spec language in section 7.6 (Bug 11192). • Update minimum value of MAX_FRAGMENT_UNIFORM_VECTORS in table 21.48 from 224 to 256, matching MAX_FRAGMENT_UNIFORM_COMPONENTS (Bug 12731). • Restore description of MAX_UNIFORM_BLOCK_SIZE in section 7.6.2, which was lost in the restructuring of the OpenGL ES 3.1 Specification, but change behavior so that exceeding the limit will cause link failure, compared to may cause link failure in the older language (Bug 12897). • Fix OpenGL ES Shading Language Specification section reference in section 7.10. • Mark STENCIL_INDEX8 as a required renderbuffer format in table 8.11, which also was lost in the restructuring (Bug 13085).

OpenGL ES 3.2 (May 5, 2022)

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

528

• Rearrange descriptions of DrawArraysOneInstance, DrawElementsOneInstance, and the actual DrawElements* commands in section 10.5 to use the term “vertex ID” when referring to the actual element index of an element transferred to the GL, and make clear that the vertex ID does include the basevertex value passed to the DrawElements*BaseVertex commands. Add the baseinstance parameter consistently throughout these commands, for consistency with similar language in the OpenGL 4.5 specification, but make clear that its value is always zero in unextended OpenGL ES (Bug 12756). Changes in the released Specification update of October 29, 2014 • Modify description of active resource list enumeration in section 7.3.1 to treat only arrays of aggregate types as top-level arrays, and clarify how this applies to GetProgramResourceiv queries TOP_LEVEL_ARRAY_SIZE and TOP_LEVEL_ARRAY_STRIDE (Bug 11753). • Restore language describing non-sequentiality of resource locations for consecutive active array elements in section 7.3.1 (Bug 12318). • Restore fix in description of LINK_STATUS for GetProgramiv in section 7.12 (Bug 9698). • Clarify that filter state is ignored for multisample texture access in sections 8.8 and 11.1.3.3 (Bug 12171). • Restore missing error for base internal format arguments to TexStorage2DMultisample in section 8.8 (Bug 12468). • Add NEAREST_MIPMAP_NEAREST to the allowed filter modes for STENCIL_INDEX textures in section 8.17 (Bug 12791). • Clarify behavior of rendering to multiple framebuffer object attachments of different sizes in section 9.2 (Bug 10403). • Moved description of SAMPLE_BUFFERS and SAMPLE_BUFFERS from section 9.4.2 to new section 9.2.3.1, and add a comment about the effective value of these parameters for framebuffer objects other than the currently bound draw framebuffer. Change references to these parameters accordingly in sections 8.6, 13.4, 16.1.2, and 16.2.1 (Bug 12360). • Clarify in section 11.1.3.2 that texel fetches are undefined when texel coordinates fall outside the computed level of detail, not the specified level of detail (Bug 9891).

OpenGL ES 3.2 (May 5, 2022)

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

529

• Add description of conditions for which multisample texel fetch operations are undefined in section 11.1.3.3 (Bug 12255). • Clarify in section 11.1.3.11 that Draw* commands are silently ignored when there is no current program object and no current program pipeline object, rather than it being an error condition (Bug 9879). • Make validation fail in section 11.1.3.11 when an empty program pipeline object (one with no code for any shader stage) is current (Bug 12176). • Remove redundant sentence fragment in section 13.7.4 (Bug 12726). • Add missing error for invalid target argument to InvalidateSubFramebuffer in section 15.2.4 (Bug 12727). • Increase minimum values of MAX_FRAGMENT_UNIFORM_COMPONENTS and MAX_COMPUTE_UNIFORM_COMPONENTS to 1024 in tables 21.48 and 21.49, respectively, for consistency with other shader stages (Bug 12731). • Restore description of UNIFORM_BUFFER_OFFSET_ALIGNMENT in table 21.50, and of SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT in table 21.52 to refer to them as minimum required alignments, while the footnote in the caption continues to note that the numeric limits are the maximum allowed values (Bug 11962). Changes in the released Specification update of June 4, 2014: • Fix minor typos and remove references to unsupported floating-point framebuffers in sections 2.1, 8.6, 9.1, 9.4.3, 15.1.5, and 16.1.3 (Bug 11899). • Fix typo in description of BeginQuery in section 4.2 (Bug 11860), and specify minimum query result size in section 4.2.1 as 32 bits for primitiveswritten queries, and 1 bit for occlusion queries (Bug 11860). • Fix error condition for UseProgram in section 7.3 (Bug 12281). • Remove dangling references to setting an image uniform with Uniform* in section 7.6.1 (Bug 11443). • Update description of internal format determination for CopyTexImage2D in section 8.6 (Bug 9807, comment 57). • Update errors for TexStorage2DMultisample in section 8.8 to include an appropriate subset of the generic errors for TexStorage* commands defined in section 8.18, and remove redundant errors in section 8.18 (Bug 11937).

OpenGL ES 3.2 (May 5, 2022)

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

530

• Change definition of the value returned from invalid image load operations in section 8.23 to (0, 0, 0, x) where the A component is undefined (Bug 11182). • Fix error condition for GetFramebufferAttachmentParameteriv in section 9.2.3 (Bug 12180). • Replace dangling reference to nonexistent FramebufferTexture3D in description of FramebufferTextureLayer in section 9.2.8 (Bug 11964). • Remove bogus framebuffer completeness condition (left over from ES 3.0 spec) in section 9.4.2 (Bug 12273). • Specify the values of gl_VertexID in the descriptions of drawing pseudocommands DrawArraysOneInstance and DrawElementsOneInstance in section 10.5 (Bug 12202). • Add missing 0 parameter for baseinstance parameter of pseudocode describing DrawArraysInstanced, DrawElements, and DrawElementsInstanced in section 10.5 (Bug 11935). • Add description of ELEMENT_ARRAY_BUFFER_BINDING. to section 10.6 (Bug 11042). • Clarify description of BindAttribLocation in section 11.1.1 (Bug 12186). • Remove spurious reference to nonexistent TEXTURE_2D_MULTISAMPLE_ARRAY in section 20.3.1 (Bug 12250). • Fix get command for DEPTH_CLEAR_VALUE in table 21.14 • Reduce minimum value of MAX_COMPUTE_SHARED_MEMORY_SIZE from 32768 to 16384 and minimum value of MAX_COMPUTE_ATOMIC_COUNTER_BUFFERS from 8 to 1 in table 21.49 (Bugs 12028, 11944). • Change values of UNIFORM_BUFFER_OFFSET_ALIGNMENT in table 21.50, and of SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT in table 21.52 to 256, and make clear that these are maximum alignment values, not minimums (Bug 11962). • Use abbreviations “max.”, “min.”, and “no.” consistently in state tables in place of “maximum”, “minimum”, and “number”. Changes in the released Specification of March 17, 2014: • Added new features as described in section E.1.

OpenGL ES 3.2 (May 5, 2022)

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

531

• Restructure the Specification following similar restructuring of the OpenGL 4.3 specification. While much language has been moved around and many new sections added, aside from new descriptions of objects and the pipeline, actual language changes resulting from restructuring are relatively small and are indentified. The restructuring includes several bugfixes initially done in the GL specification but applicable to the ES specification as well. Additional changes to more closely match the current OpenGL specification include: – Minor language tweaks throughout for greater consistency and clarity. – Moved errors for (almost) all commands into explicit Errors blocks, including adding previously-implicit errors such as INVALID_VALUE for negative sizei parameters, as described in section 2.3.1. While the Error blocks are marked as changes, in almost all cases these are existing errors that have been collected in a single place for each command, rather than new errors (despite the color coding in the version of the specification document showing changes). Phrasing is changed to a consistent “An errorname error is generated if condition.” – Add table 7.1 of shader types and refer to it from elsewhere in the spec instead of enumerating all shader types repeatedly. – Reorganized description of VertexAttrib*Format in section 10.3 to more closely match the OpenGL specification. • Change definition of API data types in section 2.2 and table 2.2 to require exact, rather than minimum bit widths. • Modify language in section 5.1.2 so that binding-related state is restored to default values after automatic unbinds. • Restructure description of queries for indexed buffer bindings in section 6.6.1 following GL spec, and remove redundant descriptions of these queries and related errors from sections 6.1.1, 7.6.3, 7.7.2, 7.8, and 12.2.2. • Add minor spec clarifications from OpenGL spec for ProgramParameteri and DeleteProgram in section 7.3, DeleteProgramPipelines and ActiveShaderProgram in section 7.4, GetUniformLocation in section 7.6, and Uniform* in section 7.6.1. • Add missing errors for TexParameter* (see section 8.10) and GetTexParameter* (see section 8.11).

OpenGL ES 3.2 (May 5, 2022)

E.2. CHANGE LOG FOR RELEASED SPECIFICATIONS

532

• Change formal parameter names for GetTexParameter* and GetTexLevelParameter* (see section 8.11) from value and data to pname and params, following the OpenGL headers and man pages. Change generated errors for GetTexLevelParameter* for consistency with other commands and with OpenGL. • Add subsection headings in section 8.11 and simplify active texture effects on queries by reference from section 8.11.1 to section 2.2.2. • Define behavior of GetTexLevelParameter* in section 8.11 for queries of multisample state from non-multisampled textures. • Change rounding mode for layer numbers of array textures in section 8.14.2 to prefer round-to-nearest-even, while still allowing old spec behavior. • Add description of DEPTH_STENCIL_TEXTURE_MODE in section 8.19, and correct its type in table 21.10. • Restructure error condition for FramebufferParameteri in section 9.2.1 to avoid ambiguity. • Define GetFramebufferAttachmentParameteriv in section 9.2.3 to return NONE when querying the object type of depth or stencil attachments, the default framebuffer is bound, and the corresponding buffer of the default framebuffer has zero bits. • Rearrange language describing integer handling in section 10.3 to differentiate between behaviors actually labelled in table 10.3 and sub-behaviors depending on the normalized argument. • Set the vertex attribute array pointer state explicitly in the pseudocode for VertexAttrib*Pointer in section 10.3.1, and remove VERTEX_BINDING_OFFSET from the vertex array object state which is looked up via the vertex attribute binding by GetVertexAttrib* in section 10.6. • Remove redundant non-local errors applying to indirect commands from section 10.3.9, as they are now described with each command. • Minor clarifications to descriptions of DrawArraysIndirect and DrawElementsIndirect in section 10.5. • Use instancecount as the formal parameter name for commands DrawArraysInstanced, DrawElementsInstanced, DrawElementsInstancedBaseVertex, and DrawElementsInstancedBaseVertex in section 10.5, instead of instanceCount or primCount, for consistency with OpenGL.

OpenGL ES 3.2 (May 5, 2022)

E.3. CREDITS AND ACKNOWLEDGEMENTS

533

• Add errors for DrawArraysIndirect and DrawElementsIndirect in section 10.5 when the default vertex array object is bound. • Clean up validation language in section 11.1.3.11 to more closely match the GL spec and remove inconsistencies about which active program objects are required. • Add missing language about stencil textures in section 14.2.1 (duplicated from vertex shader language). • Remove erroneous reference to “depth bounds test” from section 13.8. • Rewrite description of GetInternalformativ in section 20.3 to properly account for different limits on integer, depth, color, and other internal format samples. • Change error for invalid mode* parameters to BlendEquation* in section 15.1.5.1 to INVALID_ENUM. • Fix error for invalid blending function arguments in section 15.1.5.2 to INVALID_ENUM. • Replace Znumber type fields in state tables with E for enumerated state, following GL spec. • Change default value of SAMPLE_MASK_VALUE in table 21.8 to match GL spec and make it clear that all bits of each words are set. • Increased number of texture bindings from 32 to 48 in table 21.9.

E.3

Credits and Acknowledgements

OpenGL ES 3.1 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ES Working Group during the development of OpenGL ES 3.1, including the company that they represented at the time of their contributions, follow. Some major contributions made by individuals are listed together with their name. In addition, many people participated in developing desktop OpenGL specifications and extensions on which the OpenGL ES 3.1 functionality is based in large part; those individuals are listed in the respective specifications in the OpenGL Registry.

OpenGL ES 3.2 (May 5, 2022)

E.3. CREDITS AND ACKNOWLEDGEMENTS

Adrian Bucur, Samsung Alex Chalfin, AMD Alon Or-bach, Samsung Anssi Kalliolahti, NVIDIA Antti Tirronen, Qualcomm Aras Pranckevicius, Unity Ari Hirvonen, NVIDIA Barthold Lichtenbelt, NVIDIA Benj Lipchak, Apple Benji Bowman, Imagination Technologies Bill Licea-Kane, Qualcomm (framebuffer no attachments, shader atomic counters, shader image load/store, texture gather) Boguslaw Kowalik, Intel Bruce Merry Cass Everitt, NVIDIA Chris Dodd, NVIDIA Christophe Riccio, Unity Daniel Koch, NVIDIA (compute shader, program interface query, sample shading, shader bitfield operations, shader multisample interpolation) Dominik Witczak, Mobica Eric Boumaour, AMD Eric Werness, NVIDIA Evan Hart, AMD Fred Liao, Mediatek Graeme Leese, Broadcom (arrays of arrays) Graham Connor, Imagination Technologies Graham Sellers, AMD Greg Roth, NVIDIA (separate shader objects) Guangli Li, Marvell Ian Romanick, Intel Ian Stewart, NVIDIA James Helferty, NVIDIA Jan-Harald Fredriksen, ARM (vertex attrib binding) Janusz Sobczak, Mobica Jarkko Pöyry, drawElements Jason Green, Transgaming Jeff Bolz, NVIDIA Jeff Gilbert, Mozilla Jesse Hall, Google John Kessenich John Rosasco, Google Jon Leech (OpenGL ES API Specification editor)

534

Jonas Gustavsson, Sony Mobile Kalle Raita, drawElements Karol B Gasinski, Intel Kathleen Mattson, Miller and Mattson Kenneth Russell, Google Klaus Gerlicher, NVIDIA Krzysztof Kaminski, Intel Kulin Seth, Qualcomm Lijun Qu, AMD Mark Adams, NVIDIA Mark Callow, Artspark Mark Ellison, Mobica Mark Kilgard, NVIDIA Mathias Heyer, NVIDIA (texture STENCIL8 internal formats) Maurice Ribble, Qualcomm (stencil texturing) Members of the Khronos OpenGL ARB Working Group Michael Chock, NVIDIA Murat Balci, AMD Neil Trevett, NVIDIA Nick Haemel, NVIDIA Nick Hoath, Imagination Technologies (texture storage multisample) Nick Penwarden, Epic Games Pat Brown, NVIDIA (shader helper invocation, shader layout binding, shader storage buffer objects) Pierre Boudier, NVIDIA Piers Daniell, NVIDIA Piotr Czubak, Intel Pyry Haulos, drawElements Rik Cabanier, Adobe Rob Barris, NVIDIA Robert Simpson, Qualcomm (OpenGL ES Shading Language Specification editor) Robert Tray, NVIDIA Sean Ellis, ARM Season Li, NVIDIA Slawomir Cygan, Intel (explicit uniform location) Slawomir Grajewski, Intel Timo Suoranta, Broadcom (draw indirect) Tobias Hector, Imagination Technologies Tom Olson, ARM (Khronos OpenGL ES Working Group chair) Yanjun Zhang, Vivante

OpenGL ES 3.2 (May 5, 2022)

E.3. CREDITS AND ACKNOWLEDGEMENTS

535

The OpenGL ES Working Group gratefully acknowledges administrative support by the members of Gold Standard Group, including Andrew Riegel, Elizabeth Riegel, Glenn Fredericks, and Michelle Clark, and technical support from James Riordon, webmaster of Khronos.org and OpenGL.org.

OpenGL ES 3.2 (May 5, 2022)

Appendix F

Version 3.2 OpenGL ES version 3.2, released on August 10, 2015, is the fifth revision since the original version 1.0. OpenGL ES 3.2 is upward compatible with OpenGL ES version 3.1, meaning that any program that runs with an OpenGL ES 3.1 implementation will also run unchanged with an OpenGL ES 3.2 implementation. Following are brief descriptions of changes and additions to OpenGL ES 3.2.

F.1

New Features

New features in OpenGL ES 3.2 include: • Almost all features of the Android extension pack, incorporating by reference all of the following features - with the exception of the sRGB decode features of EXT_texture_sRGB_decode • Advanced blend equations. • Copying subregions between image objects • Supporting blending on a per-draw-buffer basis • Debug messages • Geometry shaders • Miscellaneous new shader functionality • ASTC texture compression (LDR profile only) • Primitive bounding boxes

536

F.1. NEW FEATURES

537

• Shader image atomic operations • Shader interface blocks • Shader multisample interpolation control • Sample shading control • Sample variables • Texture buffer objects • Texture border color • Texture cube map arrays • Tessellation shaders • STENCIL8 texture formats • Texture multisample 2D arrays and also include the following features, which are not part of the Android extension pack. • Draw calls specifying a base vertex parameter • Floating-point framebuffers • Enabling robust buffer access as described in GL_KHR_robustness (preventing crashes), but without the additional behavior defined by GL_KHR_robust_buffer_access

• Support for querying CONTEXT_FLAGS, as needed by debug and robust buffer access functionality. • Support

for querying MULTISAMPLE_LINE_WIDTH_RANGE and MULTISAMPLE_LINE_WIDTH_GRANULARITY (see section 13.6.4). Note that these are different query and enum values than desktop GL’s SMOOTH_LINE_WIDTH_*, which remain unsupported (Bug 13828).

OpenGL ES 3.2 (May 5, 2022)

F.2. CHANGE LOG DESCRIPTIONS

F.2

538

Change Log Descriptions

In the change logs below, a comment like (github #16) refers to a public Github issue on the https://github.com/KhronosGroup/OpenGL-API/ repository; (Public Bug 1432) refers to a public Bugzilla issue at https://www.khronos.org/bugzilla; (gitlab #40) refers to a private issue on Khronos’ internal servers; and (Bug 16315) refers to a private issue on Khronos’ internal servers.

F.3

Change Log for Released Specifications

Changes in the released Specification of May 5, 2022 • Remove

old query enums UNIFORM_BLOCK_REFERENCED_BY_TESS_CONTROL_SHADER, UNIFORM_BLOCK_REFERENCED_BY_TESS_EVALUATION_SHADER, and UNIFORM_BLOCK_REFERENCED_BY_GEOMETRY_SHADER from table 7.7. These old queries were never supported by OpenGL ES, instead using the program interface queries (gitlab #130).

• Update errors for TexImage3D in section 8.5.3 to validate internalformat separately from the combination of format, type, and internalformat (gitlab #147). • Update definitions of texture completeness in section 8.17 and of framebuffer completeness in section 9.4.1) to account for layered rendering to a cubemap and make the definitions more internally consistent (gitlab #141). • Remove redundant error for TexStorage3D in section 8.18 when target is TEXTURE_CUBE_MAP_ARRAY and depth is not a multiple of 6 (gitlab #135). • Add cube maps to allowed types of texture names for FramebufferTextureLayer in section 9.2.8 (gitlab #134). • Add alternate order for triangles generated by triangle strips with adjacency in table 10.2 (gitlab #150). • Update computation of element array buffer access for DrawElementsOneInstance in section 10.5 to offset in units of the size of the type being accessed (github #85). • Change the minimum value of MAX_FRAGMENT_INTERPOLATION_OFFSET to +0.5 − 1U LP in table 21.41 (gitlab #149).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

539

Changes in the released Specification of October 22, 2019 • Change the name of formal parameter bufSize to count for GetSynciv, GetInternalformativ, and GetProgramResourceiv (github #52 / #285). • Remove query objects from the list of shareable container objects in section 5 (github #60). • Fix typo in section 7.3.1.1 (github #55). • Reorganize description of state restored by ProgramBinary in section 7.5 (gitlab #120). • Specify in section 7.5 that the state of program parameter PROGRAM_SEPARABLE is restored when ProgramBinary is called successfully (gitlab #114). • Fix typo in pseudocode for GetUniformIndices in section 7.6 (github #54). • Note in section 7.6.3 that the initial value of uniform buffer object bindings is specified by the layout qualifier, if present, matching OpenGL specification behavior (github #56). • Specify that shader storage block size is restricted to the fixed size portion reported by BUFFER_DATA_SIZE in section 7.8 (github #36). • Add atomic operations on atomic counter variables to the description of ways for shaders to access buffer object memory in section 7.11 (github #14). • Clarify memory barrier requirements in section 7.11.1 (gitlab #95). • Clarify completeness requirements and interactions with multisampling for integer format textures, and specify that multisample textures never require mipmaps in section 8.17 (github #45). • Fix typo in section 11.2.2.2 (gitlab #113). • Loosen the line interpolation requirements in section 13.6.2.1 to properly handle gl_FragCoord and allow diagonal interpolation of input attributes, bring the specification in line with Vulkan and shipping hardware (gitlab #116). • Move MAX_SAMPLES from framebuffer dependent state in table 21.56, into implementation dependent state in table 21.41 (github #58)

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

540

Changes in the released Specification of February 2, 2019 • Fix “builtin” → “built-in” in a few places. • Use the term “workgroup” consistently instead of “work group” (internal issue #87 - just a typo, so not marked everywhere the change was done). • Use explicit errors for negative sizei parameters everywhere for consistency, even though there’s a generic error described in section 2.3.1 (gitlab #52). • Modify description of conversion behavior for normalized floating-point values in sections 2.3.5.1 and 8.5 so that they are not clamped on upload (Bug 13802). • Clarify enumeration of block members for arrays of uniform and shader storage buffer objects in section 7.3.1.1 (gitlab #94). • Clarify shader interface matching in section 7.4.1 so that the built-in interface blocks between vertex shaders and downstream shaders match (Bug 16124). • Generate an INVALID_OPERATION error for GetProgramBinary in section 7.5 if there are no defined binary formats (Bug 16155). • Clean up definition of which uniform blocks count against combined limits in section 7.6.2 (gitlab #69). • Allow “sensible” conversions in section 8.6 for CopyTexImage2D (Bug 12220) • Correct typo CompressedTexImage3D to CompressedTexSubImage3D in the errors section of CompressedTexSubImage3D in section 8.7 (gitlab #81). • Clarify language in section 9.2.3 about valid GetFramebufferAttachmentParameteriv queries when the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE is NONE (Bug 16198). • Specify in section 11.1.3.4 that textureSize returns undefined values when the texture is incomplete (Bug 11750). • Modified BeginTransformFeedback language about simultaneous use of buffers for transform feedback and other operations in section 12.2.2 to ensure this behavior is undefined (gitlab #83).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

541

• Specify std430 as a base layout for compute shader uniform storage requirements in section 17.1 (github #40). Changes in the released Specification of May 14, 2018 • Fix two typos (github #31). • Clarify undefined GL behavior with no current context in section 2.1 (Bug 11522). • Specify in section 2.3.1 that detection of errors based on the value of parameters is always performed prior to any state-based validation (Bug 11046). • Add a footnote to the description of Flush in section 2.3.3 explaining historical reasons for its inconsistent behavior (Bug 13916). • Note in section 2.6.6 that texture objects contain a data store. Change the feedback loop language in section 9.3 to refer to the data store rather than the texture object (Public Bug 11934). • Clarify description of the initial state of query objects for BeginQueryIndexed in section 4.2, and change the initial QUERY_RESULT_AVAILABLE state for query objects to TRUE in table 21.31 (gitlab #6, also Public Bug 1213 and Bug 16089). • Drop now-irrelevant language in section 6.1 saying that the GL may make different choices about storage location and layout based on the initial binding (github #23). • Fixed typo in an error for the shaders parameter of ShaderBinary in section 7.2 (Public Bug 1421). • Specify that only a single integer is written for several resources already described as queriable with GetProgramResourceiv in section 7.3.1.1 (github #35). • Clarify in section 7.4.1 that output blocks that match in all ways except their name are not guaranteed to match (Bug 15338). • Add clarification to std140 layout rules in section 7.6.2.2 that alignment of structures not containing types larger than a vec4 is just vec4 (Bug 10710). • Specify in section 7.7.1 that the value of UNIFORM_ARRAY_STRIDE) for atomic_uint arrays is always 4 (gitlab #31).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

542

• Added description of storage block activeness in 7.8, based on that of uniform blocks (Bug 13745). • Clarify in section 7.12 that there is always an info log for shader, program, and program pipeline objects, even if it is empty; and that the strings returned by GetShaderInfoLog, GetProgramInfoLog and GetProgramPipelineInfoLog are always null-terminated (Bug 16110). • Use the term “work group size” consistently in section 7.12 and chapter 17, instead of a mix of similar terms usually starting with “local” (Bug 11723). • Note that the implementation-dependent maximum texture dimensions are independent in each image dimension, and that implementations may fail to create textures of the maximum sizes due to resource exhaustion, in section 8.5.3. Reorganize this section to match the corresponding section of the OpenGL 4.5 Specification (Bug 12249). • Fix error for CompressedTexImage3D in section 8.7 to correctly reference the target parameter instead of internalformat (Public Bug 1440). • Fix reference to TexImage2DMultisample in section 8.8 to TexStorage2DMultisample (github #27). • Change the name of the sizedinternalformat parameter of the TexStorage*DMultisample commands in section 8.8 to internalformat, and the type to enum, to match the XML, headers, and OpenGL Specification (github #30). • Fix typo (missing comma) in prototype for TextureParameterI* in section 8.10 (Public Bug 1432). • Adjust the level-of-detail computation of f (x, y) in section 8.14.1 to consider other vectors when computing scale factors (Bug 16179). • Clean up level-of-detail computation in equation 8.11 section 8.14.3 to relax computation to match the Vulkan specification (Bug 16233). • Specify in section 8.14.4 that GenerateMipmap has no effect when the base mipmap level is undefined, or when any levelbase dimension is zero (Bug 11147, gitlab #72). • Add a requirement to the GenerateMipmap command in section 8.14.4 that any synchronization required before performing mipmap reduction will be done within the commands themselves (Public Bug 1429).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

543

• Add error for TexStorage3D in section 8.18 when target is TEXTURE_CUBE_MAP_ARRAY and depth is not a multiple of 6 (Public Bug 1439). • Describe the behavior of GetFramebufferParameteriv in section 9.2.3 when no framebuffer, or the default framebuffer is bound, by using equivalent but easier to understand language from the OpenGL Specification (gitlab #67). • Remove FramebufferTexture2D error for nonexistent layer parameter in section 9.2.8, that was accidentally left in when porting from OpenGL specification language (github #28). • Clarify in section 9.3.1 that even if writing to a given texture is disabled during rendering, a feedback loop will still occur (gitlab #46). • Remove assertions that OpenGL ES Shading Language does not support multi-dimensional arrays in sections 11.2.1.2.3 and 11.3.4.3 (gitlab #62). • Use new terminology for the “last active vertex processing stage” and “when a vertex processing stage is active”, rather than referring to “vertex shader outputs” and “when a vertex shader is active”, respectively. This allows the same language to apply to outputs from the shading pipeline coming from any of vertex, tessellation evaluation, or geometry shaders. These changes are made in a new section 12.1 and in sections 12.2, 12.2.2, 12.4, 12.5, 12.5.1, and 13.5 (github #9). • Add a footnote to the description of BindTransformFeedback in section 12.2.1 explaining that the generic bind point should not be affected by this command, and move TRANSFORM_FEEDBACK_BUFFER_BINDING generic binding state from the transform feedback object state table 21.35 to the context transformation state table 21.6 (gitlab #66). • Simplify description of per-sample shading in section 14.2.2 and clarify the requirements on the interaction between gl_SampleMaskIn and MinSampleShading (gitlab #45). • Clarify occlusion query description in section 15.1.4 to refer to fragments that are not discarded by earlier stages, rather than fragments that pass the depth test (github #19). • Add missing error for BlendFunc*i in section 15.1.5.2 to ensure buf is a valid draw buffer (Public Bug 1416).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

544

• Specify that the HSL blend equations in section 15.1.5.3 are only well defined for input color components in the range [0, 1], and update the pseudocode example to match the KHR extension (gitlab #60, Bug 16026). • Don’t require that textures passed to CopyImageSubData in section 16.2.2 be mipmap complete, by ignoring format-based completeness rules (Bug 16224). • Fix error for DispatchComputeIndirect in section 17 to refer to DISPATCH_INDIRECT_BUFFER instead of DRAW_INDIRECT_BUFFER (gitlab #41). • Clarify in appendix F.1 that the “robust buffer access” added by OpenGL ES 3.2 includes GL_KHR_robustness, but not GL_KHR_robust_buffer_access (gitlab #49). • Update backward compatibility discussion in appendix G.2 to refer to the current ES version (3.2) instead of 3.1 (Public Bug 1414). Changes in the released Specification of November 3, 2016: • Create new list of valid query targets in table 4.2, and refer to it from the errors section of BeginQuery and EndQuery in section 4.2, and GetQueryiv in section 4.2.1 (Bug 15979). • Document MapBufferRange behavior when MAP_UNSYNCHRONIZED_BIT is used in conjunction with either MAP_INVALIDATE_BUFFER_BIT or MAP_INVALIDATE_RANGE_BIT in section 6.3 (Bug 15631). • Document the error generated when loading image uniforms with Uniform* in section 7.6.1 (Bug 11443). • Allow either implementation-dependent conversion from texture source internal format to destination effective internal format for *TexSubImage* operations, or errors, in sections 8.5 and 8.6 (Bug 7110). • Clarify in sections 9.2.4 and 20.3 that the internalformat argument of RenderbufferStorageMultisample and GetInternalformativ, respectively, must be sized, (Bug 15912). • Add constraint on level for FramebufferTexture2D and FramebufferTextureLayer in section 9.2.8, and restructured the complete set of constraints (Bug 15946).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

545

• Fix constraint on layer for FramebufferTextureLayer in section 9.2.8 (Bug 15968). • Clarify behavior when a fragment shader has fewer outputs than framebuffer color attachments in section 14.2.3 and for DrawBuffers in section 15.2.1 (Bug 13825). • Attempt to rotate landscape-format state table pages when being displayed this is dependent on the PDF reader (Bug 11976). Changes in the released Specification of June 15, 2016: • Specify that queries returning unsigned integers will clamp negative state values in section 2.2.2 (Bug 14444). • Add transform feedback objects to the list of objects whose deletion causes automatic unbinding in section 5.1.2, and described the unbinding behavior for DeleteTransformFeedbacks in section 12.2.1 (Bug 14375). • Use consistent phrasing of “has/has been linked successfully” to describe the link status of programs where relevant in the descriptions of conditions for LinkProgram and UseProgram in section 7.3, GetProgramResourceLocation* in section 7.3.1.1, UseProgramStages and ActiveShaderProgram in section 7.4, GetProgramBinary in section 7.5, ProgramUniform* in section 7.6.1, GetProgramiv in section 7.12, GetAttribLocation in section 11.1.1, and GetFragDataLocation and GetFragDataIndex in section 14.2.3. This replaces a variety of previous usages including the rather wordy “has not been linked, or was linked unsuccessfully” (Bug 8640). • Add language to GetProgramResourceiv in section 7.3.1.1 specifying the returned block index when the interface block is declared as an array of block instances, and clarify the definition of active uniform blocks in section 7.6 (Bug 11938). • Add tessellation and geometry shader bits to valid bits for the UseProgramStages stages parameter in section 7.4. • Clarify shader interface matching rules in section 7.4.1 so that there is no match in the case where name, type and qualification match, but one variable has a location qualifier and the other does not (Bug 13613).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

546

• Add clarification on the method for specifying more than 32 texture units to ActiveTexture in section 8 (Bug 15774). • Cleanup description of levelbase and levelmax in sections 8.10 and 8.14.3 (Bug 15813). • Require cube map array texture support for ETC2 format textures in table 8.17 (Bug 14713). • Add image size operations to the description of format mismatch behavior for BindImageTexture in section 8.23 (Bug 15531). • Do not allow multisample textures to be bound as images by BindImageTexture in section 8.23 (Bug 15818). • Specify that the COLOR_ATTACHMENTm tokens are defined for m in the range [0, 31] in section 9.2, and specify that INVALID_OPERATION errors are generated for valid COLOR_ATTACHMENTm tokens where m is outside the range of valid color attachments for the commands GetFramebufferAttachmentParameteriv (section 9.2.3), FramebufferRenderbuffer (section 9.2.7), FramebufferTexture and FramebufferTexture2D (section 9.2.8), DrawBuffers (section 15.2.1), InvalidateSubFramebuffer (section 15.2.4), and ReadBuffer (section 16.1.1 (Bug 13858) • Add an error when an invalid target argument to BindRenderbuffer is passed, in section 9.2.4 (Bug 14283). • Specify that BindVertexBuffer can accept existing buffer object names in section 10.3.1 (Bug 15252). • Change incorrect reference of “atomic counters” to “image units” in section 11.1.3.7. • Remove description of OpenGL ES Shading Language as not supporting multi-dimensional arrays as shader inputs and outputs from sections 11.2.1.2 and 11.2.3.3 (Bug 13824). • Specify in section 11.3.1 that the mode parameter used for validating geometry shaders depends on whether or not a tessellation evaluation shader is active (Bug 14141). • Remove unusable bool and bvec* types from table 12.2 (Bug 15023). • Remove “or has been re-linked” clause from the non-local error for ResumeTransformFeedback defined in section 12.2.2 (Bugs 15414, 15415).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

547

• Update description of polygon associated data interpolation for flat shading in section 13.7.1 (Bug 15785). • Update the description of gl_SampleMaskIn in section 14.2.2 and gl_SampleMask in section 14.2.3 to refer to the maximum number of samples supported for any renderable internal format, and make corresponding changes to the definitions of the MAX_*_SAMPLES queries in table 21.41 and MAX_SAMPLES in table 21.56 (Bug 15122). • Remove references to signed normalized color buffers in section 15.1.5 (Bug 14364). • Restore ability of CopyImageSubData in section 16.2 to copy between nonoverlapping regions of the same buffer, while still not allowing BlitFramebuffer to blit between the same source and destination framebuffer (Bug 15739). • Specify in sections 16.2 and 16.2.1 that pixel copy operations generate an error when attempting to copy between identical source and destination buffers (Bug 15739). This behavior was consciously chosen in OpenGL ES 3.0 and differs from OpenGL, which allows such copies and makes them welldefined if the source and destination regions do not overlap. • Add an INVALID_OPERATION error for CopyImageSubData in section 16.2.2 when the source and destination buffers are identical, to match BlitFramebuffer (Bug 15739). • Reduce minimum value of MAX_TESS_CONTROL_TOTAL_OUTPUT_COMPONENTS from 4096 to 2048 in table 21.45 (Bug 12823). • Rewrite appendix C to refer to the Khronos Data Format Specification for the definition of formats (Bug 15253). • Specify that only the ASTC LDR Profile modes are supported in section C.1 (Feedback from Jan-Harald). Changes in the released Specification of August 10, 2015: • Added new features as described in section F.1. • Changed name of formal parameter of SamplerParameter*v in section 8.2 to params (Bug 14158).

OpenGL ES 3.2 (May 5, 2022)

F.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

548

• Add multisample texture targets to those supporting depth and stencil texture formats in section 8.5 (Bug 14158). • Drop bogus reference to PROXY texture target in section 8.5.3 (Bug 14183). • Correct command name from TextureParameterI*v to TexParameterI*v in section 8.10 (Bug 14158). • Add missing parameter range error for TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL pnames to TexParameter* in section 8.10 (Bug 14157). • Add missing parameter validation errors for FramebufferTexture2D in section 9.2.8 (Bug 14157). • Expand language about interpolated outputs in section 11.1.2.1 to match GL spec (Bug 14158). • Clarify that the primitive ID counters for tessellation control, geometry, and fragment shaders are reset to zero after each instance drawn, in sections 11.2.1.2, 11.3.4.3, and 14.2.2 (Bug 14024). • Remove redundant non-local errors for mode validation in drawing commands in section 12.2.2 (now subsumed by table 12.1), and for feedback buffer overflow detection (very difficult when using geometry shaders) (Bug 14158). • Introduce upstream shader terminology for transform feedback, and correct description of when variables written by transform feedback are undefined in section 12.2.2, to account for both geometry and tessellation shader stages (Bug 14157). • Fix language in section 13.5 to indicate that only the vertex shader can write gl_PointSize (Bug 14157). • Change require minimum value from 128 to 64 for MAX_TESS_CONTROL_OUTPUT_COMPONENTS and MAX_TESS_CONTROL_INPUT_COMPONENTS in table 21.45, MAX_TESS_EVALUATION_OUTPUT_COMPONENTS and MAX_TESS_EVALUATION_INPUT_COMPONENTS in table 21.46, and MAX_GEOMETRY_OUTPUT_COMPONENTS in table 21.47 (Bug 12823). • Fix minimum values for MAX_FRAGMENT_ATOMIC_COUNTER_BUFFERS, MAX_FRAGMENT_ATOMIC_COUNTERS and MAX_FRAGMENT_SHADER_STORAGE_BLOCKS in table 21.48, and MAX_FRAGMENT_IMAGE_UNIFORMS in table 21.52 (Bug 14157).

OpenGL ES 3.2 (May 5, 2022)

F.4. CREDITS AND ACKNOWLEDGEMENTS

F.4

549

Credits and Acknowledgements

OpenGL ES 3.2 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ES Working Group during the development of OpenGL ES 3.2, including the company that they represented at the time of their contributions, follow. Some major contributions made by individuals are listed together with their name. In addition, many people participated in developing desktop OpenGL specifications and extensions on which the OpenGL ES 3.2 functionality is based in large part; those individuals are listed in the respective specifications in the OpenGL Registry. Alon Or-bach, Samsung Andrew Garrard, Samsung Andrew Garrard, Samsung Antti Tirronen, Qualcomm Barthold Lichtenbelt, NVIDIA Benji Bowman, Imagination Technologies Bill Licea-Kane, Qualcomm Cass Everitt, Oculus Christophe Riccio, Unity Christopher Gautier, ARM Daniel Kartch, NVIDIA Daniel Koch, NVIDIA (Detailed specification review) Dominik Witczak, Samsung Etay Meiri, Intel Fred Liao, Mediatek Geoff Lang, Google Graeme Leese, Broadcom (Detailed specification review) Graham Connor, Imagination Technologies Graham Sellers, AMD Greg Roth, NVIDIA Ian Romanick, Intel Iliyan Dinev, Imagination Technologies James Jones, NVIDIA Jan-Harald Fredriksen, ARM (Detailed specification review) Jarkko Pöyry, Google Jeff Bolz, NVIDIA Jeff Gilbert, Mozilla Jesse Hall, Google John Kessenich Jon Leech (OpenGL ES API Specification editor)

Jonas Gustavsson, Sony Mobile Jonathan Putsman, Imagination Technologies Jun Wang, Imagination Technologies Kalle Raita, Google Kathleen Mattson, Miller & Mattson Mark Adams, NVIDIA Mark Callow Mathias Heyer, NVIDIA (Detailed specification review) Maurice Ribble, Qualcomm Neil Trevett, NVIDIA Olli Etuaho, NVIDIA Pat Brown, NVIDIA Patrick Doane, Blizzard Peter Deayton, Mediatek Piers Daniell, NVIDIA Pyry Haulos, Google Robert Simpson, Qualcomm (OpenGL ES Shading Language Specification editor) Robert Stepinski, Transgaming Sawato Shusaku, DMP Inc. Shannon Woods, Google Slawomir Cygan, Intel Slawomir Grajewski, Intel Steve Hill, Broadcom Tobias Hector, Imagination Technologies Tom Olson, ARM (Khronos OpenGL ES Working Group chair) Wayne Lister, Imagination Technologies Weifeng Zhang, Qualcomm Xi Ma Chen, NVIDIA Yanjun Zhang, Vivante Zeppelin Zhu, Vivante

OpenGL ES 3.2 (May 5, 2022)

F.4. CREDITS AND ACKNOWLEDGEMENTS

550

The OpenGL ES Working Group gratefully acknowledges administrative support by the members of Gold Standard Group, including Andrew Riegel, Elizabeth Riegel, Emily Stearns, Glenn Fredericks, and Michelle Clark, and technical support from James Riordon, webmaster of Khronos.org and OpenGL.org.

OpenGL ES 3.2 (May 5, 2022)

Appendix G

Backwards Compatibility The OpenGL ES 3.2 API is backward compatible with OpenGL ES 2.0. It accepts all of the same commands and their arguments, including the same token values. This appendix describes OpenGL ES 3.2 features that were carried forward from OpenGL ES 2.0 solely to maintain backward compatibility as well as those that have changed in behavior relative to OpenGL ES 2.0.

G.1

Legacy Features

The following features are present to maintain backward compatibility with OpenGL ES 2.0, but their use is not recommended as it is likely for these features to be removed in a future version. • Fixed-point (16.16) vertex attributes • Application-chosen object names (those not generated via Gen* or Create*) • Client-side vertex arrays (those not stored in buffer objects) • Luminance, alpha, and luminance alpha formats • Queryable shader range and precision (GetShaderPrecisionFormat) • Old-style non-indexed extensions query • Vector-wise uniform limits • Default vertex array object

551

G.2. DIFFERENCES IN RUNTIME BEHAVIOR

G.2

552

Differences in Runtime Behavior

The following behaviors are different in OpenGL ES 3.2 than they were in OpenGL ES 2.0. • OpenGL ES 3.2 requires that all cube map filtering be seamless. OpenGL ES 2.0 specified that a single cube map face be selected and used for filtering. See section 8.13.1. • OpenGL ES 3.2 specifies a zero-preserving mapping when converting back and forth between signed normalized fixed-point values and floating-point values. OpenGL ES 2.0 specified a mapping by which zeros are not preserved. See section 2.3.5. • OpenGL ES 3.2 requires that framebuffer objects not be shared between contexts. OpenGL ES 2.0 left it undefined whether framebuffer objects could be shared. See chapter 5.

OpenGL ES 3.2 (May 5, 2022)

Index x BITS, 502 ACTIVE ATOMIC COUNTER BUFFERS, 130, 472 ACTIVE ATTRIBUTE MAX LENGTH, 128, 469 ACTIVE ATTRIBUTES, 128, 468 ACTIVE PROGRAM, 131, 466 ACTIVE RESOURCES, 83, 473 ACTIVE TEXTURE, 13, 138, 140, 191, 455 ACTIVE UNIFORM BLOCK MAX NAME LENGTH, 129, 470 ACTIVE UNIFORM BLOCKS, 129, 470 ACTIVE UNIFORM MAX LENGTH, 129, 468 ACTIVE UNIFORMS, 128, 468 ACTIVE VARIABLES, 86, 87, 108, 474 ActiveShaderProgram, 97, 108, 531, 545 ActiveTexture, 119, 138, 546 ALIASED LINE WIDTH RANGE, 358, 486 ALIASED POINT SIZE RANGE, 356, 486 ALL BARRIER BITS, 125, 127 ALL SHADER BITS, 96 ALPHA, 149, 151, 159, 160, 169, 172, 189, 212, 373, 374, 388, 456,

457, 462, 502 ALPHA BITS, 253 ALREADY SIGNALED, 35 ALWAYS, 189, 213, 382, 383, 459 ANY SAMPLES PASSED, 39–42, 383, 384 ANY SAMPLES PASSED CONSERVATIVE, 39–42, 383, 384 ARRAY BUFFER, 52, 270, 271, 274 ARRAY BUFFER BINDING, 274, 450 ARRAY SIZE, 86, 87, 106, 107, 291, 296, 474 ARRAY STRIDE, 86, 87, 107, 113, 474 ATOMIC COUNTER BARRIER BIT, 125, 126 ATOMIC COUNTER BUFFER, 52, 53, 79, 82, 84–86, 107, 117 ATOMIC COUNTER BUFFER BINDING, 64, 478 ATOMIC COUNTER BUFFER INDEX, 86, 87, 474 ATOMIC COUNTER BUFFER SIZE, 64, 478 ATOMIC COUNTER BUFFER START, 64, 478 atomic uint, 92, 116, 541 atomicCounter, 509 atomicCounterDecrement, 509 atomicCounterIncrement, 509

553

INDEX

554

ATTACHED SHADERS, 128, 131, 467 blend support all equations, 394 AttachShader, 71 BlendBarrier, 393 BlendColor, 387, 394 BACK, 230, 233, 364, 382, 397–400, BlendEquation, 385, 385 402, 406, 407, 453 BlendEquationi, 385, 385 barrier, 313, 314 BlendEquationSeparate, 385, 385 BeginQuery, 39, 40, 40–42, 344, 383, BlendEquationSeparatei, 385, 385 529, 544 BlendFunc, 387, 388 BeginQueryIndexed, 541 BlendFunci, 387, 388 BeginTransformFeedback, 339, 339– BlendFuncSeparate, 387, 388 341, 343, 540 BlendFuncSeparatei, 388, 388 BindAttribLocation, 101, 289, 289, 291, BlitFramebuffer, 24, 354, 393, 406, 412, 519, 530 413, 520–522, 547 BindBuffer, 26, 50, 51, 51, 53, 188, 268, BLOCK INDEX, 86, 87, 107, 474 269, 275 BLUE, 189, 212, 373, 456, 457, 462, BindBufferBase, 53, 53, 63, 344, 522 502 BindBufferRange, 45, 53, 53, 54, 63, BLUE BITS, 253 115, 117, 119, 344, 519, 522 BOOL, 90 BindFramebuffer, 224, 225, 227, 250 bool, 90, 112, 546 BindImageTexture, 45, 215, 216, 217, BOOL VEC2, 90 221, 255, 546 BOOL VEC3, 90 binding, 119 BOOL VEC4, 90 BindProgramPipeline, 76, 95, 95–97, boolean, 110 131, 307, 343 BUFFER, 432 BindRenderbuffer, 234, 234, 235, 546 BUFFER ACCESS FLAGS, 52, 55, 58, BindSampler, 26, 141, 141, 143 60, 451 BindTexture, 119, 138, 139, 139, 140 BUFFER BINDING, 86, 88, 108, 474 BindTransformFeedback, 337, 338, BUFFER DATA SIZE, 86, 88, 108, 338, 543 117, 118, 474, 539 BindVertexArray, 277, 277 BUFFER MAP LENGTH, 52, 55, 58, BindVertexBuffer, 268, 269, 275, 546 60, 451 BLEND, 384, 385, 394, 459 BUFFER MAP OFFSET, 52, 55, 58, BLEND COLOR, 459 60, 451 BLEND DST ALPHA, 459 BUFFER MAP POINTER, 52, 55, 58, BLEND DST RGB, 459 60, 62, 63, 451 BLEND EQUATION ALPHA, 459 BUFFER MAPPED, 52, 55, 58, 60, 451 BLEND EQUATION RGB, 459 BUFFER SIZE, 52, 55, 56, 59, 119, BLEND SRC ALPHA, 459 185, 186, 451 BLEND SRC RGB, 459 BUFFER UPDATE BARRIER BIT, blend support, 393 124

OpenGL ES 3.2 (May 5, 2022)

INDEX

555

BUFFER USAGE, 52, 55, 57, 451 BUFFER VARIABLE, 80, 86, 87, 89 BufferData, 47, 54, 55, 60 BufferSubData, 47, 56, 123, 126 bvec*, 546 bvec2, 90, 109 bvec3, 90 bvec4, 90 BYTE, 147, 148, 150, 221, 267, 270, 411 callback, 427 CCW, 129, 363, 364, 453, 472 ccw, 316 centroid, 372 centroid in, 372 CheckFramebufferStatus, 250, 251, 252 CLAMP TO BORDER, 189, 195, 198 CLAMP TO EDGE, 189, 194, 198, 413 Clear, 24, 255, 352, 400, 401, 403 ClearBuffer{if ui}v, 402 ClearBufferfi, 402, 402, 403 ClearBufferfv, 402, 403 ClearBufferiv, 402, 403 ClearBufferuiv, 402, 403 ClearColor, 401, 402 ClearDepth, 402, 522 ClearDepthf, 401 ClearStencil, 401, 402 ClientWaitSync, 33, 34, 35, 35–37, 45 coherent, 125 COLOR, 402, 403, 405 COLOR ATTACHMENTm, 546 COLOR ATTACHMENT0, 226, 399, 406 COLOR BUFFER BIT, 401, 403, 413, 414 COLOR CLEAR VALUE, 460 COLOR WRITEMASK, 399, 460 COLORBURN, 391

COLORDODGE, 391 ColorMask, 399, 399, 400 ColorMaski, 399, 399 COMMAND BARRIER BIT, 124 COMPARE REF TO TEXTURE, 189, 213 COMPILE STATUS, 69, 71, 78, 127, 128, 465 CompileShader, 69, 69, 377 COMPRESSED R11 EAC, 178, 417, 418, 515 COMPRESSED RG11 EAC, 178, 417, 418, 515 COMPRESSED RGB8 ETC2, 178, 417, 418, 515 COMPRESSED RGB8 PUNCHTHROUGH ALPHA1 ETC2, 178, 417, 418, 515 COMPRESSED RGBA8 ETC2 EAC, 178, 417, 418, 515 COMPRESSED RGBA ASTC 10x10, 178, 418, 514 COMPRESSED RGBA ASTC 10x5, 178, 418, 514 COMPRESSED RGBA ASTC 10x6, 178, 418, 514 COMPRESSED RGBA ASTC 10x8, 178, 418, 514 COMPRESSED RGBA ASTC 12x10, 178, 419, 514 COMPRESSED RGBA ASTC 12x12, 178, 419, 514 COMPRESSED RGBA ASTC 4x4, 178, 418, 514 COMPRESSED RGBA ASTC 5x4, 178, 418, 514 COMPRESSED RGBA ASTC 5x5, 178, 418, 514 COMPRESSED RGBA ASTC 6x5,

OpenGL ES 3.2 (May 5, 2022)

INDEX

556

178, 418, 514 COMPRESSED RGBA ASTC 6x6, 178, 418, 514 COMPRESSED RGBA ASTC 8x5, 178, 418, 514 COMPRESSED RGBA ASTC 8x6, 178, 418, 514 COMPRESSED RGBA ASTC 8x8, 178, 418, 514 COMPRESSED SIGNED R11 EAC, 178, 417, 418, 515 COMPRESSED SIGNED RG11 EAC, 178, 417, 418, 515 COMPRESSED SRGB8 ALPHA8 ASTC 10x10, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 10x5, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 10x6, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 10x8, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 12x10, 179, 214, 419, 514 COMPRESSED SRGB8 ALPHA8 ASTC 12x12, 179, 214, 419, 514 COMPRESSED SRGB8 ALPHA8 ASTC 4x4, 178, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 5x4, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 5x5, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 6x5, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 -

ASTC 6x6, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 8x5, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 8x6, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ASTC 8x8, 179, 214, 418, 514 COMPRESSED SRGB8 ALPHA8 ETC2 EAC, 178, 214, 417, 418, 515 COMPRESSED SRGB8 ETC2, 178, 214, 417, 418, 515 COMPRESSED SRGB8 PUNCHTHROUGH ALPHA1 ETC2, 178, 214, 417, 418, 515 COMPRESSED TEXTURE FORMATS, 177, 488 CompressedTexImage, 182 CompressedTexImage2D, 177, 180, 181 CompressedTexImage3D, 177, 179– 181, 540, 542 CompressedTexSubImage2D, 181, 181, 182 CompressedTexSubImage3D, 181, 181, 182, 540 COMPUTE SHADER, 68, 421, 466 COMPUTE SHADER BIT, 96 COMPUTE WORK GROUP SIZE, 129, 130, 422, 467 CONDITION SATISFIED, 35 CONSTANT ALPHA, 388 CONSTANT COLOR, 388 CONTEXT FLAG DEBUG BIT, 424, 442 CONTEXT FLAG ROBUST ACCESS BIT, 442 CONTEXT FLAGS, 273, 424, 442, 489, 537 CONTEXT LOST, 14, 15, 17

OpenGL ES 3.2 (May 5, 2022)

INDEX

557

COPY READ BUFFER, 52, 61 COPY READ BUFFER BINDING, 503 COPY WRITE BUFFER, 52, 61 COPY WRITE BUFFER BINDING, 503 CopyBufferSubData, 61 CopyImageSubData, 415, 415–419, 544, 547 CopyTexImage, 170 CopyTexImage2D, 167, 174, 176, 177, 202, 529, 540 CopyTexImage3D, 174 CopyTexSubImage2D, 174, 174, 176, 177 CopyTexSubImage3D, 174, 174–177 Create*, 551 CreateProgram, 26, 71 CreateShader, 67, 68 CreateShaderProgramv, 77, 77, 78 CULL FACE, 364, 453 CULL FACE MODE, 453 CullFace, 364, 364, 367 CURRENT PROGRAM, 467 CURRENT QUERY, 42, 503 CURRENT VERTEX ATTRIB, 285, 476 CW, 129, 363, 364 cw, 316

DEBUG NEXT LOGGED MESSAGE LENGTH, 428, 499 DEBUG OUTPUT, 424, 428–430, 499 DEBUG OUTPUT SYNCHRONOUS, 433, 434, 499 DEBUG SEVERITY HIGH, 426 DEBUG SEVERITY LOW, 426, 427 DEBUG SEVERITY MEDIUM, 426 DEBUG SEVERITY NOTIFICATION, 426, 431 DEBUG SOURCE API, 16, 425 DEBUG SOURCE APPLICATION, 425, 430, 431 DEBUG SOURCE OTHER, 425 DEBUG SOURCE SHADER COMPILER, 425 DEBUG SOURCE THIRD PARTY, 425, 430, 431 DEBUG SOURCE WINDOW SYSTEM, 425 DEBUG TYPE DEPRECATED BEHAVIOR, 426 DEBUG TYPE ERROR, 16, 426 DEBUG TYPE MARKER, 426 DEBUG TYPE OTHER, 426 DEBUG TYPE PERFORMANCE, 426 DEBUG TYPE POP GROUP, 426, 431 DEBUG TYPE PORTABILITY, 426 DEBUG TYPE PUSH GROUP, 426, DARKEN, 391 431 DEBUG CALLBACK FUNCTION, DEBUG TYPE UNDEFINED BE428, 434, 441, 499 HAVIOR, 426 DEBUG CALLBACK USER DebugMessageCallback, 427, 428, 434 PARAM, 434, 441, 499 DebugMessageControl, 427, 429 DEBUG GROUP STACK DEPTH, DebugMessageInsert, 430, 430 499 DECR, 382 DEBUG LOGGED MESSAGES, 428, DECR WRAP, 382 499 DELETE STATUS, 70, 127, 128, 465, 467

OpenGL ES 3.2 (May 5, 2022)

INDEX

558

DeleteBuffers, 26, 45, 50, 51, 269 DeleteFramebuffers, 227 DeleteProgram, 77, 77, 531 DeleteProgramPipelines, 94, 95, 97, 131, 307, 531 DeleteQueries, 40, 41 DeleteRenderbuffers, 45, 235, 251 DeleteSamplers, 142, 143 DeleteShader, 70, 70 DeleteSync, 34, 35, 38 DeleteTextures, 45, 140, 216, 251 DeleteTransformFeedbacks, 45, 337, 338, 545 DeleteVertexArrays, 276, 277 DEPTH, 230, 231, 233, 402, 403, 405, 457, 462 DEPTH24 STENCIL8, 148, 164 DEPTH32F STENCIL8, 148, 164 DEPTH ATTACHMENT, 226, 239, 249, 404 DEPTH BITS, 253, 502 DEPTH BUFFER BIT, 401, 403, 413– 415 DEPTH CLEAR VALUE, 460, 530 DEPTH COMPONENT, 148, 151, 158, 159, 164, 189, 212, 248, 301, 302, 408, 456 DEPTH COMPONENT16, 148, 164 DEPTH COMPONENT24, 148, 164 DEPTH COMPONENT32F, 148, 164 DEPTH FUNC, 459 DEPTH RANGE, 452 DEPTH STENCIL, 148, 151, 154, 156–159, 164, 205, 206, 212, 239, 245, 248, 301, 302, 402– 404, 408 DEPTH STENCIL ATTACHMENT, 231, 233, 239, 245, 404, 521 DEPTH STENCIL TEXTURE MODE, 189, 205,

206, 212, 302, 456, 532 DEPTH TEST, 383, 459 DEPTH WRITEMASK, 460 DepthFunc, 383 DepthMask, 400, 400 DepthRange, 522 DepthRangef, 12, 13, 348 DetachShader, 72 dFdx, 437 dFdy, 437 DIFFERENCE, 391 Disable, 273, 352, 356, 364, 367–369, 381, 383, 384, 396, 424, 433, 440 Disablei, 384, 385 DisableVertexAttribArray, 271, 285 DISPATCH INDIRECT BUFFER, 52, 124, 276, 422, 423, 544 DISPATCH INDIRECT BUFFER BINDING, 485 DispatchCompute, 421, 422 DispatchComputeIndirect, 124, 275, 276, 422, 544 DITHER, 396, 459 do, 314 DONT CARE, 429, 430, 437, 438, 484 Draw*, 520, 529 DRAW BUFFERi, 461 DRAW FRAMEBUFFER, 224, 225, 227, 228, 230, 233, 238, 240– 242, 251, 252, 404, 412, 460, 521 DRAW FRAMEBUFFER BINDING, 201, 227, 253, 397, 460 DRAW INDIRECT BUFFER, 52, 124, 276, 280, 284, 544 DRAW INDIRECT BUFFER BINDING, 450 DrawArrays, 256, 258, 272, 277, 278, 279, 304

OpenGL ES 3.2 (May 5, 2022)

INDEX

559

DrawArraysIndirect, 275, 276, 279, 280, 532, 533 DrawArraysIndirectCommand, 280 DrawArraysInstanced, 279, 279, 282, 530, 532 DrawArraysOneInstance, 277, 278, 530 DrawBuffer, 396, 400, 403 DrawBuffers, 396, 397, 397, 398, 519, 545, 546 DrawElements, 121, 272, 273, 275, 277, 281, 282, 283, 530 DrawElementsBaseVertex, 275, 283 DrawElementsIndirect, 275, 276, 283, 284, 532, 533 DrawElementsIndirectCommand, 284 DrawElementsInstanced, 273, 275, 282, 530, 532 DrawElementsInstancedBaseVertex, 275, 283, 532 DrawElementsOneInstance, 280, 280, 281, 283, 530, 538 DrawRangeElements, 273, 275, 282, 283, 488 DrawRangeElementsBaseVertex, 275, 283, 283 DST ALPHA, 388 DST COLOR, 388 DYNAMIC COPY, 52, 55 DYNAMIC DRAW, 52, 55 DYNAMIC READ, 52, 55

381, 383, 384, 396, 424, 433, 440 Enablei, 384, 385 EnableVertexAttribArray, 271, 277, 285 EndPrimitive, 255 EndQuery, 41, 41, 383, 544 EndTransformFeedback, 47, 48, 339, 339, 343 EQUAL, 129, 189, 213, 382, 383, 472 equal spacing, 316, 322 EXCLUSION, 391 EXT texture sRGB decode, 536 EXTENSIONS, 441–443, 489

FALSE, 10, 12, 38, 40–42, 51, 52, 55, 60, 69, 70, 73, 76–78, 95, 100, 101, 109, 127–129, 135, 136, 140, 143, 211, 212, 215–218, 227, 232, 233, 236, 244, 267, 273, 277, 285, 305, 306, 337, 352, 370, 374, 384, 424, 429, 433, 440, 449–451, 453, 454, 456, 457, 459, 461, 462, 465– 467, 472, 477, 479, 481, 499, 503 FASTEST, 437, 438 FenceSync, 26, 33, 33, 34, 37, 38, 47 Finish, 18, 18, 33, 47, 511 FIRST VERTEX CONVENTION, 334 FIXED, 267 flat, 330, 345, 365, 547 early fragment tests, 378 FLOAT, 89, 147, 148, 150, 193, 219– ELEMENT ARRAY BARRIER BIT, 221, 231, 267, 286, 408, 410, 123 411, 449, 523 ELEMENT ARfloat, 89, 112, 290, 342 RAY BUFFER, 52, 123, 275, FLOAT 32 UNSIGNED INT 24 8 284, 286 REV, 148, 150, 152–155 ELEMENT ARRAY BUFFER BIND- FLOAT MAT2, 90 ING, 286, 449, 530 FLOAT MAT2x3, 90 Enable, 273, 352, 356, 364, 367–369, FLOAT MAT2x4, 90

OpenGL ES 3.2 (May 5, 2022)

INDEX

560

FLOAT MAT3, 90 FLOAT MAT3x2, 90 FLOAT MAT3x4, 90 FLOAT MAT4, 90 FLOAT MAT4x2, 90 FLOAT MAT4x3, 90 FLOAT UNSIGNED INT, 155 FLOAT VEC2, 89 FLOAT VEC3, 89 FLOAT VEC4, 89 Flush, 18, 18, 37, 511, 541 FlushMappedBufferRange, 47, 58, 59, 59 for, 314 FRACTIONAL EVEN, 129 fractional even spacing, 316 FRACTIONAL ODD, 129 fractional odd spacing, 316 FRAGMENT INTERPOLATION OFFSET BITS, 372, 487 FRAGMENT SHADER, 68, 133, 134, 466 FRAGMENT SHADER BIT, 96 FRAGMENT SHADER DERIVATIVE HINT, 437, 484 FRAMEBUFFER, 225, 228, 230, 233, 238, 240–242, 251, 252, 404, 432 FRAMEBUFFER ALPHA SIZE, 171 FRAMEBUFFER ATTACHMENT x SIZE, 462 FRAMEBUFFER ATTACHMENT ALPHA SIZE, 231 FRAMEBUFFER ATTACHMENT BLUE SIZE, 231 FRAMEBUFFER ATTACHMENT COLOR ENCODING, 171, 173, 232, 386, 395, 413, 462, 519

FRAMEBUFFER ATTACHMENT COMPONENT TYPE, 231, 233, 462, 519 FRAMEBUFFER ATTACHMENT DEPTH SIZE, 231 FRAMEBUFFER ATTACHMENT GREEN SIZE, 231 FRAMEBUFFER ATTACHMENT LAYERED, 232, 233, 244, 462 FRAMEBUFFER ATTACHMENT OBJECT NAME, 231–233, 239, 244, 248, 462 FRAMEBUFFER ATTACHMENT OBJECT TYPE, 231– 233, 239, 244, 248, 253, 462, 540 FRAMEBUFFER ATTACHMENT RED SIZE, 231 FRAMEBUFFER ATTACHMENT STENCIL SIZE, 231 FRAMEBUFFER ATTACHMENT TEXTURE CUBE MAP FACE, 232, 244, 462 FRAMEBUFFER ATTACHMENT TEXTURE LAYER, 233, 244, 254, 462, 520 FRAMEBUFFER ATTACHMENT TEXTURE LEVEL, 201, 202, 232, 244, 246, 462, 518 FRAMEBUFFER BARRIER BIT, 124, 126 FRAMEBUFFER BINDING, 227 FRAMEBUFFER BLUE SIZE, 171 FRAMEBUFFER COMPLETE, 252 FRAMEBUFFER DEFAULT, 231 FRAMEBUFFER DEFAULT FIXED SAMPLE LOCATIONS, 228, 230, 461

OpenGL ES 3.2 (May 5, 2022)

INDEX

561

FRAMEBUFFER DEFAULT HEIGHT, 228, 230, 249, 461 FRAMEBUFFER DEFAULT LAYERS, 228, 230, 461 FRAMEBUFFER DEFAULT SAMPLES, 228, 230, 461 FRAMEBUFFER DEFAULT WIDTH, 228, 230, 249, 461 FRAMEBUFFER GREEN SIZE, 171 FRAMEBUFFER INCOMPLETE ATTACHMENT, 249 FRAMEBUFFER INCOMPLETE DIMENSIONS, 249 FRAMEBUFFER INCOMPLETE LAYER TARGETS, 250 FRAMEBUFFER INCOMPLETE MISSING ATTACHMENT, 250 FRAMEBUFFER INCOMPLETE MULTISAMPLE, 250 FRAMEBUFFER RED SIZE, 171 FRAMEBUFFER UNDEFINED, 249 FRAMEBUFFER UNSUPPORTED, 250, 251 FramebufferParameteri, 227, 230, 532 FramebufferRenderbuffer, 238, 239, 251, 546 FramebufferTexture, 45, 240, 243, 244, 546 FramebufferTexture2D, 241, 242–244, 543, 544, 546, 548 FramebufferTextureLayer, 242, 244, 530, 538, 544, 545 FRONT, 364, 382, 400 FRONT AND BACK, 364, 382, 400 FRONT FACE, 453 FrontFace, 363, 364, 374 FUNC ADD, 387, 394, 459 FUNC REVERSE SUBTRACT, 387

FUNC SUBTRACT, 387 fwidth, 437 Gen*, 551 GenBuffers, 25, 26, 50, 50, 51, 268, 269 GENERATE MIPMAP HINT, 437, 484 GenerateMipmap, 204, 521, 542 GenFramebuffers, 224, 226, 226, 227 GenProgramPipelines, 94, 94–97, 130, 131, 307 GenQueries, 39, 39, 40 GenRenderbuffers, 234, 235, 235 GenSamplers, 141, 141–144 GenTextures, 139, 139, 140 GenTransformFeedbacks, 336, 337, 338 GenVertexArrays, 276, 276, 277 GEOMETRY INPUT TYPE, 129, 130, 328, 469 GEOMETRY OUTPUT TYPE, 129, 130, 330, 469 GEOMETRY SHADER, 68, 327, 466 GEOMETRY SHADER BIT, 96 GEOMETRY SHADER INVOCATIONS, 129, 130, 469 GEOMETRY VERTICES OUT, 129, 130, 330, 332, 469 GEQUAL, 189, 213, 382, 383 GetActiveUniformBlockiv, 471 GetActiveAttrib, 291, 292, 469 GetActiveUniform, 106, 109, 468 GetActiveUniformBlockiv, 107, 108 GetActiveUniformBlockName, 107 GetActiveUniformsiv, 106, 107, 470, 471 GetAttachedShaders, 131, 467 GetAttribLocation, 289, 292, 469, 545 GetBooleani v, 399, 440, 460, 479 GetBooleanv, 12, 273, 399, 439, 446, 454, 460, 481, 487, 488

OpenGL ES 3.2 (May 5, 2022)

INDEX

562

GetBufferParameteri64v, 62, 451 GetBufferParameteriv, 62, 451 GetBufferPointerv, 62, 62, 451 GetDebugMessageLog, 428, 434, 434, 435 GetError, 13, 14, 17, 503 GetFloatv, 9, 12, 348, 356, 439, 446, 452–454, 459, 460, 486, 487 GetFragDataIndex, 545 GetFragDataLocation, 377, 545 GetFramebufferAttachmentParameteriv, 462 GetFramebufferParameteriv, 461 GetFramebufferAttachmentParameteriv, 230, 231, 253, 519, 521, 530, 532, 540, 546 GetFramebufferParameteriv, 230, 461, 543 GetGraphicsResetStatus, 16, 16, 17 GetInteger64i v, 63, 440, 449, 478, 480–482 GetInteger64v, 12, 36, 281, 439, 446, 486, 487, 496, 498 GetIntegeri v, 63, 370, 394, 422, 440, 449, 454, 459, 478–482, 495 GetIntegerv, 12, 13, 17, 103, 111, 112, 115–118, 138, 141, 227, 234, 235, 273, 282, 286, 334, 395, 398, 399, 406, 408, 409, 422, 439, 442, 446, 448–450, 452, 453, 455, 459–461, 464, 467, 478, 480, 482, 484–503 GetInternalformativ, 184, 237, 443, 533, 539, 544 GetMultisamplefv, 299, 354, 355, 502 GetnUniformfv, 134 GetnUniformiv, 134 GetnUniformuiv, 134 GetObjectLabel, 435, 436, 449, 451, 456, 458, 462, 463, 465–467,

477, 481 GetObjectPtrLabel, 436, 483 GetPointerv, 434, 441, 499 GetProgramInterfaceiv, 473 GetProgramResourceiv, 474, 475 GetProgramBinary, 100, 100–102, 467, 540, 545 GetProgramInfoLog, 75, 101, 132, 132, 467, 542 GetProgramInterfaceiv, 83, 85 GetProgramiv, 73, 100, 101, 128, 128, 131, 132, 296, 306, 309, 328, 330, 332, 422, 467–470, 472, 528, 545 GetProgramPipelineInfoLog, 132, 132, 466, 542 GetProgramPipelineiv, 130, 132, 307, 466 GetProgramResourceIndex, 84 GetProgramResourceiv, 86, 86, 87, 107, 108, 112, 113, 117, 528, 539, 541, 545 GetProgramResourceLocation, 93, 93 GetProgramResourceName, 85 GetQueryiv, 42, 503, 544 GetQueryObjectuiv, 17, 42, 477 GetRenderbufferParameteriv, 238, 253, 463 GetSamplerParameter, 144, 458 GetSamplerParameterfv, 458 GetSamplerParameterI{i ui}v, 144 GetSamplerParameterIiv, 144 GetSamplerParameterIuiv, 12, 144 GetSamplerParameteriv, 458 GetShaderPrecisionFormat, 488 GetShaderInfoLog, 69, 132, 132, 465, 542 GetShaderiv, 69, 70, 127, 132, 133, 465 GetShaderPrecisionFormat, 69, 133, 551

OpenGL ES 3.2 (May 5, 2022)

INDEX

563

GetShaderSource, 133, 465 GetString, 441, 442, 443, 489 GetStringi, 442, 489 GetSynciv, 17, 34, 38, 38, 483, 539 GetTexLevelParameter, 192, 192, 457 GetTexLevelParameteriv, 457 GetTexParameter, 191, 207, 220, 456 GetTexParameterfv, 456 GetTexParameterI, 191 GetTexParameterIiv, 191 GetTexParameterIuiv, 191 GetTexParameteriv, 456 GetTransformFeedbackVarying, 469 GetTransformFeedbackVarying, 296 GetUniform, 468 GetUniformBlockIndex, 106 GetUniformfv, 134 GetUniformIndices, 105, 539 GetUniformiv, 134 GetUniformLocation, 105, 119, 120, 468, 531 GetUniformuiv, 134 GetVertexAttribPointerv, 449 GetVertexAttribfv, 284, 285, 476 GetVertexAttribIiv, 285, 285 GetVertexAttribIuiv, 285, 285 GetVertexAttribiv, 285, 285, 449 GetVertexAttribPointerv, 286 gl , 83 gl BoundingBox, 312, 313, 325, 352 gl BoundingBox[0], 352 gl BoundingBox[1], 352 gl FragColor, 376, 377 gl FragCoord, 361, 374, 539 gl FragCoord.z, 506 gl FragData[n], 376, 377 gl FragDepth, 376, 377, 506 gl FrontFacing, 374 gl in, 311, 324, 325 gl in[], 331

gl InstanceID, 278, 281, 292, 304 gl InvocationID, 309, 311, 313, 331 GL KHR robust buffer access, 537, 544 GL KHR robustness, 537, 544 gl Layer, 255, 333, 334, 487 gl MaxPatchVertices, 311, 325, 326 gl NumSamples, 375 gl NumWorkGroups, 421 gl out, 312 gl PatchVerticesIn, 311, 325 gl PerVertex, 97, 99 gl PointCoord, 357 gl PointSize, 97, 99, 304, 356, 548 gl Position, 294, 304, 311, 312, 325, 326, 331, 333, 347, 512 gl PrimitiveID, 311, 325, 333, 374 gl PrimitiveIDIn, 331 gl SampleID, 375 gl SampleMask, 376, 547 gl SampleMaskIn, 375, 543, 547 gl SamplePosition, 375 gl TessCoord, 315, 325, 509 gl TessLevelInner, 312, 313, 325, 326 gl TessLevelInner[1], 325 gl TessLevelOuter, 312, 313, 325, 326 gl TessLevelOuter[2], 325 gl TessLevelOuter[3], 325 gl VertexID, 278, 280, 292, 304, 375, 522, 530 glBlitFramebuffer, 16 GREATER, 189, 213, 382, 383 GREEN, 189, 212, 373, 456, 457, 462, 502 GREEN BITS, 253 GUILTY CONTEXT RESET, 16 HALF FLOAT, 147, 148, 150, 221, 267, 410, 411 HARDLIGHT, 391

OpenGL ES 3.2 (May 5, 2022)

INDEX

564

HIGH FLOAT, 133 HIGH INT, 133 highp, 341 Hint, 437 HSL COLOR, 393 HSL HUE, 393 HSL LUMINOSITY, 393 HSL SATURATION, 393 if, 79, 314 iimage2D, 91 iimage2DArray, 92 iimage3D, 91 iimageBuffer, 91 iimageCube, 91 iimageCubeArray, 92 image2D, 91 image2DArray, 91 image3D, 91 IMAGE 2D, 91 IMAGE 2D ARRAY, 91 IMAGE 3D, 91 IMAGE BINDING ACCESS, 479 IMAGE BINDING FORMAT, 479 IMAGE BINDING LAYER, 479 IMAGE BINDING LAYERED, 479 IMAGE BINDING LEVEL, 479 IMAGE BINDING NAME, 479 IMAGE BUFFER, 91 IMAGE CUBE, 91 IMAGE CUBE MAP ARRAY, 91 IMAGE FORMAT COMPATIBILITY BY CLASS, 220 IMAGE FORMAT COMPATIBILITY BY SIZE, 220 IMAGE FORMAT COMPATIBILITY TYPE, 191, 220 imageBuffer, 91 imageCube, 91 imageCubeArray, 91

IMPLEMENTATION COLOR READ FORMAT, 253, 408, 409, 502 IMPLEMENTATION COLOR READ TYPE, 253, 408, 409, 502 in, 325 INCR, 382 INCR WRAP, 382 INFO LOG LENGTH, 127, 128, 131, 132, 465–467 INNOCENT CONTEXT RESET, 16 INT, 90, 147, 148, 150, 169, 193, 219, 221, 231, 267, 270, 408, 411 int, 90, 112, 342 INT 2 10 10 10 REV, 267, 268, 274 INT IMAGE 2D, 91 INT IMAGE 2D ARRAY, 92 INT IMAGE 3D, 91 INT IMAGE BUFFER, 91 INT IMAGE CUBE, 91 INT IMAGE CUBE MAP ARRAY, 92 INT SAMPLER 2D, 91 INT SAMPLER 2D ARRAY, 91 INT SAMPLER 2D MULTISAMPLE, 91 INT SAMPLER 2D MULTISAMPLE ARRAY, 91 INT SAMPLER 3D, 91 INT SAMPLER BUFFER, 91 INT SAMPLER CUBE, 91 INT SAMPLER CUBE MAP ARRAY, 91 INT VEC2, 90 INT VEC3, 90 INT VEC4, 90 INTERLEAVED ATTRIBS, 129, 136, 295, 296, 341, 469 interpolateAtOffset, 487 interpolateAtCentroid, 372

OpenGL ES 3.2 (May 5, 2022)

INDEX

565

interpolateAtOffset, 372 interpolateAtSample, 372 INVALID ENUM, 14, 15, 34, 38, 40– 43, 53, 54, 56, 59, 62, 63, 68, 71, 77, 78, 83–85, 92, 94, 102, 128, 130, 131, 134, 138, 139, 143–145, 166, 173, 180–182, 184, 185, 190–193, 204, 208– 210, 217, 225, 228, 230, 233, 235–238, 240–242, 252, 264, 268, 278, 281, 286, 295, 338, 339, 355, 364, 385, 389, 398, 403, 404, 407, 414, 419, 429– 432, 436, 438, 440–443, 445, 533 INVALID FRAMEBUFFER OPERATION, 15, 176, 252, 414 INVALID INDEX, 84 INVALID OPERATION, 15, 40, 41, 43, 53, 56, 59, 60, 62, 63, 69– 72, 74, 75, 77, 83–85, 92–97, 100, 102, 110, 111, 116, 120, 128, 130–133, 135, 139, 142– 144, 149, 152, 153, 159, 166, 171, 173, 176, 180–185, 190, 204, 208–210, 217, 226, 229, 230, 233, 237, 238, 240–243, 268, 269, 271, 272, 277, 280, 284, 291, 292, 295, 305–309, 327, 328, 334, 337–340, 343, 344, 378, 385, 393, 394, 398, 404, 407, 409, 410, 412, 414, 415, 419, 420, 422, 423, 429, 519, 521, 540, 546, 547 INVALID VALUE, 14, 15, 34–36, 38– 40, 50, 51, 54, 56, 59–61, 63, 68–72, 74, 75, 77, 78, 83– 85, 92–97, 100, 102, 110, 111, 115, 116, 125, 127, 128, 130– 133, 135, 139–143, 145, 166,

173–176, 180–182, 184, 185, 190, 193, 208, 217, 227, 228, 235–237, 241–243, 264, 266– 269, 271, 272, 276–278, 280, 283, 284, 286, 291, 292, 295, 307, 337, 349, 355, 358, 369, 370, 385, 389, 398, 399, 401, 403, 404, 409, 414, 419, 422, 423, 429–431, 433, 435, 436, 440, 441, 443, 445, 519, 531 Invalidate[Sub]Framebuffer, 521, 523 InvalidateFramebuffer, 405 InvalidateSubFramebuffer, 404, 404, 529, 546 INVERT, 382 IS PER PATCH, 86, 88 IS ROW MAJOR, 86, 88, 107, 474 isampler2D, 91 isampler2DArray, 91 isampler2DMS, 91 isampler2DMSArray, 91 isampler3D, 91 isamplerBuffer, 91 isamplerCube, 91 isamplerCubeArray, 91 IsBuffer, 51, 51 IsEnabled, 395, 440, 440, 446, 450, 453, 454, 459, 499 IsEnabledi, 385, 394, 440, 441, 459 IsFramebuffer, 227, 227 ISOLINES, 129 isolines, 314, 322, 325 IsProgram, 77, 77 IsProgramPipeline, 95, 95 IsQuery, 42, 42 IsRenderbuffer, 236, 236 IsSampler, 141, 143, 143 IsShader, 70, 70 IsSync, 38, 38 IsTexture, 140, 140

OpenGL ES 3.2 (May 5, 2022)

INDEX

566

IsTransformFeedback, 337, 337 IsVertexArray, 277, 277 ivec2, 90, 342 ivec3, 90, 342 ivec4, 90, 219, 342

LinkProgram, 71, 73, 74–76, 80, 82, 98, 101, 102, 115, 120, 128, 289, 291, 293, 295, 305, 332, 343, 545 LOCATION, 86, 88, 93, 474 location, 83, 98, 545 KEEP, 382, 459 LOSE CONTEXT ON RESET, 17 LOW FLOAT, 133 LAST VERTEX CONVENTION, 334 LOW INT, 133 LAYER PROVOKING VERTEX, 334, lowp, 341 487 LUMINANCE, 149, 151, 157, 159, 160, layout, 83, 88, 112, 113, 115, 117–119, 169, 172, 374 219, 308, 312, 314, 316, 328, LUMINANCE ALPHA, 149, 151, 157, 329, 331, 378, 422, 539 159, 160, 169, 172, 374 LEQUAL, 189, 212, 213, 382, 383, 456, 458 main, 314 LESS, 189, 213, 382, 383, 459 MAJOR VERSION, 442, 489 LIGHTEN, 391 MAP FLUSH EXPLICIT BIT, 58–60 LINE LOOP, 258, 328, 334, 340 MAP INVALLINE STRIP, 129, 258, 328, 329, 334, IDATE BUFFER BIT, 58, 59, 340 544 LINE STRIP ADJACENCY, 260, 329, MAP INVALIDATE RANGE BIT, 334 57–59, 544 LINE WIDTH, 453 MAP READ BIT, 57–59 LINEAR, 171, 173, 189, 195, 198, 199, MAP UNSYNCHRONIZED BIT, 58, 201, 203, 205, 206, 212, 232, 59, 544 246, 299, 413–415, 456, 458, MAP WRITE BIT, 57–59 519 MapBuffer, 344 LINEAR MIPMAP LINEAR, 189, MapBufferRange, 53, 56, 57, 57–59, 202, 203, 246 268, 344, 544 LINEAR MIPMAP NEAREST, 189, matC, 112 201–203, 246 matCxR, 112 LINES, 129, 258, 328, 334, 339, 340 mat2, 90, 290, 342 lines, 328 mat2x3, 90, 290, 342 LINES ADJACENCY, 129, 260, 329, mat2x4, 90, 290, 342 334 mat3, 90, 109, 290, 342 lines adjacency, 329 mat3x2, 90, 290, 342 LineWidth, 357 mat3x4, 90, 290, 342 LINK STATUS, 73, 100, 101, 128, 467, mat4, 90, 290, 342 522, 528 mat4x2, 90, 290, 342

OpenGL ES 3.2 (May 5, 2022)

INDEX

567

mat4x3, 90, 290, 342 MATRIX STRIDE, 86, 88, 107, 112, 113, 474 MAX, 387 MAX 3D TEXTURE SIZE, 165, 192, 243, 486 MAX ARRAY TEXTURE LAYERS, 184, 243, 486 MAX ATOMIC COUNTER BUFFER BINDINGS, 64, 117, 497 MAX ATOMIC COUNTER BUFFER SIZE, 497 MAX COLOR ATTACHMENTS, 224, 233, 239–242, 252, 397, 404, 407, 486 MAX COLOR TEXTURE SAMPLES, 444, 487 MAX COMBINED ATOMIC COUNTER BUFFERS, 116, 497 MAX COMBINED ATOMIC COUNTERS, 303, 497 MAX COMBINED COMPUTE UNIFORM COMPONENTS, 104, 495 MAX COMBINED FRAGMENT UNIFORM COMPONENTS, 104, 496 MAX COMBINED GEOMETRY UNIFORM COMPONENTS, 104, 496 MAX COMBINED IMAGE UNIFORMS, 303, 498 MAX COMBINED SHADER OUTPUT RESOURCES, 221, 497 MAX COMBINED SHADER STORAGE BLOCKS, 118, 304, 306, 498

MAX COMBINED TESS CONTROL UNIFORM COMPONENTS, 104, 496 MAX COMBINED TESS EVALUATION UNIFORM COMPONENTS, 104, 496 MAX COMBINED TEXTURE IMAGE UNITS, 110, 138, 142, 301, 497 MAX COMBINED UNIFORM BLOCKS, 112, 115, 496 MAX COMBINED VERTEX UNIFORM COMPONENTS, 104, 496 MAX COMPUTE ATOMIC COUNTER BUFFERS, 116, 495, 530 MAX COMPUTE ATOMIC COUNTERS, 303, 495 MAX COMPUTE IMAGE UNIFORMS, 303, 498 MAX COMPUTE SHADER STORAGE BLOCKS, 118, 304, 495 MAX COMPUTE SHARED MEMORY SIZE, 423, 495, 530 MAX COMPUTE TEXTURE IMAGE UNITS, 300, 495 MAX COMPUTE UNIFORM BLOCKS, 111, 495 MAX COMPUTE UNIFORM COMPONENTS, 103, 495, 529 MAX COMPUTE WORK GROUP COUNT, 422, 495 MAX COMPUTE WORK GROUP INVOCATIONS, 422, 495 MAX COMPUTE WORK GROUP SIZE, 422, 495

OpenGL ES 3.2 (May 5, 2022)

INDEX

568

MAX CUBE MAP TEXTURE SIZE, 165, 192, 242, 486 MAX DEBUG GROUP STACK DEPTH, 431, 500 MAX DEBUG LOGGED MESSAGES, 428, 500 MAX DEBUG MESSAGE LENGTH, 427, 430, 431, 500 MAX DEPTH TEXTURE SAMPLES, 444, 487 MAX DRAW BUFFERS, 385, 389, 398, 399, 403, 486 MAX ELEMENT INDEX, 281, 486 MAX ELEMENTS INDICES, 282, 488 MAX ELEMENTS VERTICES, 282, 488 MAX FRAGMENT ATOMIC COUNTER BUFFERS, 116, 494, 548 MAX FRAGMENT ATOMIC COUNTERS, 303, 494, 548 MAX FRAGMENT IMAGE UNIFORMS, 303, 498, 548 MAX FRAGMENT INPUT COMPONENTS, 375, 494 MAX FRAGMENT INTERPOLATION OFFSET, 372, 487, 538 MAX FRAGMENT SHADER STORAGE BLOCKS, 118, 304, 494, 548 MAX FRAGMENT UNIFORM BLOCKS, 111, 494 MAX FRAGMENT UNIFORM COMPONENTS, 103, 494, 527, 529 MAX FRAGMENT UNIFORM VECTORS, 103, 494, 527 MAX FRAME-

BUFFER HEIGHT, 228, 248, 486 MAX FRAMEBUFFER LAYERS, 228, 248, 486 MAX FRAMEBUFFER SAMPLES, 228, 248, 486 MAX FRAMEBUFFER WIDTH, 228, 248, 486 MAX GEOMETRY ATOMIC COUNTER BUFFERS, 116, 493 MAX GEOMETRY ATOMIC COUNTERS, 302, 493 MAX GEOMETRY IMAGE UNIFORMS, 303, 498 MAX GEOMETRY INPUT COMPONENTS, 332, 493 MAX GEOMETRY OUTPUT COMPONENTS, 333, 493, 548 MAX GEOMETRY OUTPUT VERTICES, 332, 493 MAX GEOMETRY SHADER INVOCATIONS, 493 MAX GEOMETRY SHADER STORAGE BLOCKS, 118, 304, 493 MAX GEOMETRY TEXTURE IMAGE UNITS, 300, 493 MAX GEOMETRY TOTAL OUTPUT COMPONENTS, 332, 493 MAX GEOMETRY UNIFORM BLOCKS, 111, 493 MAX GEOMETRY UNIFORM COMPONENTS, 103, 493 MAX IMAGE UNITS, 120, 215, 217, 498 MAX INTEGER SAMPLES, 237, 444, 487

OpenGL ES 3.2 (May 5, 2022)

INDEX

569

MAX LABEL LENGTH, 433, 500 MAX NAME LENGTH, 83–85, 473 MAX NUM ACTIVE VARIABLES, 83, 84, 473 MAX PATCH VERTICES, 264, 491 MAX PROGRAM TEXEL OFFSET, 196, 494 MAX PROGRAM TEXTURE GATHER OFFSET, 196, 201, 494 MAX RECTANGLE TEXTURE SIZE, 165 MAX RENDERBUFFER SIZE, 237, 486 MAX SAMPLE MASK WORDS, 370, 454, 487 MAX SAMPLES, 237, 444, 486, 487, 539, 547 MAX SERVER WAIT TIMEOUT, 36, 487 MAX SHADER STORAGE BLOCK SIZE, 118, 498 MAX SHADER STORAGE BUFFER BINDINGS, 64, 498 MAX TESS CONTROL ATOMIC COUNTER BUFFERS, 116, 491 MAX TESS CONTROL ATOMIC COUNTERS, 302, 491 MAX TESS CONTROL IMAGE UNIFORMS, 303, 498 MAX TESS CONTROL INPUT COMPONENTS, 312, 491, 548 MAX TESS CONTROL OUTPUT COMPONENTS, 313, 491, 548 MAX TESS CONTROL SHADER STORAGE BLOCKS, 118, 303, 491

MAX TESS CONTROL TEXTURE IMAGE UNITS, 300, 491 MAX TESS CONTROL TOTAL OUTPUT COMPONENTS, 313, 491, 547 MAX TESS CONTROL UNIFORM BLOCKS, 111, 491 MAX TESS CONTROL UNIFORM COMPONENTS, 103, 491 MAX TESS EVALUATION ATOMIC COUNTER BUFFERS, 116, 492 MAX TESS EVALUATION ATOMIC COUNTERS, 302, 492 MAX TESS EVALUATION IMAGE UNIFORMS, 303, 498 MAX TESS EVALUATION INPUT COMPONENTS, 326, 492, 548 MAX TESS EVALUATION OUTPUT COMPONENTS, 327, 492, 548 MAX TESS EVALUATION SHADER STORAGE BLOCKS, 118, 304, 492 MAX TESS EVALUATION TEXTURE IMAGE UNITS, 300, 492 MAX TESS EVALUATION UNIFORM BLOCKS, 111, 492 MAX TESS EVALUATION UNIFORM COMPONENTS, 103, 492 MAX TESS GEN LEVEL, 316, 491 MAX TESS PATCH COMPONENTS, 313, 326, 491 MAX TEXTURE BUFFER SIZE, 186, 488

OpenGL ES 3.2 (May 5, 2022)

INDEX

570

MAX TEXTURE IMAGE UNITS, 300, 494 MAX TEXTURE LOD BIAS, 196, 486 MAX TEXTURE SIZE, 165, 184, 192, 242, 243, 486 MAX TRANSFORM FEEDBACK INTERLEAVED COMPONENTS, 296, 501 MAX TRANSFORM FEEDBACK SEPARATE ATTRIBS, 64, 295, 341, 501 MAX TRANSFORM FEEDBACK SEPARATE COMPONENTS, 296, 501 MAX UNIFORM BLOCK SIZE, 112, 495, 496, 527 MAX UNIFORM BUFFER BINDINGS, 64, 116, 496 MAX UNIFORM LOCATIONS, 104, 497 MAX VARYING COMPONENTS, 294, 497 MAX VARYING VECTORS, 294, 497 MAX VERTEX ATOMIC COUNTER BUFFERS, 116, 490 MAX VERTEX ATOMIC COUNTERS, 302, 490 MAX VERTEX ATTRIB BINDINGS, 269, 272, 286, 488 MAX VERTEX ATTRIB RELATIVE OFFSET, 268, 488 MAX VERTEX ATTRIB STRIDE, 269, 271, 488 MAX VERTEX ATTRIBS, 265, 266, 268, 269, 271, 272, 286, 291, 293, 490 MAX VERTEX IMAGE UNIFORMS, 303, 498

MAX VERTEX OUTPUT COMPONENTS, 294, 312, 313, 326, 327, 332, 333, 376, 490 MAX VERTEX SHADER STORAGE BLOCKS, 118, 303, 490 MAX VERTEX TEXTURE IMAGE UNITS, 300, 490 MAX VERTEX UNIFORM BLOCKS, 111, 490 MAX VERTEX UNIFORM COMPONENTS, 103, 490 MAX VERTEX UNIFORM VECTORS, 103, 490 MAX VIEWPORT DIMS, 348, 405, 487 MEDIUM FLOAT, 133 MEDIUM INT, 133 mediump, 341 MemoryBarrier, 123, 125–127 memoryBarrier, 122, 126 MemoryBarrierByRegion, 126, 127 MIN, 387 MIN FRAGMENT INTERPOLATION OFFSET, 372, 487 MIN PROGRAM TEXEL OFFSET, 196, 494 MIN PROGRAM TEXTURE GATHER OFFSET, 196, 201, 494 MIN SAMPLE SHADING VALUE, 356, 454 MINOR VERSION, 442, 489 MinSampleShading, 356, 543 MIRRORED REPEAT, 189, 198 MULTIPLY, 391 MULTISAMPLE LINE WIDTH GRANULARITY, 363, 486, 537 MULTISAMPLE LINE WIDTH -

OpenGL ES 3.2 (May 5, 2022)

INDEX

571 RANGE, 363, 486, 537

NAME LENGTH, 86, 88, 107, 108, 475 NEAREST, 189, 194, 197, 201, 203, 205, 206, 213, 246, 299, 413, 414 NEAREST MIPMAP LINEAR, 189, 201–203, 212, 246, 458 NEAREST MIPMAP NEAREST, 189, 201–203, 206, 213, 246, 528 NEVER, 189, 213, 382, 383 NICEST, 437, 438 NO ERROR, 14, 16, 17 NO RESET NOTIFICATION, 17 NONE, 176, 189, 193, 206, 211, 212, 231–233, 244, 248, 301, 381, 393, 396–399, 402, 403, 406, 409, 456–458, 462, 519, 521, 532, 540 NOTEQUAL, 189, 213, 382, 383 NULL, 427, 435, 436, 449, 451, 499, 522 NUM ACTIVE VARIABLES, 86–88, 108, 475 NUM COMPRESSED TEXTURE FORMATS, 177, 488 NUM EXTENSIONS, 443, 489 NUM PROGRAM BINARY FORMATS, 100, 101, 488 NUM SAMPLE COUNTS, 444 NUM SAMPLES COUNTS, 445 NUM SHADER BINARY FORMATS, 66, 70, 488 OBJECT TYPE, 34, 38, 483 ObjectLabel, 432, 433, 435 ObjectPtrLabel, 433 OFFSET, 86, 89, 107, 475

ONE, 189, 373, 387, 388, 394, 459 ONE MINUS CONSTANT ALPHA, 388 ONE MINUS CONSTANT COLOR, 388 ONE MINUS DST ALPHA, 388 ONE MINUS DST COLOR, 388 ONE MINUS SRC ALPHA, 388 ONE MINUS SRC COLOR, 388 out, 312 OUT OF MEMORY, 14, 15, 208 OVERLAY, 391 PACK ALIGNMENT, 408, 464 PACK IMAGE HEIGHT, 521 PACK ROW LENGTH, 408, 464 PACK SKIP IMAGES, 521 PACK SKIP PIXELS, 408, 464 PACK SKIP ROWS, 408, 464 patch, 88, 309 patch in, 326 patch out, 313 PATCH VERTICES, 264, 448 PATCHES, 263, 273, 309, 487 PatchParameteri, 264 PauseTransformFeedback, 339, 340 PIXEL BUFFER BARRIER BIT, 124 PIXEL PACK BUFFER, 52, 124, 145, 407 PIXEL PACK BUFFER BINDING, 411, 464 PIXEL UNPACK BUFFER, 52, 124, 145 PIXEL UNPACK BUFFER BINDING, 149, 177, 464 PixelStorei, 144, 145, 145, 408, 420 point mode, 316 POINTS, 129, 258, 328, 329, 334, 339, 340 points, 328

OpenGL ES 3.2 (May 5, 2022)

INDEX

572

POLYGON OFFSET FACTOR, 453 PushDebugGroup, 431, 431 POLYGON OFFSET FILL, 367, 453 QUADS, 129, 472 POLYGON OFFSET UNITS, 453 quads, 314, 316, 319, 325 PolygonOffset, 366 QUERY, 432 PopDebugGroup, 431, 431 PRIMITIVE BOUNDING BOX, 312, QUERY RESULT, 40, 43, 477 QUERY RESULT 452 AVAILABLE, 17, 40, 42, 43, PRIMITIVE RESTART FIXED IN477, 541 DEX, 273, 450 PRIMITIVE RESTART FOR R11F G11F B10F, 147, 160, 162, 417 PATCHES SUPPORTED, R16F, 148, 162, 171, 187, 418 273, 487 R16I, 148, 162, 187, 418 PrimitiveBoundingBox, 352 R16UI, 148, 162, 187, 418 PRIMITIVES GENERATED, 39, 344 R32F, 148, 162, 171, 187, 219, 221, PROGRAM, 432 237, 417, 444 PROGRAM BINARY FORMATS, r32f, 219 101, 488 R32I, 148, 162, 187, 219, 221, 417 PROGRAM BINARY LENGTH, 100, r32i, 219 467 R32UI, 148, 162, 187, 219, 221, 417, PROGRAM BINARY RE479 TRIEVABLE HINT, 76, 77, r32ui, 219 102, 130, 467 R8, 148, 162, 172, 187, 418, 457 PROGRAM INPUT, 79, 82, 86, 87, 93, R8 SNORM, 148, 162, 418 291, 292 R8I, 148, 162, 187, 418 PROGRAM OUTPUT, 80, 82, 86, 87, R8UI, 148, 162, 187, 418 93, 377, 378 RASTERIZER DISCARD, 252, 352, PROGRAM PIPELINE, 432 453 PROGRAM PIPELINE BINDING, READ BUFFER, 406, 461 467 READ FRAMEBUFFER, 224, 225, PROGRAM SEPARABLE, 76–78, 96, 227, 228, 230, 233, 238, 240– 101, 130, 305, 467, 539 242, 251, 252, 404, 412, 460, ProgramBinary, 75, 76, 100, 101, 102, 521 128, 343, 539 READ FRAMEBUFFER BINDING, ProgramParameteri, 76, 102, 531 176, 227, 253, 409, 410, 460 ProgramUniform, 110 READ ONLY, 216, 479 ProgramUniform{1234}ui, 110 READ WRITE, 216 ProgramUniform{1234}uiv, 110 ReadBuffer, 397, 406, 420, 546 ProgramUniformMatrix{234}, 111 ReadnPixels, 408, 409 ProgramUniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}, 111

OpenGL ES 3.2 (May 5, 2022)

INDEX

573

ReadPixels, 124, 144, 153, 169, 219, 220, 252, 255, 344, 406, 407, 408, 408–410 RED, 148, 151, 159, 162, 178, 189, 193, 212, 221, 373, 374, 409, 412, 456, 457, 462, 502 RED BITS, 253 RED INTEGER, 148, 151, 221 REFERENCED BY COMPUTE SHADER, 86, 89, 475 REFERENCED BY FRAGMENT SHADER, 86, 89, 108, 475 REFERENCED BY GEOMETRY SHADER, 86, 89, 475 REFERENCED BY TESS CONTROL SHADER, 86, 89, 475 REFERENCED BY TESS EVALUATION SHADER, 86, 89, 475 REFERENCED BY VERTEX SHADER, 86, 89, 108, 475 ReleaseShaderCompiler, 69, 69 RENDERBUFFER, 231, 232, 234–236, 238–240, 253, 416, 419, 432, 443, 444, 460 RENDERBUFFER ALPHA SIZE, 238, 463 RENDERBUFFER BINDING, 235, 460 RENDERBUFFER BLUE SIZE, 238, 463 RENDERBUFFER DEPTH SIZE, 238, 463 RENDERBUFFER GREEN SIZE, 238, 463 RENDERBUFFER HEIGHT, 236, 238, 463

RENDERBUFFER INTERNAL FORMAT, 236, 238, 463 RENDERBUFFER RED SIZE, 238, 463 RENDERBUFFER SAMPLES, 234, 236, 238, 250, 463 RENDERBUFFER STENCIL SIZE, 238, 463 RENDERBUFFER WIDTH, 236, 238, 463 RenderbufferStorage, 236, 237 RenderbufferStorage*, 251 RenderbufferStorageMultisample, 228, 236, 236, 237, 544 renderbuffertarget, 239 RENDERER, 441, 442, 489 REPEAT, 189, 198, 212, 458 REPLACE, 382 RESET NOTIFICATION STRATEGY, 17, 503 ResumeTransformFeedback, 339, 339, 340, 343, 546 RG, 148, 151, 159, 162, 163, 178, 374, 409, 412 RG16, 417 RG16F, 148, 162, 171, 187, 417 RG16I, 148, 162, 187, 417 RG16UI, 148, 163, 187, 417 RG32F, 148, 162, 171, 187, 237, 417, 444 RG32I, 148, 163, 187, 417 RG32UI, 148, 163, 187, 417 RG8, 148, 162, 172, 187, 418 RG8 SNORM, 148, 162, 418 RG8I, 148, 162, 187, 418 RG8UI, 148, 162, 187, 418 RG INTEGER, 148, 151 RGB, 147, 149, 151, 154, 156, 159– 163, 169, 172, 178, 248, 374, 388, 409, 410, 412

OpenGL ES 3.2 (May 5, 2022)

INDEX

574

RGB10 A2, 147, 162, 170, 408, 417 RGB10 A2UI, 147, 162, 417 RGB16F, 147, 162, 171, 417 RGB16I, 147, 163, 417 RGB16UI, 147, 163, 417 RGB32F, 147, 162, 171, 187, 417 RGB32I, 148, 163, 187, 417 RGB32UI, 148, 163, 187, 417 RGB565, 147, 160, 162, 172 RGB5 A1, 147, 160, 162, 172 RGB8, 147, 160, 162, 172, 418 RGB8 SNORM, 147, 162, 418 RGB8I, 147, 163, 418 RGB8UI, 147, 163, 418 RGB9 E5, 147, 160, 162, 176, 215, 417, 522 RGB INTEGER, 147, 148, 151 RGBA, 147, 149, 151, 154, 156, 159, 160, 162, 163, 169, 172, 178, 179, 211, 219–221, 248, 374, 408, 409, 457 RGBA10 A2, 172 RGBA16F, 147, 162, 171, 187, 219, 221, 237, 417, 444 rgba16f, 219 RGBA16I, 147, 163, 187, 219, 221, 417 rgba16i, 219 RGBA16UI, 147, 163, 187, 219, 221, 417 rgba16ui, 219 RGBA32F, 147, 162, 171, 187, 219, 220, 237, 417, 444 rgba32f, 219 RGBA32I, 147, 163, 187, 219, 221, 417 rgba32i, 219 RGBA32UI, 147, 163, 187, 219, 221, 417 rgba32ui, 219 RGBA4, 147, 160, 162, 172, 463 RGBA8, 147, 160, 162, 172, 187, 219,

221, 417 rgba8, 219 RGBA8 SNORM, 147, 162, 219, 221, 417 rgba8 snorm, 219 RGBA8I, 147, 163, 187, 219, 221, 417 rgba8i, 219 RGBA8UI, 147, 163, 187, 219, 221, 417 rgba8ui, 219 RGBA INTEGER, 147, 151, 154, 169, 219, 221, 408 sample, 372, 375 sample in, 355, 372 SAMPLE ALPHA TO COVERAGE, 379, 381, 454 SAMPLE BUFFERS, 121, 176, 233, 234, 354–357, 362, 367, 369, 372, 379, 396, 400, 409, 414, 502, 519, 520, 528 SAMPLE COVERAGE, 369, 375, 454 SAMPLE COVERAGE INVERT, 369, 454 SAMPLE COVERAGE VALUE, 369, 370, 454 SAMPLE MASK, 369, 375, 454 SAMPLE MASK VALUE, 13, 369, 370, 454, 533 SAMPLE POSITION, 354, 355, 502 SAMPLE SHADING, 356, 454 SampleCoverage, 370 SampleMaski, 370 SAMPLER, 432 sampler*, 119 sampler*Shadow, 301 sampler2D, 90, 119 sampler2DArray, 90 sampler2DArrayShadow, 90 sampler2DMS, 90

OpenGL ES 3.2 (May 5, 2022)

INDEX

575

sampler2DMSArray, 90 sampler2DShadow, 90 sampler3D, 90 samplerCubeArray, 90 SAMPLER 2D, 90 SAMPLER 2D ARRAY, 90 SAMPLER 2D ARRAY SHADOW, 90 SAMPLER 2D MULTISAMPLE, 90 SAMPLER 2D MULTISAMPLE ARRAY, 90 SAMPLER 2D SHADOW, 90 SAMPLER 3D, 90 SAMPLER BINDING, 141, 455 SAMPLER BUFFER, 90 SAMPLER CUBE, 90 SAMPLER CUBE MAP ARRAY, 90 SAMPLER CUBE MAP ARRAY SHADOW, 90 SAMPLER CUBE SHADOW, 90 samplerBuffer, 90 samplerCube, 90 samplerCubeArrayShadow, 90 samplerCubeShadow, 90 SamplerParameter, 142 SamplerParameterI{i ui}v, 142 SamplerParameterIiv, 142 SamplerParameterIuiv, 142 SamplerParameteriv, 142 SAMPLES, 184, 233, 234, 354–356, 444, 445, 502, 519 Scissor, 368 SCISSOR BOX, 459 SCISSOR TEST, 368, 459 SCREEN, 391 SEPARATE ATTRIBS, 129, 295, 296, 341 SHADER, 432 SHADER BINARY FORMATS, 70, 71, 488

SHADER COMPILER, 66, 488 SHADER IMAGE ACCESS BARRIER BIT, 124, 126 SHADER SOURCE LENGTH, 127, 128, 133, 465 SHADER STORAGE BARRIER BIT, 125, 127 SHADER STORAGE BLOCK, 80, 84, 86 SHADER STORAGE BUFFER, 52, 53, 119 SHADER STORAGE BUFFER BINDING, 64, 480 SHADER STORAGE BUFFER OFFSET ALIGNMENT, 64, 498, 529, 530 SHADER STORAGE BUFFER SIZE, 64, 480 SHADER STORAGE BUFFER START, 64, 480 SHADER TYPE, 127, 128, 135, 465 ShaderBinary, 70, 71, 541 ShaderSource, 68, 68, 69, 133 SHADING LANGUAGE VERSION, 441, 442, 489 shared, 103, 423, 519 SHORT, 147, 148, 150, 221, 267, 270, 411 SIGNALED, 17, 34, 38 SIGNED NORMALIZED, 193, 231 SOFTLIGHT, 391 SRC ALPHA, 388 SRC ALPHA SATURATE, 388 SRC COLOR, 388 SRGB, 171, 173, 232, 386, 395, 413 SRGB8, 147, 162, 214, 418 SRGB8 ALPHA8, 147, 162, 214, 417 SRGB ALPHA8, 172 STACK OVERFLOW, 15, 431 STACK UNDERFLOW, 15, 431

OpenGL ES 3.2 (May 5, 2022)

INDEX

576

STATIC COPY, 52, 55 STATIC DRAW, 52, 55, 451 STATIC READ, 52, 55 std140, 103, 113–115, 118, 519, 541 std430, 114, 115, 118, 423, 541 STENCIL, 230, 231, 233, 402, 403, 405, 457, 462 STENCIL8, 537 STENCIL ATTACHMENT, 226, 239, 249, 404 STENCIL BACK FAIL, 459 STENCIL BACK FUNC, 459 STENCIL BACK PASS DEPTH FAIL, 459 STENCIL BACK PASS DEPTH PASS, 459 STENCIL BACK REF, 459 STENCIL BACK VALUE MASK, 459 STENCIL BACK WRITEMASK, 460 STENCIL BITS, 253, 502 STENCIL BUFFER BIT, 401, 403, 413–415 STENCIL CLEAR VALUE, 460 STENCIL FAIL, 459 STENCIL FUNC, 459 STENCIL INDEX, 148, 151, 158, 159, 164, 189, 205, 206, 212, 248, 302, 408, 528 STENCIL INDEX8, 148, 164, 527 STENCIL PASS DEPTH FAIL, 459 STENCIL PASS DEPTH PASS, 459 STENCIL REF, 459 STENCIL TEST, 381, 459 STENCIL VALUE MASK, 459 STENCIL WRITEMASK, 13, 460 StencilFunc, 381, 382, 383, 511 StencilFuncSeparate, 381, 382 StencilMask, 400, 400, 511 StencilMaskSeparate, 400, 400 StencilOp, 381, 382

StencilOpSeparate, 381, 382 STREAM COPY, 52, 55 STREAM DRAW, 52, 54 STREAM READ, 52, 55 SUBPIXEL BITS, 486 switch, 314 SYNC, 436 SYNC CONDITION, 34, 38, 483 SYNC FENCE, 34, 38, 483 SYNC FLAGS, 34, 38, 483 SYNC FLUSH COMMANDS BIT, 18, 35–37 SYNC GPU COMMANDS COMPLETE, 34, 38, 483 SYNC STATUS, 17, 34, 38, 483 TESS CONTROL OUTPUT VERTICES, 129, 130, 309, 472 TESS CONTROL SHADER, 68, 309, 466 TESS CONTROL SHADER BIT, 96 TESS EVALUATION SHADER, 68, 322, 466 TESS EVALUATION SHADER BIT, 96 TESS GEN MODE, 129, 130, 472 TESS GEN POINT MODE, 129, 130, 472 TESS GEN SPACING, 129, 130, 472 TESS GEN VERTEX ORDER, 129, 130, 472 TexBuffer, 186 TexBufferRange, 185 texelFetch, 298 TexImage, 138, 174, 175 TexImage*, 170, 522 TexImage*D, 144 TexImage2D, 145, 163, 166, 166, 167, 169, 174, 177, 181, 202, 219

OpenGL ES 3.2 (May 5, 2022)

INDEX

577

TexImage3D, 145, 157, 158, 163, 165– 167, 174, 177, 181, 202, 538 TexParameter, 47, 138, 142, 188 TexParameterI, 188 TexParameterIiv, 188 TexParameterIuiv, 188 TexParameteriv, 188 TexStorage2D, 208 TexStorage2DMultisample, 183, 183, 228, 528, 529, 542 TexStorage3D, 209, 538, 543 TexStorage3DMultisample, 183, 183, 184 TexSubImage, 124, 174, 175 TexSubImage*, 522 TexSubImage*D, 144 TexSubImage2D, 145, 174, 174, 176, 181 TexSubImage3D, 145, 173, 174, 175, 181, 220 TEXTURE, 231, 232, 244, 432 TEXTURE0, 138, 455 TEXTURE31, 138 TEXTURE x SIZE, 457 TEXTURE x TYPE, 457 TEXTURE xD, 455 TEXTURE 2D, 119, 138, 158, 166, 169, 173, 174, 180, 182, 188, 191, 192, 204, 208, 209, 217, 241, 242 TEXTURE 2D ARRAY, 139, 157, 159, 174, 180, 182, 188, 191, 192, 204, 210, 217, 455 TEXTURE 2D MULTISAMPLE, 139, 158, 183, 188, 190–192, 241, 242, 443, 444, 455 TEXTURE 2D MULTISAMPLE ARRAY, 139, 159, 183, 188, 190– 192, 443, 444, 455 TEXTURE 3D, 139, 157, 174, 180,

182, 188, 191, 192, 204, 209, 210, 217 TEXTURE ALPHA SIZE, 193 TEXTURE ALPHA TYPE, 192 TEXTURE BASE LEVEL, 189–191, 207, 212, 246, 456, 548 TEXTURE BINDING xD, 455 TEXTURE BINDING 2D ARRAY, 455 TEXTURE BINDING 2D MULTISAMPLE, 455 TEXTURE BINDING 2D MULTISAMPLE ARRAY, 455 TEXTURE BINDING BUFFER, 455 TEXTURE BINDING CUBE MAP, 455 TEXTURE BINDING CUBE MAP ARRAY, 455 TEXTURE BLUE SIZE, 193 TEXTURE BLUE TYPE, 192 TEXTURE BORDER COLOR, 142– 144, 188–191, 198, 212, 456, 458 TEXTURE BUFFER, 52, 139, 185, 188, 192, 217, 416, 419, 455 TEXTURE BUFFER BINDING, 503 TEXTURE BUFFER DATA STORE BINDING, 457 TEXTURE BUFFER OFFSET, 457 TEXTURE BUFFER OFFSET ALIGNMENT, 185, 488 TEXTURE BUFFER SIZE, 457 TEXTURE COMPARE FUNC, 189, 212, 456, 458 TEXTURE COMPARE MODE, 189, 206, 212, 213, 301, 456, 458 TEXTURE COMPRESSED, 457 TEXTURE CUBE MAP, 139, 159,

OpenGL ES 3.2 (May 5, 2022)

INDEX

578

167, 188, 191, 192, 204, 209, 217, 455 TEXTURE CUBE MAP ARRAY, 139, 157, 159, 164, 166, 167, 174, 180, 182, 188, 191, 192, 204, 210, 217, 455, 538, 543 TEXTURE CUBE MAP NEGATIVE X, 194, 216 TEXTURE CUBE MAP NEGATIVE Y, 194, 216 TEXTURE CUBE MAP NEGATIVE Z, 194, 216 TEXTURE CUBE MAP POSITIVE X, 194, 216 TEXTURE CUBE MAP POSITIVE Y, 194, 216 TEXTURE CUBE MAP POSITIVE Z, 194, 216 TEXTURE DEPTH, 193, 457 TEXTURE DEPTH SIZE, 193 TEXTURE DEPTH TYPE, 192 TEXTURE FETCH BARRIER BIT, 124, 127 TEXTURE FIXED SAMPLE LOCATIONS, 184, 193, 250, 457 TEXTURE GREEN SIZE, 193 TEXTURE GREEN TYPE, 192 TEXTURE HEIGHT, 183, 184, 193, 457 TEXTURE IMMUTABLE FORMAT, 185, 191, 207, 208, 212, 456, 518 TEXTURE IMMUTABLE LEVELS, 190, 191, 208, 212, 241, 243, 456, 518 TEXTURE INTERNAL FORMAT, 184, 193, 457 TEXTURE MAG FILTER, 189, 205, 212, 213, 456, 458

TEXTURE MAX LEVEL, 189, 190, 207, 212, 246, 456, 548 TEXTURE MAX LOD, 189, 190, 212, 456, 458 TEXTURE MIN FILTER, 141, 189, 197, 198, 201, 202, 205, 207, 212, 213, 246, 456, 458 TEXTURE MIN LOD, 189, 190, 212, 456, 458 TEXTURE RED SIZE, 193 TEXTURE RED TYPE, 192 TEXTURE SAMPLES, 183, 184, 193, 234, 250, 457 TEXTURE SHARED SIZE, 193, 457 TEXTURE STENCIL SIZE, 193 TEXTURE SWIZZLE A, 189, 212, 373, 456 TEXTURE SWIZZLE B, 189, 212, 373, 456 TEXTURE SWIZZLE G, 189, 212, 373, 456 TEXTURE SWIZZLE R, 189, 212, 373, 456 TEXTURE UPDATE BARRIER BIT, 124 TEXTURE WIDTH, 183, 184, 193, 457 TEXTURE WRAP R, 189, 198, 456, 458 TEXTURE WRAP S, 189, 198, 456, 458 TEXTURE WRAP T, 189, 198, 456, 458 textureGather, 196, 199–201, 494 textureGatherOffset, 199 textureGatherOffsets, 201 textureSize, 300, 540 TIMEOUT EXPIRED, 35 TIMEOUT IGNORED, 36 TOP LEVEL ARRAY SIZE, 87, 89,

OpenGL ES 3.2 (May 5, 2022)

INDEX

579

475, 528 TOP LEVEL ARRAY STRIDE, 87, 89, 475, 528 TRANSFORM FEEDBACK, 337, 338, 432 TRANSFORM FEEDBACK ACTIVE, 481 TRANSFORM FEEDBACK BARRIER BIT, 125 TRANSFORM FEEDBACK BINDING, 452 TRANSFORM FEEDBACK BUFFER, 52, 53, 340, 344 TRANSFORM FEEDBACK BUFFER BINDING, 64, 338, 452, 481, 543 TRANSFORM FEEDBACK BUFFER MODE, 129, 469 TRANSFORM FEEDBACK BUFFER SIZE, 64, 481 TRANSFORM FEEDBACK BUFFER START, 64, 481 TRANSFORM FEEDBACK PAUSED, 481 TRANSFORM FEEDBACK PRIMITIVES WRITTEN, 39, 42, 344 TRANSFORM FEEDBACK VARYING, 80, 82, 84, 86, 87, 296, 297 TRANSFORM FEEDBACK VARYING MAX LENGTH, 129, 469 TRANSFORM FEEDBACK VARYINGS, 129, 296, 469 TransformFeedbackVaryings, 80, 82, 84, 295, 295, 296, 341 TRIANGLE FAN, 260, 264, 329, 334,

340 TRIANGLE STRIP, 129, 259, 264, 329, 334, 340, 469 TRIANGLE STRIP ADJACENCY, 263, 264, 329, 334 TRIANGLES, 129, 260, 264, 329, 334, 339, 340, 469 triangles, 314, 316, 317, 329 TRIANGLES ADJACENCY, 129, 262, 264, 329, 334 triangles adjacency, 329 TRUE, 10, 12, 17, 38, 39, 42, 43, 51, 52, 58, 60, 66, 69–71, 73, 76–78, 95, 101, 102, 109, 127–130, 140, 143, 184, 185, 208, 215, 217, 218, 227, 232, 236, 244, 250, 267, 273, 277, 285, 306, 337, 369, 370, 374, 384, 399, 424, 429, 440, 457, 459, 460, 477, 488, 499, 541 TYPE, 87, 89, 106, 107, 291, 296, 475 uimage2D, 92 uimage2DArray, 92 uimage3D, 92 uimageBuffer, 92 uimageCube, 92 uimageCubeArray, 92 uint, 90, 112, 116, 342 UNDEFINED VERTEX, 334 UNIFORM, 79, 82, 86, 87, 89, 93, 105, 106 Uniform, 9 Uniform1f, 10 Uniform1i, 10 Uniform2f, 10 Uniform2i, 10 Uniform3f, 10 Uniform3i, 10

OpenGL ES 3.2 (May 5, 2022)

INDEX

580

Uniform4f, 9, 10 Uniform4i, 10 UNIFORM ARRAY STRIDE, 107, 117, 471, 541 UNIFORM BARRIER BIT, 124, 127 UNIFORM BLOCK, 79, 84, 86, 106, 107 UNIFORM BLOCK ACTIVE UNIFORM INDICES, 108, 471 UNIFORM BLOCK ACTIVE UNIFORMS, 108, 471 UNIFORM BLOCK BINDING, 108, 471 UNIFORM BLOCK DATA SIZE, 108, 115, 471 UNIFORM BLOCK INDEX, 107, 470 UNIFORM BLOCK NAME LENGTH, 108, 471 UNIFORM BLOCK REFERENCED BY FRAGMENT SHADER, 108, 471 UNIFORM BLOCK REFERENCED BY GEOMETRY SHADER, 538 UNIFORM BLOCK REFERENCED BY TESS CONTROL SHADER, 538 UNIFORM BLOCK REFERENCED BY TESS EVALUATION SHADER, 538 UNIFORM BLOCK REFERENCED BY VERTEX SHADER, 108, 471 UNIFORM BUFFER, 52, 53, 115 UNIFORM BUFFER BINDING, 64, 482 UNIFORM BUFFER OFFSET ALIGNMENT, 64, 496, 529, 530 UNIFORM BUFFER SIZE, 64, 482

UNIFORM BUFFER START, 64, 482 UNIFORM IS ROW MAJOR, 107, 471 UNIFORM MATRIX STRIDE, 107, 471 UNIFORM NAME LENGTH, 107, 470 UNIFORM OFFSET, 107, 470 UNIFORM SIZE, 107, 470 UNIFORM TYPE, 107, 470 Uniform{1234}{if ui}, 108 Uniform{1234}{if ui}v, 108 UniformBlockBinding, 115, 115 UniformMatrix2x4fv, 109 UniformMatrix3fv, 109 UniformMatrix{234}fv, 108 UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}fv, 108 UNKNOWN CONTEXT RESET, 16 UnmapBuffer, 47, 50, 55, 58, 60, 60 UNPACK ALIGNMENT, 145, 152, 158, 464 UNPACK IMAGE HEIGHT, 145, 158, 464 UNPACK ROW LENGTH, 145, 152, 158, 464 UNPACK SKIP IMAGES, 145, 158, 167, 464 UNPACK SKIP PIXELS, 145, 152, 464 UNPACK SKIP ROWS, 145, 152, 464 UNSIGNALED, 34, 38, 483 UNSIGNED BYTE, 147–150, 160, 169, 221, 267, 270, 273, 280, 281, 408, 411 UNSIGNED INT, 90, 147, 148, 150, 155, 169, 193, 219, 221, 231, 267, 270, 273, 280, 281, 408, 411 UNSIGNED INT 10F 11F -

OpenGL ES 3.2 (May 5, 2022)

INDEX

581

11F REV, 147, 150, 154–156, 410, 411 UNSIGNED INT 24 8, 148, 150, 154, 155 UNSIGNED INT 2 10 10 10 REV, 147, 150, 154, 155, 169, 267, 268, 274, 408, 411 UNSIGNED INT 5 9 9 9 REV, 147, 150, 154–156, 161 UNSIGNED INT ATOMIC COUNTER, 92 UNSIGNED INT IMAGE 2D, 92 UNSIGNED INT IMAGE 2D ARRAY, 92 UNSIGNED INT IMAGE 3D, 92 UNSIGNED INT IMAGE BUFFER, 92 UNSIGNED INT IMAGE CUBE, 92 UNSIGNED INT IMAGE CUBE MAP ARRAY, 92 UNSIGNED INT SAMPLER 2D, 91 UNSIGNED INT SAMPLER 2D ARRAY, 91 UNSIGNED INT SAMPLER 2D MULTISAMPLE, 91 UNSIGNED INT SAMPLER 2D MULTISAMPLE ARRAY, 91 UNSIGNED INT SAMPLER 3D, 91 UNSIGNED INT SAMPLER BUFFER, 91 UNSIGNED INT SAMPLER CUBE, 91 UNSIGNED INT SAMPLER CUBE MAP ARRAY, 91 UNSIGNED INT VEC2, 90 UNSIGNED INT VEC3, 90 UNSIGNED INT VEC4, 90 UNSIGNED NORMALIZED, 193, 231 UNSIGNED SHORT, 147, 148, 150, 154, 221, 267, 270, 273, 280,

281, 411 UNSIGNED SHORT 4 4 4 4, 147, 149, 150, 154, 160, 411 UNSIGNED SHORT 5 5 5 1, 147, 149, 150, 154, 160, 411 UNSIGNED SHORT 5 6 5, 147, 149, 150, 154, 160, 411 usampler2D, 91 usampler2DArray, 91 usampler2DMS, 91 usampler2DMSArray, 91 usampler3D, 91 usamplerBuffer, 91 usamplerCube, 91 usamplerCubeArray, 91 UseProgram, 75, 75, 76, 95, 108, 305– 307, 343, 529, 545 UseProgramStages, 76, 96, 96, 130, 305, 343, 545 uvec2, 90, 342 uvec3, 90, 342 uvec4, 90, 219, 342 VALIDATE STATUS, 128, 131, 306, 307, 466, 467 ValidateProgram, 128, 306, 306, 519 ValidateProgramPipeline, 131, 307 vec2, 89, 290, 342 vec3, 89, 290, 342 vec4, 89, 109, 114, 115, 219, 290, 312, 342, 541 VENDOR, 441, 442, 489 VERSION, 441, 442, 489 VERTEX ARRAY, 432 VERTEX ARRAY BINDING, 13, 280, 284, 285, 450 VERTEX ATTRIB ARRAY BARRIER BIT, 123

OpenGL ES 3.2 (May 5, 2022)

INDEX

582

VERTEX ATTRIB ARRAY BUFFER, 123 VERTEX ATTRIB ARRAY BUFFER BINDING, 274, 285, 449 VERTEX ATTRIB ARRAY DIVISOR, 285, 449 VERTEX ATTRIB ARRAY ENABLED, 285, 449 VERTEX ATTRIB ARRAY INTEGER, 285, 449 VERTEX ATTRIB ARRAY NORMALIZED, 285, 449 VERTEX ATTRIB ARRAY POINTER, 271, 286, 449 VERTEX ATTRIB ARRAY SIZE, 285, 449 VERTEX ATTRIB ARRAY STRIDE, 271, 285, 449 VERTEX ATTRIB ARRAY TYPE, 285, 449 VERTEX ATTRIB BINDING, 285, 449 VERTEX ATTRIB RELATIVE OFFSET, 285, 449 VERTEX BINDING BUFFER, 285, 449 VERTEX BINDING DIVISOR, 285, 449 VERTEX BINDING OFFSET, 449, 532 VERTEX BINDING STRIDE, 271, 449 VERTEX SHADER, 68, 133, 134, 466 VERTEX SHADER BIT, 96 VertexAttribBinding, 269, 271, 274 VertexAttribDivisor, 272, 278, 279, 281 VertexAttribFormat, 267, 268, 270, 285 VertexAttribI4, 265

VertexAttribIFormat, 267, 268, 270, 285 VertexAttribIPointer, 268, 270 VertexAttribPointer, 268, 270, 274, 277 VertexBindingDivisor, 272 vertices, 309 VIEW CLASS 128 BITS, 417 VIEW CLASS 16 BITS, 418 VIEW CLASS 24 BITS, 418 VIEW CLASS 32 BITS, 417 VIEW CLASS 48 BITS, 417 VIEW CLASS 64 BITS, 417 VIEW CLASS 8 BITS, 418 VIEW CLASS 96 BITS, 417 VIEW CLASS ASTC 10x10 RGBA, 418 VIEW CLASS ASTC 10x5 RGBA, 418 VIEW CLASS ASTC 10x6 RGBA, 418 VIEW CLASS ASTC 10x8 RGBA, 418 VIEW CLASS ASTC 12x10 RGBA, 419 VIEW CLASS ASTC 12x12 RGBA, 419 VIEW CLASS ASTC 4x4 RGBA, 418 VIEW CLASS ASTC 5x4 RGBA, 418 VIEW CLASS ASTC 5x5 RGBA, 418 VIEW CLASS ASTC 6x5 RGBA, 418 VIEW CLASS ASTC 6x6 RGBA, 418 VIEW CLASS ASTC 8x5 RGBA, 418 VIEW CLASS ASTC 8x6 RGBA, 418 VIEW CLASS ASTC 8x8 RGBA, 418 VIEW CLASS EAC R11, 418 VIEW CLASS EAC RG11, 418 VIEW CLASS ETC2 EAC RGBA, 418 VIEW CLASS ETC2 RGB, 418 VIEW CLASS ETC2 RGBA, 418

OpenGL ES 3.2 (May 5, 2022)

INDEX

583

VIEWPORT, 452 Viewport, 348 WAIT FAILED, 35 WaitSync, 33–35, 36, 36, 37, 45, 47, 487 while, 314 WRITE ONLY, 216 ZERO, 189, 373, 382, 387, 388, 394, 459

OpenGL ES 3.2 (May 5, 2022)

Related documents

• OpenSL ES 1.1 Specification — OpenSL ES 1.1 Specification#601026
• OpenVG 1.1 Specification — OpenVG 1.1 Specification#601027
• OpenVX 1.3.2 Specification — The OpenVX Specification#601028
• OpenXR 1.1 Specification — The OpenXR 1.1.62 Specification (with all registered extensions)#601029
• SYCL 2020 Specification — SYCL 2020 Specification (revision 12)#601030