Khronosopen access

OpenGL 4.6 Specification — OpenGL 4.6 Specification

Khronos Group · Khronos Group

Khronos · Standards · License: Open Access

khronos, specification, standard, opengl, 4.6

The OpenGL® Graphics System: A Specification (Version 4.6 (Core Profile) - May 5, 2022) Mark Segal Kurt Akeley Editor (version 1.1): Chris Frazier Editor (versions 1.2-4.6 ): Jon Leech Editor (version 2.0): Pat Brown

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 Specification . . . . . . . . . . . . . . 1.1.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 What is the OpenGL Graphics System? . . . . . . . . . . . . . . 1.2.1 Programmer’s View of OpenGL . . . . . . . . . . . . . . 1.2.2 Implementor’s View of OpenGL . . . . . . . . . . . . . . 1.2.3 Our View . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.4 Fixed-function Hardware and the Compatibility Profile . . 1.2.5 The Deprecation Model . . . . . . . . . . . . . . . . . . 1.3 Related APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.1 OpenGL Shading Language . . . . . . . . . . . . . . . . 1.3.2 OpenGL ES . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.3 OpenGL ES Shading Language . . . . . . . . . . . . . . 1.3.4 SPIR-V . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.5 WebGL . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.6 Window System Bindings . . . . . . . . . . . . . . . . . 1.3.7 OpenCL . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Filing Bug Reports . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 1 1 2 2 2 3 3 3 4 4 4 5 5 6 6 7 7

OpenGL Fundamentals 2.1 Execution Model . . . . . . . . . . . . . . . . . . . . . . . . . . 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 . . . . . . . . . . . . . . . . . . . . . .

8 8 10 12 14 15 16 19 20

CONTENTS

2.4 2.5 2.6

2.3.4 Numeric Representation and Computation . . . . . . . . . 2.3.5 Fixed-Point Data Conversions . . . . . . . . . . . . . . . Rendering Commands . . . . . . . . . . . . . . . . . . . . . . . Context State . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5.1 Generic Context State Queries . . . . . . . . . . . . . . . Objects and the Object Model . . . . . . . . . . . . . . . . . . . 2.6.1 Object Management . . . . . . . . . . . . . . . . . . . . 2.6.2 Buffer Objects . . . . . . . . . . . . . . . . . . . . . . . 2.6.3 Shader Objects . . . . . . . . . . . . . . . . . . . . . . . 2.6.4 Program Objects . . . . . . . . . . . . . . . . . . . . . . 2.6.5 Program Pipeline Objects . . . . . . . . . . . . . . . . . 2.6.6 Texture Objects . . . . . . . . . . . . . . . . . . . . . . . 2.6.7 Sampler Objects . . . . . . . . . . . . . . . . . . . . . . 2.6.8 Renderbuffer Objects . . . . . . . . . . . . . . . . . . . . 2.6.9 Framebuffer Objects . . . . . . . . . . . . . . . . . . . . 2.6.10 Vertex Array Objects . . . . . . . . . . . . . . . . . . . . 2.6.11 Transform Feedback Objects . . . . . . . . . . . . . . . . 2.6.12 Query Objects . . . . . . . . . . . . . . . . . . . . . . . 2.6.13 Sync Objects . . . . . . . . . . . . . . . . . . . . . . . . 2.6.14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

21 24 26 27 27 27 28 29 29 29 30 30 30 31 31 31 31 32 32 32

Dataflow Model

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 Types and Targets . . . . . . . . . . . . . . 4.2.2 Query Object Creation and Activation . . . . . . . . . . . 4.2.3 Query Object Queries . . . . . . . . . . . . . . . . . . . 4.3 Time Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36 36 38 40 41 42 42 43 47 51

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 . . . . . . . . . . . . . . . .

53 54 54 54 54 55

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

iii

5.3

Propagating Changes to Objects . . . . . . . . . . . . . . . . . . 5.3.1 Determining Completion of Changes to an object . . . . . 5.3.2 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.3 Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . .

55 56 57 57

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 . . . . . . . . 6.2.1 Clearing Buffer Object Data Stores . . . . . . . . . . . . 6.3 Mapping and Unmapping Buffer Data . . . . . . . . . . . . . . . 6.3.1 Unmapping Buffers . . . . . . . . . . . . . . . . . . . . . 6.3.2 Effects of Mapping Buffers on Other GL Commands . . . 6.4 Effects of Accessing Outside Buffer Bounds . . . . . . . . . . . . 6.5 Invalidating Buffer Data . . . . . . . . . . . . . . . . . . . . . . 6.6 Copying Between Buffers . . . . . . . . . . . . . . . . . . . . . . 6.7 Buffer Object Queries . . . . . . . . . . . . . . . . . . . . . . . . 6.7.1 Indexed Buffer Object Limits and Binding Queries . . . . 6.8 Buffer Object State . . . . . . . . . . . . . . . . . . . . . . . . .

59 60 62 65 71 73 78 79 79 80 80 81 83 84

Programs and Shaders 7.1 Shader Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Shader Binaries . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 Shader Specialization . . . . . . . . . . . . . . . . . . . . 7.3 Program Objects . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1 Program Interfaces . . . . . . . . . . . . . . . . . . . . . 7.4 Program Pipeline Objects . . . . . . . . . . . . . . . . . . . . . . 7.4.1 Shader Interface Matching . . . . . . . . . . . . . . . . . 7.4.2 SPIR-V Shader Interface Matching . . . . . . . . . . . . 7.4.3 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 Invocation Groups . . . . . . . . . . . . . . . . . . . . . . . . . .

87 88 91 93 94 102 121 125 128 129 130 133 140 144 148 149 150 150 151 153

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

7.10 Subroutine Uniform Variables . . . . . . . . . . . . . . . . . . . 7.11 Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.12 Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.13 Shader Memory Access . . . . . . . . . . . . . . . . . . . . . . . 7.13.1 Shader Memory Access Ordering . . . . . . . . . . . . . 7.13.2 Shader Memory Access Synchronization . . . . . . . . . 7.14 Shader, Program, and Program Pipeline Queries . . . . . . . . . . 7.15 Required State . . . . . . . . . . . . . . . . . . . . . . . . . . . .

154 157 158 159 159 161 166 175

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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4.4 Transfer of Pixel Rectangles . . . . . . . . . . . . . . . . 8.4.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Texture Image Specification . . . . . . . . . . . . . . . . . . . . 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 . . . . . . . . . . . . . . 8.11.4 Texture Image 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 . . . . . . . . .

178 179 183 188 188 189 190 190 190 203 203 206 207 211 218 225 225 233 234 238 242 242 242 243 246 252 253 253 254 256 259

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

8.14.3 Mipmapping . . . . . . . . . . . . . . . . . . . . . . . . 8.14.4 Manual Mipmap Generation . . . . . . . . . . . . . . . . 8.14.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 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 Texture Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.19 Immutable-Format Texture Images . . . . . . . . . . . . . . . . . 8.19.1 Behavior of Immutable-Format Texture Images . . . . . . 8.20 Invalidating Texture Image Data . . . . . . . . . . . . . . . . . . 8.21 Clearing Texture Image Data . . . . . . . . . . . . . . . . . . . . 8.22 Texture State and Proxy State . . . . . . . . . . . . . . . . . . . . 8.23 Texture Comparison Modes . . . . . . . . . . . . . . . . . . . . . 8.23.1 Depth Texture Comparison Mode . . . . . . . . . . . . . 8.24 sRGB Texture Color Conversion . . . . . . . . . . . . . . . . . . 8.25 Shared Exponent Texture Color Conversion . . . . . . . . . . . . 8.26 Texture Image Loads and Stores . . . . . . . . . . . . . . . . . . 8.26.1 Image Unit Queries . . . . . . . . . . . . . . . . . . . . .

264 266 267 267 268 268 269 270 270 270 275 281 281 282 284 287 287 288 289 289 298

Framebuffers and Framebuffer Objects 9.1 Framebuffer Overview . . . . . . . . . . . . . . . . . . . . . . . 9.2 Binding and Managing Framebuffer Objects . . . . . . . . . . . . 9.2.1 Framebuffer Object Parameters . . . . . . . . . . . . . . 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 . . . . . . . . 9.3 Feedback Loops Between Textures and the Framebuffer . . . . . . 9.3.1 Rendering Feedback Loops . . . . . . . . . . . . . . . . . 9.3.2 Texture Copying Feedback Loops . . . . . . . . . . . . . 9.4 Framebuffer Completeness . . . . . . . . . . . . . . . . . . . . . 9.4.1 Framebuffer Attachment Completeness . . . . . . . . . . 9.4.2 Whole Framebuffer Completeness . . . . . . . . . . . . . 9.4.3 Required Framebuffer Formats . . . . . . . . . . . . . . .

299 299 301 305 306 307 312 316 317 317 319 325 325 326 327 328 329 332

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

9.4.4

9.5 9.6 9.7 9.8

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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.6 Triangle Strips . . . . . . . . . . . . . . . . . . . . . . . 10.1.7 Triangle Fans . . . . . . . . . . . . . . . . . . . . . . . . 10.1.8 Separate Triangles . . . . . . . . . . . . . . . . . . . . . 10.1.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1.11 Lines with Adjacency . . . . . . . . . . . . . . . . . . . 10.1.12 Line Strips with Adjacency . . . . . . . . . . . . . . . . . 10.1.13 Triangles with Adjacency . . . . . . . . . . . . . . . . . 10.1.14 Triangle Strips with Adjacency . . . . . . . . . . . . . . . 10.1.15 Separate Patches . . . . . . . . . . . . . . . . . . . . . . 10.1.16 General Considerations For Polygon Primitives . . . . . . 10.1.17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Current Vertex Attribute Values . . . . . . . . . . . . . . . . . . . 10.2.1 Current Generic Attributes . . . . . . . . . . . . . . . . . 10.2.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.3 Vertex Attribute Queries . . . . . . . . . . . . . . . . . . 10.2.4 Required State . . . . . . . . . . . . . . . . . . . . . . . 10.3 Vertex Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.1 Vertex Array Objects . . . . . . . . . . . . . . . . . . . . 10.3.2 Specifying Arrays for Generic Vertex Attributes . . . . . . 10.3.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.4 Vertex Attribute Divisors . . . . . . . . . . . . . . . . . . 10.3.5 Transferring Array Elements . . . . . . . . . . . . . . . . 10.3.6 Primitive Restart . . . . . . . . . . . . . . . . . . . . . . OpenGL 4.6 (Core Profile) - May 5, 2022

332 333 333 334 334 335 337 339 339 339 339 339 340 340 341 341 341 341 341 343 343 344 345 346 346 347 347 349 349 349 349 350 352 359 359 360 361

CONTENTS

vii

10.3.7 Robust Buffer Access . . . . . . . . . . . . . . . . . . . . 10.3.8 Packed Vertex Data Formats . . . . . . . . . . . . . . . . 10.3.9 Vertex Arrays in Buffer Objects . . . . . . . . . . . . . . 10.3.10 Array Indices in Buffer Objects . . . . . . . . . . . . . . 10.3.11 Indirect Commands in Buffer Objects . . . . . . . . . . . 10.4 Drawing Commands Using Vertex Arrays . . . . . . . . . . . . . 10.4.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.5 Vertex Array and Vertex Array Object Queries . . . . . . . . . . . 10.6 Required State . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.9 Conditional Rendering . . . . . . . . . . . . . . . . . . . . . . . 10.10Submission Queries . . . . . . . . . . . . . . . . . . . . . . . . .

362 362 363 363 364 364 377 377 380 381 381 381 383

11 Programmable Vertex Processing 11.1 Vertex Shaders . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1.1 Vertex Attributes . . . . . . . . . . . . . . . . . . . . . . 11.1.2 Vertex Shader Variables . . . . . . . . . . . . . . . . . . 11.1.3 Shader Execution . . . . . . . . . . . . . . . . . . . . . . 11.1.4 Vertex Shader Queries . . . . . . . . . . . . . . . . . . . 11.2 Tessellation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2.1 Tessellation Control Shaders . . . . . . . . . . . . . . . . 11.2.2 Tessellation Primitive Generation . . . . . . . . . . . . . 11.2.3 Tessellation Evaluation Shaders . . . . . . . . . . . . . . 11.2.4 Tessellation Shader Queries . . . . . . . . . . . . . . . . 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 . . . . . . . . . 11.3.5 Geometry Shader Queries . . . . . . . . . . . . . . . . .

384 384 384 390 396 409 409 410 415 424 429 430 430 432 433 433 439

440

13 Fixed-Function Vertex Post-Processing 13.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2 The Last Vertex Processing Stage . . . . . . . . . . . . . . . . . . 13.3 Transform Feedback . . . . . . . . . . . . . . . . . . . . . . . . 13.3.1 Transform Feedback Objects . . . . . . . . . . . . . . . . 13.3.2 Transform Feedback Primitive Capture . . . . . . . . . .

441 441 442 442 442 444

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

viii

13.3.3 Transform Feedback Draw Operations . . . . . . . . . . . 13.4 Primitive Queries . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5 Transform Feedback Overflow Queries . . . . . . . . . . . . . . . 13.6 Flatshading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7 Primitive Clipping . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7.1 Clipping Shader Outputs . . . . . . . . . . . . . . . . . . 13.7.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7.3 Primitive Clipping Queries . . . . . . . . . . . . . . . . . 13.8 Coordinate Transformations . . . . . . . . . . . . . . . . . . . . 13.8.1 Controlling the Viewport . . . . . . . . . . . . . . . . . . 13.9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

450 451 452 452 453 456 457 457 457 458 461

14 Fixed-Function Primitive Assembly and Rasterization 14.1 Discarding Primitives Before Rasterization . . . . . . . . . . . . 14.2 Invariance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3 Antialiasing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3.1 Multisampling . . . . . . . . . . . . . . . . . . . . . . . 14.4 Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.4.1 Basic Point Rasterization . . . . . . . . . . . . . . . . . . 14.4.2 Point Rasterization State . . . . . . . . . . . . . . . . . . 14.4.3 Point Multisample Rasterization . . . . . . . . . . . . . . 14.5 Line Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.5.1 Basic Line Segment Rasterization . . . . . . . . . . . . . 14.5.2 Other Line Segment Features . . . . . . . . . . . . . . . . 14.5.3 Line Rasterization State . . . . . . . . . . . . . . . . . . 14.5.4 Line Multisample Rasterization . . . . . . . . . . . . . . 14.6 Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.6.1 Basic Polygon Rasterization . . . . . . . . . . . . . . . . 14.6.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.6.3 Antialiasing . . . . . . . . . . . . . . . . . . . . . . . . . 14.6.4 Options Controlling Polygon Rasterization . . . . . . . . 14.6.5 Depth Offset . . . . . . . . . . . . . . . . . . . . . . . . 14.6.6 Polygon Multisample Rasterization . . . . . . . . . . . . 14.6.7 Polygon Rasterization State . . . . . . . . . . . . . . . . 14.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.9 Early Per-Fragment Tests . . . . . . . . . . . . . . . . . . . . . . 14.9.1 Pixel Ownership Test . . . . . . . . . . . . . . . . . . . . 14.9.2 Scissor Test . . . . . . . . . . . . . . . . . . . . . . . . . 14.9.3 Multisample Fragment Operations . . . . . . . . . . . . .

462 464 464 464 466 468 470 470 471 471 471 474 476 476 477 477 480 480 480 480 482 483 483 483 483 484 484 486

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

14.9.4 The Early Fragment Test Qualifier . . . . . . . . . . . . . 487 15 Programmable Fragment Processing 15.1 Fragment Shader Variables . . . . . . . . . . . . . . . . . . . . . 15.2 Shader Execution . . . . . . . . . . . . . . . . . . . . . . . . . . 15.2.1 Texture Access . . . . . . . . . . . . . . . . . . . . . . . 15.2.2 Shader Inputs . . . . . . . . . . . . . . . . . . . . . . . . 15.2.3 Shader Outputs . . . . . . . . . . . . . . . . . . . . . . . 15.2.4 Early Fragment Tests . . . . . . . . . . . . . . . . . . . . 15.3 Fragment Shader Queries . . . . . . . . . . . . . . . . . . . . . .

488 488 489 490 491 493 498 498

499

17 Writing Fragments and Samples to the Framebuffer 17.1 Antialiasing Application . . . . . . . . . . . . . . . . . . . . . . 17.2 Multisample Point Fade . . . . . . . . . . . . . . . . . . . . . . . 17.3 Per-Fragment Operations . . . . . . . . . . . . . . . . . . . . . . 17.3.1 Alpha To Coverage . . . . . . . . . . . . . . . . . . . . . 17.3.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.3.3 Stencil Test . . . . . . . . . . . . . . . . . . . . . . . . . 17.3.4 Depth Buffer Test . . . . . . . . . . . . . . . . . . . . . . 17.3.5 Occlusion Queries . . . . . . . . . . . . . . . . . . . . . 17.3.6 Blending . . . . . . . . . . . . . . . . . . . . . . . . . . 17.3.7 sRGB Conversion . . . . . . . . . . . . . . . . . . . . . 17.3.8 Dithering . . . . . . . . . . . . . . . . . . . . . . . . . . 17.3.9 Logical Operation . . . . . . . . . . . . . . . . . . . . . 17.3.10 Additional Multisample Fragment Operations . . . . . . . 17.4 Whole Framebuffer Operations . . . . . . . . . . . . . . . . . . . 17.4.1 Selecting Buffers for Writing . . . . . . . . . . . . . . . . 17.4.2 Fine Control of Buffer Updates . . . . . . . . . . . . . . 17.4.3 Clearing the Buffers . . . . . . . . . . . . . . . . . . . . 17.4.4 Invalidating Framebuffer Contents . . . . . . . . . . . . . 17.4.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

500 500 500 501 501 503 503 504 505 506 513 514 514 515 516 517 521 523 526 528

18 Reading and Copying Pixels 18.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2 Reading Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2.1 Selecting Buffers for Reading . . . . . . . . . . . . . . . 18.2.2 ReadPixels . . . . . . . . . . . . . . . . . . . . . . . . . 18.2.3 Obtaining Pixels from the Framebuffer . . . . . . . . . .

529 529 529 529 531 532

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

18.2.4 Conversion of RGBA values . . . . . . . . . . . . . . . . 18.2.5 Conversion of Depth values . . . . . . . . . . . . . . . . 18.2.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2.8 Final Conversion . . . . . . . . . . . . . . . . . . . . . . 18.2.9 Placement in Pixel Pack Buffer or Client Memory . . . . . 18.3 Copying Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.3.1 Blitting Pixel Rectangles . . . . . . . . . . . . . . . . . . 18.3.2 Copying Between Images . . . . . . . . . . . . . . . . . 18.4 Pixel Draw and Read State . . . . . . . . . . . . . . . . . . . . .

534 534 534 535 535 536 536 538 541 544

19 Compute Shaders 546 19.1 Compute Shader Variables . . . . . . . . . . . . . . . . . . . . . 548 19.2 Compute Shader Queries . . . . . . . . . . . . . . . . . . . . . . 548 20 Debug Output 20.1 Debug Messages . . . . . . . . . . . . . . . . . . . . . . . . . . 20.2 Debug Message Callback . . . . . . . . . . . . . . . . . . . . . . 20.3 Debug Message Log . . . . . . . . . . . . . . . . . . . . . . . . 20.4 Controlling Debug Messages . . . . . . . . . . . . . . . . . . . . 20.5 Externally Generated Messages . . . . . . . . . . . . . . . . . . . 20.6 Debug Groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.7 Debug Labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20.8 Asynchronous and Synchronous Debug Output . . . . . . . . . . 20.9 Debug Output Queries . . . . . . . . . . . . . . . . . . . . . . .

549 550 552 553 553 555 555 557 558 559

21 Special Functions 21.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.5 Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

562 562 562 562 562 562 563

22 Context State Queries 22.1 Simple Queries . . . . . . . . . . . . . . . . . . . . . . . . . . . 22.2 Pointer, String, and Related Context Queries . . . . . . . . . . . . 22.3 Internal Format Queries . . . . . . . . . . . . . . . . . . . . . . . 22.3.1 Supported Operation Queries . . . . . . . . . . . . . . . . 22.3.2 Other Internal Format Queries . . . . . . . . . . . . . . .

564 564 566 569 570 573

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

22.4 Transform Feedback State Queries . . . . . . . . . . . . . . . . . 580 22.5 Indexed Binding State Queries . . . . . . . . . . . . . . . . . . . 581 23 State Tables

582

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 . . . . . . . . . . . . . . . . . . . . . . . .

657 657 658 658 660 662 663

B Corollaries

664

C The OpenGL SPIR-V Execution Environment C.1 Required Versions and Formats . . . . . . . . . . . . . . . . . . . C.2 Valid SPIR-V Built-In Variable Decorations . . . . . . . . . . . . C.3 Valid SPIR-V Capabilities . . . . . . . . . . . . . . . . . . . . . C.4 Validation Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . C.5 Precision and Operation of SPIR-V Instructions . . . . . . . . . . C.6 Precision of GLSL.std.450 Instructions . . . . . . . . . . . . . . .

666 666 666 666 667 669 670

D Compressed Texture Image Formats D.1 RGTC Compressed Texture Image Formats . . . . . . . . . . . . D.2 BPTC Compressed Texture Image Formats . . . . . . . . . . . . D.3 ETC Compressed Texture Image Formats . . . . . . . . . . . . .

674 674 675 675

E Profiles and the Deprecation Model E.1 Core and Compatibility Profiles . . . . . . . . . . . . . . . . . . E.2 Deprecated and Removed Features . . . . . . . . . . . . . . . . . E.2.1 Deprecated But Still Supported Features . . . . . . . . . . E.2.2 Removed Features . . . . . . . . . . . . . . . . . . . . .

677 678 678 678 679

F Version 4.2 F.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . F.2 Deprecation Model . . . . . . . . . . . . . . . . . . . . . . . . . F.3 Changed Tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . F.4 Change Log for Released Specifications . . . . . . . . . . . . . . F.5 Credits and Acknowledgements . . . . . . . . . . . . . . . . . .

684 684 685 685 686 688

OpenGL 4.6 (Core Profile) - May 5, 2022

CONTENTS

xii

G Version 4.3 G.1 Restructuring . . . . . . . . . . . . . . . . . . . . . . . . . . . . G.2 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . G.3 Deprecation Model . . . . . . . . . . . . . . . . . . . . . . . . . G.4 Changed Tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . G.5 Change Log for Released Specifications . . . . . . . . . . . . . . G.6 Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G.7 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . .

691 691 692 693 693 694 701 703

H Version 4.4 H.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . H.2 Deprecation Model . . . . . . . . . . . . . . . . . . . . . . . . . H.3 Change Log for Released Specifications . . . . . . . . . . . . . . H.4 Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H.5 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . .

704 704 705 705 716 717

Version 4.5 I.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . I.2 Deprecation Model . . . . . . . . . . . . . . . . . . . . . . . . . I.3 Change Log for Released Specifications . . . . . . . . . . . . . . I.4 Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I.5 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . .

718 718 719 719 731 732

Version 4.6 J.1 New Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . J.2 Deprecation Model . . . . . . . . . . . . . . . . . . . . . . . . . J.3 Change Log Descriptions . . . . . . . . . . . . . . . . . . . . . . J.4 Change Log for Released Specifications . . . . . . . . . . . . . . J.5 Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J.6 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . .

733 733 734 734 734 743 744

K OpenGL Registry, Header Files, and ARB Extensions K.1 OpenGL Registry . . . . . . . . . . . . . . . . . . . . . . . . . . K.2 Header Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . K.3 ARB and Khronos Extensions . . . . . . . . . . . . . . . . . . . K.3.1 Naming Conventions . . . . . . . . . . . . . . . . . . . . K.3.2 Promoting Extensions to Core Features . . . . . . . . . . K.3.3 Extension Summaries . . . . . . . . . . . . . . . . . . .

745 745 745 746 747 747 747

Index

779

OpenGL 4.6 (Core Profile) - May 5, 2022

List of Figures 3.1

Block diagram of the GL pipeline. . . . . . . . . . . . . . . . . .

8.1 8.2 8.3 8.4

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

190 195 217 261

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. . . . . . . . . . . . . . . . . . . .

337 340 341 343 344

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

416 419 422 424

14.1 Rasterization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2 Visualization of Bresenham’s algorithm. . . . . . . . . . . . . . . 14.3 Rasterization of non-antialiased wide lines. . . . . . . . . . . . . 14.4 The region used in rasterizing an antialiased line segment. . . . .

462 472 474 476

17.1 Per-fragment operations. . . . . . . . . . . . . . . . . . . . . . . 501 18.1 Operation of ReadPixels. . . . . . . . . . . . . . . . . . . . . . . 529

xiii

List of Tables 1.1

OpenGL ES to OpenGL version relationship. . . . . . . . . . . .

2.1 2.2 2.3

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

12 13 18

4.1

Initial properties of a sync object created with FenceSync. . . . .

6.1 6.2 6.3 6.4

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

61 61 68

6.5

75 85

CreateShader type values and the corresponding shader stages. . 89 GetProgramResourceiv properties and supported interfaces . . . 112 OpenGL Shading Language type tokens . . . . . . . . . . . . . . 119 Query targets for default uniform block storage, in components. . 134 Query targets for combined uniform block storage, in components. 134 GetProgramResourceiv properties used by GetActiveUniformsiv. 138 GetProgramResourceiv properties used by GetActiveUniformBlockiv. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 7.8 GetProgramResourceiv properties used by GetActiveAtomicCounterBufferiv. . . . . . . . . . . . . . . . . . . . . . . . . . . 140 7.9 Interfaces for active subroutines . . . . . . . . . . . . . . . . . . 155 7.10 Interfaces for active subroutine uniforms . . . . . . . . . . . . . . 155

7.1 7.2 7.3 7.4 7.5 7.6 7.7

8.1 8.2

PixelStore* parameters. . . . . . . . . . . . . . . . . . . . . . . 189 Pixel data types. . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

xiv

LIST OF TABLES

8.3 8.4 8.5 8.6

Pixel data formats. . . . . . . . . . . . . . . . . . . . . . . . . . Swap Bytes bit ordering. . . . . . . . . . . . . . . . . . . . . . . Packed pixel formats. . . . . . . . . . . . . . . . . . . . . . . . . UNSIGNED_BYTE formats. Bit numbers are indicated for each component. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 UNSIGNED_SHORT formats . . . . . . . . . . . . . . . . . . . . . 8.8 UNSIGNED_INT formats . . . . . . . . . . . . . . . . . . . . . . 8.9 FLOAT_UNSIGNED_INT formats . . . . . . . . . . . . . . . . . . 8.10 Packed pixel field assignments. . . . . . . . . . . . . . . . . . . . 8.11 Conversion from RGBA, depth, and stencil pixel components to internal texture components. . . . . . . . . . . . . . . . . . . . . 8.12 Sized internal color formats. . . . . . . . . . . . . . . . . . . . . 8.13 Sized internal depth and stencil formats. . . . . . . . . . . . . . . 8.14 Generic and specific compressed internal formats. . . . . . . . . . 8.15 Valid texture target parameters . . . . . . . . . . . . . . . . . . . 8.16 Internal formats for buffer textures . . . . . . . . . . . . . . . . . 8.17 Texture parameters and their values. . . . . . . . . . . . . . . . . 8.18 Texture return values. . . . . . . . . . . . . . . . . . . . . . . . . 8.19 Selection of cube map images. . . . . . . . . . . . . . . . . . . . 8.20 Texel location wrap mode application. . . . . . . . . . . . . . . . 8.21 Legal texture targets for TextureView. . . . . . . . . . . . . . . . 8.22 Compatible internal formats for TextureView . . . . . . . . . . . 8.23 Depth texture comparison functions. . . . . . . . . . . . . . . . . 8.24 sRGB texture internal formats. . . . . . . . . . . . . . . . . . . . 8.25 Mapping of image load, store, and atomic texel coordinate components to texel numbers. . . . . . . . . . . . . . . . . . . . . . . . 8.26 Supported image unit formats, with equivalent format layout qualifiers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.27 Texel sizes, compatibility classes, and pixel format/type combinations for each image format. . . . . . . . . . . . . . . . . . . . . 9.1 9.2 9.3

194 194 197 198 199 200 201 202 205 210 211 212 221 238 240 249 253 260 272 273 288 289 293 296 298

Buffer selection for default framebuffer attachment queries . . . . 309 Framebuffer attachment points. . . . . . . . . . . . . . . . . . . . 319 Layer numbers for cube map texture faces. . . . . . . . . . . . . . 336

10.1 Triangles generated by triangle strips with adjacency. . . . . . . . 345 10.2 Alternative ordering to triangles generated by triangle strips with adjacency (see table 10.1). . . . . . . . . . . . . . . . . . . . . . 346 10.3 Vertex array sizes (values per vertex) and data types for generic vertex attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . 354

OpenGL 4.6 (Core Profile) - May 5, 2022

LIST OF TABLES

xvi

10.4 Packed component layout for non-BGRA formats. . . . . . . . . . 10.5 Packed component layout for BGRA format. . . . . . . . . . . . . 10.6 Packed component layout for UNSIGNED_INT_10F_11F_11F_REV format. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.7 Indirect commands and corresponding indirect buffer targets. . . .

362 362 363 365

11.1 Generic attribute components accessed by attribute variables. . . . 385 11.2 Generic attributes and vector types used by column vectors of matrix variables bound to generic attribute index i. . . . . . . . . . . 386 11.3 Scalar and vector vertex attribute types . . . . . . . . . . . . . . . 386 13.1 Transform feedback modes . . . . . . . . . . . . . . . . . . . . . 446 13.2 Provoking vertex selection. . . . . . . . . . . . . . . . . . . . . . 453 15.1 Correspondence of filtered texture components to texture base components. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 17.1 RGB and alpha blend equations. . . . . . . . . . . . . . . . . . . 17.2 Blending functions. . . . . . . . . . . . . . . . . . . . . . . . . . 17.3 Logical operations . . . . . . . . . . . . . . . . . . . . . . . . . 17.4 Buffer selection for the default framebuffer . . . . . . . . . . . . 17.5 Buffer selection for a framebuffer object . . . . . . . . . . . . . . 17.6 DrawBuffers buffer selection for the default framebuffer . . . . .

509 511 515 518 518 519

18.1 PixelStore parameters. . . . . . . . . . . . . . . . . . . . . . . . 18.2 ReadPixels GL data types and reversed component conversion formulas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.3 ReadPixels index masks. . . . . . . . . . . . . . . . . . . . . . . 18.4 Compatible internal formats for copying . . . . . . . . . . . . . .

532 537 538 544

20.1 Sources of debug output messages . . . . . . . . . . . . . . . . . 20.2 Types of debug output messages . . . . . . . . . . . . . . . . . . 20.3 Severity levels of messages . . . . . . . . . . . . . . . . . . . . . 20.4 Object namespace identifiers . . . . . . . . . . . . . . . . . . . .

550 551 551 557

21.1 Hint targets and descriptions . . . . . . . . . . . . . . . . . . . . 563 22.1 Context profile bits . . . . . . . . . . . . . . . . . . . . . . . . . 567 22.2 Internal format targets . . . . . . . . . . . . . . . . . . . . . . . . 570 23.1 State Variable Types . . . . . . . . . . . . . . . . . . . . . . . . . 583 23.2 Current Values and Associated Data . . . . . . . . . . . . . . . . 584 OpenGL 4.6 (Core Profile) - May 5, 2022

LIST OF TABLES

xvii

23.3 Vertex Array Object State (cont.) . . . . . . . . . . . . . . . . . . 23.4 Vertex Array Object State (cont.) . . . . . . . . . . . . . . . . . . 23.5 Vertex Array Data (not in Vertex Array objects) . . . . . . . . . . 23.6 Buffer Object State . . . . . . . . . . . . . . . . . . . . . . . . . 23.7 Transformation state . . . . . . . . . . . . . . . . . . . . . . . . 23.8 Coloring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.9 Rasterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.10Rasterization (cont.) . . . . . . . . . . . . . . . . . . . . . . . . . 23.11Multisampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.12Textures (state per texture unit) . . . . . . . . . . . . . . . . . . . 23.13Textures (state per texture unit) (cont.) . . . . . . . . . . . . . . . 23.14Textures (state per texture object) . . . . . . . . . . . . . . . . . . 23.15Textures (state per texture object) (cont.) . . . . . . . . . . . . . . 23.16Textures (state per texture image) . . . . . . . . . . . . . . . . . . 23.17Textures (state per texture image) (cont.) . . . . . . . . . . . . . . 23.18Textures (state per sampler object) . . . . . . . . . . . . . . . . . 23.19Texture Environment and Generation . . . . . . . . . . . . . . . . 23.20Pixel Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.21Pixel Operations (cont.) . . . . . . . . . . . . . . . . . . . . . . . 23.22Framebuffer Control . . . . . . . . . . . . . . . . . . . . . . . . 23.23Framebuffer (state per target binding point) . . . . . . . . . . . . 23.24Framebuffer (state per framebuffer object) . . . . . . . . . . . . . 23.25Framebuffer (state per attachment point) . . . . . . . . . . . . . . 23.26Renderbuffer (state per target and binding point) . . . . . . . . . . 23.27Renderbuffer (state per renderbuffer object) . . . . . . . . . . . . 23.28Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.29Pixels (cont.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23.30Shader Object State . . . . . . . . . . . . . . . . . . . . . . . . . 23.31Program Pipeline Object State . . . . . . . . . . . . . . . . . . . 23.32Program Object State . . . . . . . . . . . . . . . . . . . . . . . . 23.33Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.34Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.35Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.36Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.37Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.38Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.39Program Object State (cont.) . . . . . . . . . . . . . . . . . . . . 23.40Program Interface State . . . . . . . . . . . . . . . . . . . . . . . 23.41Program Object Resource State . . . . . . . . . . . . . . . . . . . 23.42Program Object Resource State (cont.) . . . . . . . . . . . . . . . OpenGL 4.6 (Core Profile) - May 5, 2022

585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624

LIST OF TABLES

xviii

23.43Vertex and Geometry Shader State . . . . . . . . . . . . . . . . . 625 23.44Query Object State . . . . . . . . . . . . . . . . . . . . . . . . . 626 23.45Image State (state per image unit) . . . . . . . . . . . . . . . . . 627 23.46Atomic Counter Buffer Binding State . . . . . . . . . . . . . . . 628 23.47Shader Storage Buffer Binding State . . . . . . . . . . . . . . . . 629 23.48Transform Feedback Object State . . . . . . . . . . . . . . . . . . 630 23.49Uniform Buffer Binding State . . . . . . . . . . . . . . . . . . . 631 23.50Sync Object State . . . . . . . . . . . . . . . . . . . . . . . . . . 632 23.51Hints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 633 23.52Compute Dispatch State . . . . . . . . . . . . . . . . . . . . . . 634 23.53Implementation Dependent Values . . . . . . . . . . . . . . . . . 635 23.54Implementation Dependent Values (cont.) . . . . . . . . . . . . . 636 23.55Implementation Dependent Values (cont.) . . . . . . . . . . . . . 637 23.56Implementation Dependent Version and Extension Support . . . . 638 23.57Implementation Dependent Vertex Shader Limits . . . . . . . . . 639 23.58Implementation Dependent Tessellation Shader Limits . . . . . . 640 23.59Implementation Dependent Tessellation Shader Limits (cont.) . . 641 23.60Implementation Dependent Geometry Shader Limits . . . . . . . 642 23.61Implementation Dependent Fragment Shader Limits . . . . . . . . 643 23.62Implementation Dependent Compute Shader Limits . . . . . . . . 644 23.63Implementation Dependent Aggregate Shader Limits . . . . . . . 645 23.64Implementation Dependent Aggregate Shader Limits (cont.) . . . 646 23.65Implementation Dependent Aggregate Shader Limits (cont.) . . . 647 23.66Implementation Dependent Aggregate Shader Limits (cont.) . . . 648 23.67Debug Output State . . . . . . . . . . . . . . . . . . . . . . . . . 649 23.68Implementation Dependent Debug Output State . . . . . . . . . . 650 23.69Implementation Dependent Values (cont.) † These queries return the maximum no. of samples for all internal formats required to support multisampled rendering. . . . . . . . . . . 651

23.70Implementation

Dependent

Values

(cont.)

† “ULP” means “Units in the Last Place” or “Unit of Least Precision”. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 652

23.71Internal Format Dependent Values . . . . . . . . . . . . . . . . . 23.72Implementation Dependent Transform Feedback Limits . . . . . . 23.73Framebuffer Dependent Values . . . . . . . . . . . . . . . . . . . 23.74Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . . . . .

653 654 655 656

C.1 Built-in Variable Decorations . . . . . . . . . . . . . . . . . . . . 671 C.2 Valid SPIR-V Capabilities . . . . . . . . . . . . . . . . . . . . . 672 C.3 32-Bit Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . 673 OpenGL 4.6 (Core Profile) - May 5, 2022

LIST OF TABLES

xix

C.4 GLSL.std.450 Precision . . . . . . . . . . . . . . . . . . . . . . . 673 D.1 Mapping of OpenGL RGTC formats to descriptions. . . . . . . . 675 D.2 Mapping of OpenGL BPTC formats to descriptions. . . . . . . . . 675 D.3 Mapping of OpenGL ETC formats to descriptions. . . . . . . . . 676 F.1

New token names . . . . . . . . . . . . . . . . . . . . . . . . . . 686

G.1 New token names . . . . . . . . . . . . . . . . . . . . . . . . . . 694

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 1

Introduction This document, referred to as the “OpenGL Specification” or just “Specification” hereafter, describes the OpenGL 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 computer 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 Registry, located at URL http://www.opengl.org/registry/

1.1

Formatting of the OpenGL Specification

Starting with version 4.3, the OpenGL Specification has undergone major restructuring to focus on programmable shading, and to describe important concepts and objects in the context of the entire API before describing details of their use in the graphics pipeline.

1.1.1 This subsection is only defined in the compatibility profile.

1.1.2 This subsection is only defined in the compatibility profile.

1.2. WHAT IS THE OPENGL GRAPHICS SYSTEM?

1.2

What is the OpenGL Graphics System?

OpenGL (for “Open Graphics Library”) 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 requires that the graphics hardware contain a framebuffer. Many OpenGL calls control drawing geometric objects such as points, lines, 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 and its properties. Some commands explicitly manage the framebuffer.

1.2.1

Programmer’s View of OpenGL

To the programmer, OpenGL is a set of commands that allow the specification of shader programs or shaders, data used by shaders, and state controlling aspects of OpenGL 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 begins with calls to open a window into the framebuffer into which the program will draw. Then, calls are made to allocate an OpenGL context and associate it with the window. Once a context is allocated, OpenGL 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.

1.2.2

Implementor’s View of OpenGL

To the implementor, OpenGL is a set of commands that control the operation of the GPU. Modern GPUs accelerate almost all OpenGL operations, storing data and framebuffer images in GPU memory and executing shaders in dedicated GPU processors. However, OpenGL 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 API, while dividing the work for each OpenGL command

OpenGL 4.6 (Core Profile) - May 5, 2022

1.2. WHAT IS THE OPENGL GRAPHICS SYSTEM?

between the CPU and the graphics hardware as appropriate for the capabilities of the GPU. OpenGL 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. Most of these objects and state are available to the programmer, who can set, manipulate, and query their values through OpenGL commands. Some of it, however, is derived state visible only by the effect it has on how OpenGL operates. One of the main goals of this Specification is to describe OpenGL objects and context state explicitly, to elucidate how they change in response to OpenGL commands, and to indicate what their effects are.

1.2.3

Our View

We view OpenGL as a pipeline having some programmable stages and some statedriven 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.

1.2.4

Fixed-function Hardware and the Compatibility Profile

Older generations of graphics hardware were not programmable using shaders, although they were configurable by setting state controlling specific details of their operation. The compatibility profile of OpenGL continues to support the legacy OpenGL commands developed for such fixed-function hardware, although they are typically implemented by writing shaders which reproduce the operation of such hardware. Fixed-function OpenGL commands and operations are described as alternative interfaces following descriptions of the corresponding shader stages.

1.2.5

The Deprecation Model

Features marked as deprecated in one version of the Specification are expected to be removed in a future version, allowing applications time to transition away from use of deprecated features. The deprecation model is described in more detail, together with a summary of the commands and state deprecated from this version of the API, in appendix E.

OpenGL 4.6 (Core Profile) - May 5, 2022

1.3. RELATED APIS

1.3

Related APIs

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

1.3.1

OpenGL Shading Language

The OpenGL Specification should be read together with a companion document titled The OpenGL Shading Language. The latter document (referred to as the OpenGL Shading Language Specification hereafter) defines the syntax and semantics of the programming language used to write shaders (see chapter 7). Descriptions of shaders later in this document may include references to concepts and terms (such as shading language variable types) defined in the OpenGL Shading Language Specification. OpenGL 4.6 implementations are guaranteed to support version 4.60 of the OpenGL Shading Language. All references to sections of that specification refer to that version. The latest supported version of the shading language may be queried as described in section 22.2. The core profile of OpenGL 4.6 is also guaranteed to support all previous versions of the OpenGL Shading Language back to version 1.40. In some implementations the core profile may also support earlier versions of the OpenGL Shading Language, and may support compatibility profile versions of the OpenGL Shading Language for versions 1.40 and earlier. In this case, errors will be generated when using language features such as compatibility profile built-ins not supported by the core profile API. The #version strings for all supported versions of the OpenGL Shading Language may be queried as described in section 22.2. The OpenGL Shading Language Specification is available in the OpenGL Registry.

1.3.2

OpenGL ES

OpenGL ES is a royalty-free, cross-platform API for full-function 2D and 3D graphics on embedded systems such as mobile phones, game consoles, and vehicles. It consists of well-defined subsets of OpenGL. Each version of OpenGL ES implements a subset of a corresponding OpenGL version as shown in table 1.1. OpenGL ES versions also include some additional functionality taken from later OpenGL versions or specific to OpenGL ES. It is straightforward to port code written for OpenGL ES to corresponding versions of OpenGL.

OpenGL 4.6 (Core Profile) - May 5, 2022

1.3. RELATED APIS OpenGL ES Version OpenGL ES 1.1 OpenGL ES 2.0 OpenGL ES 3.0 OpenGL ES 3.1

5 OpenGL Version it subsets OpenGL 1.5 OpenGL 2.0 OpenGL 3.3 OpenGL 4.3

Table 1.1: OpenGL ES to OpenGL version relationship.

OpenGL and OpenGL ES are developed in parallel within the Khronos Group, which controls both standards. OpenGL 4.3 and 4.5 include additional functionality initially defined in OpenGL ES 3.0 and OpenGL ES 3.1, respectively, for increased compatibility between OpenGL and OpenGL ES implementations. The OpenGL ES Specifications are available in the Khronos API Registry at URL http://www.khronos.org/registry/

1.3.3

OpenGL ES Shading Language

The Specification should also be read together with companion documents titled The OpenGL ES Shading Language. Versions 1.00, 3.00, and 3.10 should be read. These documents define versions of the OpenGL Shading Language designed for implementations of OpenGL ES 2.0, 3.0, and 3.1 respectively, but also supported by OpenGL implementations. References to the OpenGL Shading Language Specification hereafter include both OpenGL and OpenGL ES versions of the Shading Language; references to specific sections are to those sections in version 4.60 of the OpenGL Shading Language Specification. OpenGL 4.6 implementations are guaranteed to support versions 1.00, 3.00, and 3.10 of the OpenGL ES Shading Language. The #version strings for all supported versions of the OpenGL Shading Language may be queried as described in section 22.2. The OpenGL ES Shading Language Specifications are available in the Khronos API Registry.

1.3.4

SPIR-V

SPIR-V is a binary intermediate language for representing graphical-shader stages and compute kernels for multiple Khronos APIs, such as OpenCL, OpenGL, and Vulkan. OpenGL 4.6 (Core Profile) - May 5, 2022

1.3. RELATED APIS

The SPIR-V Specification, and the related SPIR-V Extended Instructions for the OpenGL Shading Language Specification, are available in the Khronos API Registry.

1.3.5

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 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.

1.3.6

Window System Bindings

OpenGL 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. 1.3.6.1

GLX - X Window System Bindings

OpenGL Graphics with the X Window System, referred to as the GLX Specification hereafter, describes the GLX API for use of OpenGL in the X Window System. It is primarily directed at Linux and Unix systems, but GLX implementations also exist for Microsoft Windows, MacOS X, and some other platforms where X is available. The GLX Specification is available in the OpenGL Registry. 1.3.6.2

WGL - Microsoft Windows Bindings

The WGL API supports use of OpenGL with Microsoft Windows. WGL is documented in Microsoft’s MSDN system, although no full specification exists. 1.3.6.3

MacOS X Window System Bindings

Several APIs exist supporting use of OpenGL with Quartz, the MacOS X window system, including CGL, AGL, and NSOpenGLView. These APIs are documented on Apple’s developer website.

OpenGL 4.6 (Core Profile) - May 5, 2022

1.4. FILING BUG REPORTS 1.3.6.4

EGL - Mobile and Embedded Device Bindings

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 supporting OpenGL may be available on some desktop platforms as well. The EGL Specification is available in the Khronos API Registry.

1.3.7

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 buffer and texture objects, and to coordinate control of and transfer of data between OpenCL and OpenGL. This allows applications to split processing of data between OpenCL and OpenGL; for example, by using OpenCL to implement a physics model and then rendering and interacting with the resulting dynamic geometry using OpenGL. The OpenCL Specification is available in the Khronos API Registry.

1.4

Filing Bug Reports

Bug reports on the OpenGL API Specification should be filed in the Khronos OpenGL-API Github repository, located at URL https://github.com/KhronosGroup/OpenGL-API Bug reports on the OpenGL Shading Language Specification should be filed in the Khronos OpenGL-GLSL Github repository, located at URL https://github.com/KhronosGroup/OpenGL-GLSL It is best to file bugs against the most recently released versions, since older versions are usually not updated for bugfixes.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 2

OpenGL Fundamentals This chapter introduces fundamental concepts including the OpenGL 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

Execution Model

OpenGL (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, patch, 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 8

2.1. EXECUTION MODEL

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 state or the framebuffer must be complete before any subsequent command can have any such effects. Data binding occurs on call. This means that data passed to a GL 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, although shaders can be written to generate such objects. In other words, OpenGL 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 be made current to any one of these contexts. 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 4.6 (Core Profile) - May 5, 2022

2.2. COMMAND SYNTAX

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 1.3.6. Allocation and initialization of GL contexts is also done using these companion APIs. GL contexts can 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 is designed to be run on a range of platforms with varying capabilities, memory, 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 (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 commands as functions or procedures using ANSI C syntax. Languages such as C++ and Javascript which 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 4.6 (Core Profile) - May 5, 2022

2.2. COMMAND SYNTAX

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}{ b s i i64 f d ub us ui ui64}{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 ); 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 v3 ); OpenGL 4.6 (Core Profile) - May 5, 2022

2.2. COMMAND SYNTAX Type Descriptor b s i i64 f d ub us ui ui64

12 Corresponding GL Type byte short int int64 float double ubyte ushort uint uint64

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

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 types. 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. The types clampf and clampd are no longer used, replaced by float and double respectively together with specification language requiring parameter clamping1 .

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 1

These changes are backwards-compatible at the compilation and linking levels, and are being propagated to man pages and header files as well.

OpenGL 4.6 (Core Profile) - May 5, 2022

2.2. COMMAND SYNTAX GL Type

13 Description

boolean byte

Bit Width 8 8

ubyte char short

8 8 16

ushort int

16 32

uint fixed

32 32

int64

uint64 sizei enum intptr

64 32 32 ptrbits

sizeiptr sync bitfield half

ptrbits ptrbits 32 16

float clampf double clampd

32 32 64 64

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 16.16 scaled integer Signed two’s complement binary integer Unsigned binary integer Non-negative binary integer size Enumerated binary integer value Signed twos 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] 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 large enough to store any CPU address. sync is defined as an anonymous struct pointer in the C language bindings while intptr and sizeiptr are defined as integer types large enough to hold a pointer. OpenGL 4.6 (Core Profile) - May 5, 2022

2.2. COMMAND SYNTAX

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. • 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, with or without normalization as described for specific commands. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

• If a command returning signed or unsigned 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 DepthRange value, or a depth buffer clear value. In these cases, the query command converts the floating-point value to an integer according to the INT entry of table 18.2; a value not in [−1, 1] converts to an undefined value. • If a command returning floating-point data is called, such as GetFloatv or GetDoublev, a boolean value of TRUE or FALSE is interpreted as 1.0 or 0.0, respectively. An integer value is coerced to floating-point. Single- and double-precision floating-point values are converted as necessary. 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. 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 DepthRange parameters are returned in the order n followed by f.

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 1.3.6.

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

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 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 may define behavior currently considered as an error. Several error generation conditions are implicit in the description of every GL command2 . • 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. 2

For historical reasons, some older errors do not follow the patterns described below, and cannot easily be changed. The explicit error language with each command always controls behavior when in conflict with this section.

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

• If a command that requires an enumerated value is passed a symbolic constant that is never 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 or values pointed to are not allowable for the given command. In some cases, a symbolic constant is allowable for a command, but is forbidden in combination with current GL state and/or a value passed for another parameter of that command. These cases are documented explicitly, and an INVALID_OPERATION error is generated instead. • 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. The Specification attempts to explicitly describe these implicit error conditions (with the exception of CONTEXT_LOST3 and OUT_OF_MEMORY4 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 the 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 detected before any processing based on current state5 . When an error is generated, the GL may also generate a debug output message describing its cause (see chapter 20). 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. 3

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. 4 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. 5 This ensures consistent behavior for commands including language which ignores certain parameters under some conditions, such as glBlitFramebuffer treatment of mask and filter.

OpenGL 4.6 (Core Profile) - 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

enum argument out of range Numeric argument out of range Operation illegal in current state Framebuffer object is not complete Not enough memory left to execute command Command would cause a stack overflow Command would cause a stack underflow

OUT_OF_MEMORY STACK_OVERFLOW STACK_UNDERFLOW

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

Table 2.3: Summary of GL errors

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. 2.3.1.1

No Error Mode

If the GL context was created with the no error mode enabled, then any place where the driver would have generated an error instead has undefined behavior. This could include application termination. All calls to GetError will return NO_ERROR or OUT_OF_MEMORY. OUT_OF_MEMORY errors are a special case because they already allow for undefined behavior and are more difficult for application developers to predict than other errors. OUT_OF_MEMORY errors may be delayed, which can be useful for optimizing multithreaded drivers, but eventually the OUT_OF_MEMORY

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

error will be reported if an implementation would have reported this error. Since behavior of OUT_OF_MEMORY errors are undefined there is some implementation flexibility here. However, this behavior may provide useful information on some implementations that do report OUT_OF_MEMORY without crashing. 6

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. 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_ERROR7 . 6

Jon - “should” be reported? 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. 7

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

If the behavior is LOSE_CONTEXT_ON_RESET, a graphics reset will result in the loss of all context state, requiring the recreation of all associated objects. 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 context8 . 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 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 time9 (al8

The values returned by GetGraphicsResetStatus in the different contexts may differ. 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. 9

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

though such commands may still be executing when Flush returns). The command void Finish( void ); forces all previously issued GL commands to complete. Finish does not return until all effects from such 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 under ”Floating-Point Computation” 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 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. 2.3.4.1

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 floatingpoint 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. OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

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:   (−1)S × 0.0, E = 0, M = 0    M  (−1)S × 2−14 × 210 , E = 0, M 6= 0  V = (−1)S × 2E−15 × 1 + 2M10 , 0 < E < 31    (−1)S × 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

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

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

  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 . 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. 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 −14   × 32 , E = 0, M 6= 0 2 M E−15 V = 2 × 1 + 32 , 0 < E < 31    Inf , E = 31, M = 0     NaN , E = 31, M 6= 0

OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

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. 2.3.4.5

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 OpenGL 4.6 (Core Profile) - May 5, 2022

2.3. COMMAND EXECUTION

the 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. For framebuffer and renderbuffer alpha components, b must be at least 2 if the buffer does not contain an alpha component, or if there is only one bit of alpha in the buffer. The signed and unsigned fixed-point representations are assumed to be b-bit binary twos-complement integers and binary unsigned integers, respectively. 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 floating-point value f is performed using c f = max , −1.0 . (2.2) 2b−1 − 1 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 values10 , 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 10

This is a behavior change in OpenGL 4.2. 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 4.6 (Core Profile) - May 5, 2022

2.4. RENDERING COMMANDS

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 section 2.2.2) and returning integers11 , 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 sometimes treated specially by other GL operations such as conditional rendering (see section 10.9). Such commands are called rendering commands, and include the drawing commands *Draw* (see section 10.4), as well as these additional commands: • BlitFramebuffer (see section 18.3.1) • Clear (see section 17.4.3) • ClearBuffer* (see section 17.4.3.1) • DispatchCompute* (see section 19) 11

This is a behavior change in OpenGL 4.2. 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 4.6 (Core Profile) - May 5, 2022

2.5. CONTEXT STATE

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 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.

2.5.1

Generic Context State Queries

Context state queries are described in detail in chapter 22.

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. OpenGL 4.6 (Core Profile) - May 5, 2022

2.6. OBJECTS AND THE OBJECT MODEL

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). 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. Objects may also be created directly by functions that return a new name or names representing a freshly initialized object. Some functions return a single object name directly whereas others are able to create a large number of new objects, returning their names in an array. Examples of the former are CreateProgram for program objects and FenceSync for fence sync objects. Examples of the latter are CreateBuffers, CreateTextures and CreateVertexArrays for buffers, textures and vertex arrays, respectively. 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, OpenGL 4.6 (Core Profile) - May 5, 2022

2.6. OBJECTS AND THE OBJECT MODEL

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 1.3.6. 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. 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. Under certain circumstances, the data store of a buffer object may be shared between the client and server and accessed simultaneously by both. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

2.6. OBJECTS AND THE OBJECT MODEL

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 a data store containing a collection of texture images built from arrays of image elements. The image elements are 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. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

2.6. OBJECTS AND THE OBJECT MODEL

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 shared12 . 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. 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. 12

Framebuffer objects created with the commands defined by the GL_EXT_framebuffer_object extension are defined to be shared, while FBOs created with commands defined by the OpenGL core or GL_ARB_framebuffer_object extension are defined to not be shared. Undefined behavior results when using FBOs created by EXT commands through non-EXT interfaces, or vice-versa.

OpenGL 4.6 (Core Profile) - May 5, 2022

2.6. OBJECTS AND THE OBJECT MODEL

Transform feedback objects are container objects including references to buffer objects, and are not shared. They are described in detail in section 13.3.1.

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

2.6.14 This subsection is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - 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 a single input primitive. 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, stenciling, and other logical operations on fragment values. 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

34 of how the GL is implemented, and we present it only as a means to organize the various operations of the GL.

OpenGL 4.6 (Core Profile) - May 5, 2022

Tessellation Control Shader

Vertex Buffer Object b

Figure 3.1. Block diagram of the GL pipeline.

OpenGL 4.6 (Core Profile) - May 5, 2022 t – Texture Binding

b – Buffer Binding

Programmable Stage

Fixed Function Stage

Legend

Transform Feedback Buffer b

Vertex Shader

Draw Indirect Buffer b

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 Puller

Element Array Buffer b

From Application

Pixel Pack Buffer b

Texture Image t

Pixel Unpack Buffer b

Compute Shader

Dispatch

From Application

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. 36

4.1. SYNC OBJECTS AND FENCES

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 4.6 (Core Profile) - May 5, 2022

4.1. SYNC OBJECTS AND FENCES

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 4.6 (Core Profile) - May 5, 2022

4.1. SYNC OBJECTS AND FENCES

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 pname 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 4.6 (Core Profile) - May 5, 2022

4.1. SYNC OBJECTS AND FENCES 4.1.1.1

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. 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 or unsignaled. The initial values of sync object state are defined as specified by FenceSync. 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 4.6 (Core Profile) - May 5, 2022

4.1. SYNC OBJECTS AND FENCES

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 until it is no longer associated with any fence command and no longer blocking any *WaitSync command. OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

4.2

Query Objects and Asynchronous Queries

Asynchronous queries provide a mechanism to return information about the processing of a sequence of GL commands.

4.2.1

Query Object Types and Targets

Query types supported by the GL include: • Primitive queries with a target of PRIMITIVES_GENERATED (see section 13.4) return information on the number of primitives processed by the GL. There may be at most the value of MAX_VERTEX_STREAMS active queries of this type. • Primitive queries with a target of TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN (see section 13.4) return information on the number of primitives written to one or more buffer objects. There may be at most the value of MAX_VERTEX_STREAMS active queries of this type. • Transform feedback overflow queries with a target of TRANSFORM_FEEDBACK_OVERFLOW or TRANSFORM_FEEDBACK_STREAM_OVERFLOW return information on whether or not transform feedback overflow happened for one or more streams (see section 13.5). • Occlusion queries with a target of SAMPLES_PASSED, ANY_SAMPLES_PASSED or ANY_SAMPLES_PASSED_CONSERVATIVE (see section 17.3.5) count the number of fragments or samples that pass the depth test, or set a boolean to true when any fragments or samples pass the depth test. There may be at most one active query of this type. • Time elapsed queries with a target of TIME_ELAPSED (see section 4.3) record the amount of time needed to fully process a sequence of commands. There may be at most one active query of this type. • Timer queries with a target of TIMESTAMP (see section 4.3) record the current time of the GL. There may be at most one active query of this type. • Submission queries with a target of VERTICES_SUBMITTED and PRIMITIVES_SUBMITTED return information on the number of vertices and primitives transferred to the GL, respectively (see section 10.10). • Vertex shader queries with a target of VERTEX_SHADER_INVOCATIONS return information on the number of times the vertex shader has been invoked (see section 11.1.4). OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

• Tessellation shader queries with a target of TESS_CONTROL_SHADER_PATCHES and TESS_EVALUATION_SHADER_INVOCATIONS return information on the number of patches processed by the tessellation control shader stage and the number of times the tessellation evaluation shader has been invoked, respectively (see section 11.2.4). • Geometry shader queries with a target of GEOMETRY_SHADER_INVOCATIONS and GEOMETRY_SHADER_PRIMITIVES_EMITTED return information on the number of times the geometry shader has been invoked and the number of primitives it emitted (see section 11.3.5). • Primitive clipping queries with a target of CLIPPING_INPUT_PRIMITIVES and CLIPPING_OUTPUT_PRIMITIVES return information on the number of primitives that were processed in the primitive clipping stage and the number of primitives that were output by the primitive clipping stage and are further processed by the rasterization stage, respectively (see section 13.7.3). • Fragment shader queries with a target of FRAGMENT_SHADER_INVOCATIONS return information on the number of times the fragment shader has been invoked (see section 15.3). • Compute shader queries with a target of COMPUTE_SHADER_INVOCATIONS return information on the number of times the compute shader has been invoked (see section 19.2).

4.2.2

Query Object Creation and Activation

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 4.2.3 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, BeginQueryIndexed, or QueryCounter (see section 4.3). OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

Errors An INVALID_VALUE error is generated if n is negative. Query objects may also be created with the command void CreateQueries( enum target, sizei n, uint *ids ); CreateQueries returns n previously unused query object names in ids, each representing a new query object with the specified target. target must be one of the query object targets described in section 4.2.1. The initial state of the resulting query object is that the result is marked available (the value of QUERY_RESULT_AVAILABLE for the query object is TRUE) 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 query object targets described in section 4.2.1. An INVALID_VALUE error is generated if n is negative. 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, other than timer queries of type TIMESTAMP, supported by the GL has an active query object name for each of the possible active queries. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

void BeginQueryIndexed( enum target, uint index, uint id ); target indicates the type of query to be performed. The valid values of target are discussed in more detail in subsequent sections. 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. Objects specified with target SAMPLES_PASSED may only be reused for that target. index is the index of the query, and must be between zero and a target-specific maximum. 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. The active query object name for target and index is set to id. Errors An INVALID_ENUM error is generated if target is TIMESTAMP, or is not one of the query object targets described in section 4.2.1. An INVALID_VALUE error is generated if target is PRIMITIVES_GENERATED, TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not in the range zero to the value of MAX_VERTEX_STREAMS minus one. An INVALID_VALUE error is generated if target is a valid target other than PRIMITIVES_GENERATED, TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not zero. An INVALID_OPERATION error is generated if id is not a name returned from a previous call to GenQueries or CreateQueries, or if such a name has since been deleted 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 and index

OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

• the active query object for conditional rendering (see section 10.9). An INVALID_OPERATION error is generated if the active query object name for target and index is non-zero. The command void BeginQuery( enum target, uint id ); is equivalent to BeginQueryIndexed(target, 0, id); The command void EndQueryIndexed( enum target, uint index ); marks the end of the sequence of commands to be tracked for the active query specified by target and index. target and index have the same meaning as for BeginQueryIndexed. The corresponding active query object is updated to indicate that query results are not available, and the active query object name for target and index is reset to zero. When the commands issued prior to EndQueryIndexed have completed and a final query result is available, the query object active when EndQueryIndexed was called is updated to contain the query result and to indicate that the query result is available. Errors An INVALID_ENUM error is generated if target is TIMESTAMP, or is not one of the query object targets described in section 4.2.1. An INVALID_VALUE error is generated if target is PRIMITIVES_GENERATED TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not in the range zero to the value of MAX_VERTEX_STREAMS minus one. An INVALID_VALUE error is generated if target is a valid target other than PRIMITIVES_GENERATED, TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not zero. An INVALID_OPERATION error is generated if the active query object name for target and index is zero.

OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

The command void EndQuery( enum target ); is equivalent to EndQueryIndexed(target, 0); 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 is implementation-dependent and may be determined as described in section 4.2.3. The initial state of a query object depends on whether it was created with CreateQueries or BeginQueryIndexed, as described above. 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. The necessary state for each possible active query target and index 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.

4.2.3

Query Object Queries

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 may be queried with the command void GetQueryIndexediv( enum target, uint index, enum pname, int *params ); target and index specify the active query, and have the same meaning as for BeginQueryIndexed. If pname is CURRENT_QUERY, the name of the currently active query object for target and index, or zero if no query is active, will be placed in params. If target is TIMESTAMP, zero is always returned. OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

If pname is QUERY_COUNTER_BITS, index is ignored and the implementationdependent number of bits used to hold the query result for target will be placed in params. The number of query counter bits may be zero, in which case the counter contains no useful information. For primitive queries (PRIMITIVES_GENERATED and TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN) if the number of bits is non-zero, the minimum number of bits allowed is 32. For transform feedback overflow queries (TRANSFORM_FEEDBACK_OVERFLOW and TRANSFORM_FEEDBACK_STREAM_OVERFLOW) if the number of bits is non-zero, the minimum number of bits is 1. For occlusion queries with target ANY_SAMPLES_PASSED or ANY_SAMPLES_PASSED_CONSERVATIVE, if the number of bits is non-zero, the minimum number of bits is 1. For occlusion queries with target SAMPLES_PASSED, if the number of bits is non-zero, the minimum number of bits allowed is 32. For timer queries (target TIME_ELAPSED and TIMESTAMP), if the number of bits is non-zero, the minimum number of bits allowed is 30. This will allow at least one second of timing. For pipeline statistics queries (target VERTICES_SUBMITTED, PRIMITIVES_SUBMITTED, VERTEX_SHADER_INVOCATIONS, TESS_CONTROL_SHADER_PATCHES, TESS_EVALUATION_SHADER_INVOCATIONS, GEOMETRY_SHADER_INVOCATIONS, FRAGMENT_SHADER_INVOCATIONS, COMPUTE_SHADER_INVOCATIONS, GEOMETRY_SHADER_PRIMITIVES_EMITTED, CLIPPING_INPUT_PRIMITIVES, and CLIPPING_OUTPUT_PRIMITIVES), if the number of bits is non-zero, the minimum number of bits allowed is 32. Errors An INVALID_ENUM error is generated if target is not one of the query object targets described in section 4.2.1. An INVALID_VALUE error is generated if target is PRIMITIVES_GENERATED, TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not in the range zero to the value of MAX_VERTEX_STREAMS minus one. An INVALID_VALUE error is generated if target is a valid target other than PRIMITIVES_GENERATED, TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW, and index is not zero. An INVALID_ENUM error is generated if pname is not CURRENT_QUERY

OpenGL 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

or QUERY_COUNTER_BITS. The command void GetQueryiv( enum target, enum pname, int *params ); is equivalent to GetQueryIndexediv(target, 0, pname, params); The state of a query object may be queried with the commands void GetQueryObjectiv( uint id, enum pname, int *params ); void GetQueryObjectuiv( uint id, enum pname, uint *params ); void GetQueryObjecti64v( uint id, enum pname, int64 *params ); void GetQueryObjectui64v( uint id, enum pname, uint64 *params ); void GetQueryBufferObjectiv( uint id, uint buffer, enum pname, intptr offset ); void GetQueryBufferObjectuiv( uint id, uint buffer, enum pname, intptr offset ); void GetQueryBufferObjecti64v( uint id, uint buffer, enum pname, intptr offset ); void GetQueryBufferObjectui64v( uint id, uint buffer, enum pname, intptr offset ); id is the name of a query object. For GetQueryBufferObject*, buffer is the name of a buffer object and offset is an offset into buffer at which the queried value is written. For GetQueryObject*, the queried value may be returned either in client memory or in a buffer object. If zero is bound to the current query result buffer binding point (see QUERY_RESULT in section 6.1), then params is treated as a pointer into client memory at which the queried value is written. Otherwise, params is treated as an offset into the query result buffer object at which the queried value is written. 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 4.6 (Core Profile) - May 5, 2022

4.2. QUERY OBJECTS AND ASYNCHRONOUS QUERIES

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 eventually. If pname is QUERY_TARGET, then the target of the query object is returned as a single integer. If pname is QUERY_RESULT, then the query object’s result value is returned as a single integer. 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. If the number of query counter bits for target is zero, then the result is returned as a single integer with the value zero. Querying QUERY_RESULT for any given query object forces that query to complete within a finite amount of time. If pname is QUERY_RESULT_NO_WAIT, then the query object’s result value is returned as a single integer if the result is available at the time of the state query. If the result is not available then the query return value is not written. If multiple queries are issued using the same object name prior to calling these query commands, 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_OPERATION error is generated by GetQueryBufferObject* if buffer is not the name of an existing buffer object. An INVALID_ENUM error is generated if pname is not QUERY_RESULT, QUERY_RESULT_AVAILABLE, QUERY_RESULT_NO_WAIT, or QUERY_TARGET. An INVALID_OPERATION error is generated if the query writes to a buffer object, and the specified buffer offset would cause data to be written beyond the bounds of that buffer object. An INVALID_VALUE error is generated by GetQueryBufferObject* if offset is negative.

OpenGL 4.6 (Core Profile) - May 5, 2022

4.3. TIME QUERIES

4.3

Time Queries

Query objects may also be used to track the amount of time needed to fully complete a set of GL commands (a time elapsed query), or to determine the current time of the GL (a timer query). When BeginQuery and EndQuery are called with a target of TIME_ELAPSED, the GL prepares to start and stop the timer used for time elapsed queries. The timer is started or stopped when the effects from all previous commands on the GL client and server state and the framebuffer have been fully realized. The BeginQuery and EndQuery commands may return before the timer is actually started or stopped. When the time elapsed query timer is finally stopped, the elapsed time (in nanoseconds) is written to the corresponding query object as the query result value, and the query result for that object is marked as available. A timer query object is created with the command void QueryCounter( uint id, enum target ); target must be TIMESTAMP. If id is an unused query object name, the name is marked as used and associated with a new query object of type TIMESTAMP. Otherwise id must be the name of an existing query object of that type. Alternatively, TIMESTAMP query objects can be created by calling CreateQueries with target set to TIMESTAMP. When QueryCounter is called, the GL records the current time into the corresponding query object. The time is recorded after all previous commands on the GL client and server state and the framebuffer have been fully realized. When the time is recorded, the query result for that object is marked available. Timer queries can be used within a BeginQuery / EndQuery block where the target is TIME_ELAPSED, and it does not affect the result of that query object. The current time of the GL may be queried by calling GetIntegerv or GetInteger64v with the symbolic constant TIMESTAMP. This will return the GL time after all previous commands have reached the GL server but have not yet necessarily executed. By using a combination of this synchronous get command and the asynchronous timestamp query object target, applications can measure the latency between when commands reach the GL server and when they are realized in the framebuffer. Errors An INVALID_ENUM error is generated if target is not TIMESTAMP. An INVALID_OPERATION error is generated if id is not a name returned

OpenGL 4.6 (Core Profile) - May 5, 2022

4.3. TIME QUERIES

from a previous call to GenQueries, or if such a name has since been deleted with DeleteQueries. An INVALID_OPERATION error is generated if id is the name of an existing query object whose type is not TIMESTAMP.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 5

Shared Objects and Multiple Contexts This chapter describes special considerations for objects shared between multiple OpenGL contexts, including deletion behavior and how changes to shared objects are propagated between contexts. Objects that may 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). Some of these objects may contain views (alternate interpretations) of part or all of the data store of another object. Examples are texture buffer objects, which contain a view of a buffer object’s data store, and texture views, which contain a view of another texture object’s data store. Views act as references on the object whose data store is viewed. 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 versions or different profiles of the same OpenGL version (see appendix E). 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

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.

5.1.2

Automatic Unbinding of Deleted Objects

When a buffer, texture, or renderbuffer object is 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, 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, texture, sampler, renderbuffer, query, or sync 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, texture, sampler, or renderbuffer object is in use if any of the following conditions are satisfied: • the object is attached to any container object • the object is bound to a context bind point in any context • any other object contains a view of the data store of the object. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

5.2. SYNC OBJECTS AND MULTIPLE CONTEXTS

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 while it is in use (as defined above). Following its deletion, the object’s name may be returned by Gen* or Create* commands. The underlying object state and data for such a deleted, but still in use object may still be read and written by the GL, but cannot be accessed by name. 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, context binding points, views, fence commands, active queries, etc. are removed. 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.

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 signaled, 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

5.3. PROPAGATING CHANGES TO OBJECTS

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: • 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, even if T’s data store was modified via a different view of the data store.

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 command 1 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 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 4.6 (Core Profile) - May 5, 2022

5.3. PROPAGATING CHANGES TO OBJECTS

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: • 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 updates 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 13.3.3), results will be undefined.

OpenGL 4.6 (Core Profile) - May 5, 2022

5.3. PROPAGATING CHANGES TO OBJECTS

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. 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. 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 4.6 (Core Profile) - 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 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. In addition to generating an unused name and then binding it to a target with BindBuffer, a buffer object may also be created with the command void CreateBuffers( sizei n, uint *buffers ); CreateBuffers returns n previously unused buffer names in buffers, each representing a new buffer object initialized as if it had been bound to an unspecified target.

6.1. CREATING AND BINDING BUFFER OBJECTS

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. 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 a name returned by GenBuffers 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, 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 a name returned by GenBuffers to any of the valid targets are formally equivalent. OpenGL 4.6 (Core Profile) - May 5, 2022

6.1. CREATING AND BINDING BUFFER OBJECTS

Target name

Purpose

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

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

Described in section(s) 10.3.9 7.7 6.6 6.6 19 10.3.11 10.3.10 10.4 18.2, 22 8.4 4.2.3 7.8 8.9 13.3 7.6.2

Table 6.1: Buffer object binding targets. Name

Type int64 enum

BUFFER_SIZE BUFFER_USAGE

BUFFER_ACCESS

enum

BUFFER_ACCESS_FLAGS BUFFER_IMMUTABLE_STORAGE BUFFER_MAPPED BUFFER_MAP_POINTER BUFFER_MAP_OFFSET BUFFER_MAP_LENGTH BUFFER_STORAGE_FLAGS

int boolean boolean void* int64 int64 int

Initial Value 0 STATIC_DRAW

READ_WRITE

0 FALSE FALSE NULL

0 0 0

Table 6.2: Buffer object parameters and their values.

OpenGL 4.6 (Core Profile) - May 5, 2022

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

See section 6.3 TRUE, FALSE TRUE, FALSE address any non-negative integer any non-negative integer See section 6.2

6.1. CREATING AND BINDING BUFFER OBJECTS

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 section 5.1. Initially, each buffer object target is bound to zero. 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 buffer is not zero or a name returned from a previous call to GenBuffers, or if such a name has since been deleted with DeleteBuffers. An INVALID_OPERATION error is generated by client attempts to modify or query buffer object state for a target bound to zero, since there is no buffer object corresponding to the name zero,

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 ); target must be one of ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BUFFER, TRANSFORM_FEEDBACK_BUFFER or UNIFORM_BUFFER. Additional 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 MapBuffer. Both commands bind the OpenGL 4.6 (Core Profile) - May 5, 2022

6.1. CREATING AND BINDING BUFFER OBJECTS 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, 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. 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 13.3.2. 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 6.7.1. An INVALID_OPERATION error is generated if buffer is not zero 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 by BindBufferRange if 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 6.7.1. The commands void BindBuffersBase( enum target, uint first, sizei count, OpenGL 4.6 (Core Profile) - May 5, 2022

6.1. CREATING AND BINDING BUFFER OBJECTS

const uint *buffers ); void BindBuffersRange( enum target, uint first, sizei count, const uint *buffers, const intptr *offsets, const sizeiptr *sizes ); bind count existing buffer objects to bindings numbered first through f irst + count − 1 in the array of buffer binding points corresponding to target. If buffers is not NULL, it specifies an array of count values, each of which must be zero or the name of an existing buffer object. For BindBuffersRange, offsets and sizes specify arrays of count values indicating the range of each buffer to bind. If buffers is NULL, all bindings from first to f irst + count − 1 are reset to their unbound (zero) state. In this case, the offsets and sizes associated with the binding points are set to default values, ignoring offsets and sizes. BindBuffersBase is equivalent (assuming no errors are generated) to: for (i = 0; i < count; i++) { if (buf f ers == NULL) { BindBufferBase(target, f irst + i, 0); } else { BindBufferBase(target, f irst + i, buf f ers[i]); } } except that the single general buffer binding corresponding to target is unmodified, and that buffers will not be created if they do not exist. BindBuffersRange is equivalent (assuming no errors are generated) to: for (i = 0; i < count; i++) { if (buf f ers == NULL) { BindBufferRange(target, f irst + i, 0, 0, 0); } else { BindBufferRange(target, f irst + i, buf f ers[i], of f sets[i], sizes[i]); } } except that the single general buffer binding corresponding to target is unmodified, and that buffers will not be created if they do not exist. The values specified in buffers, offsets, and sizes will be checked separately for each binding point. When values for a specific binding point are invalid, the state for that binding point will be unchanged and an error will be generated. When OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

such an error occurs, state for other binding points will still be changed if their corresponding values are valid. Errors An INVALID_ENUM error is generated if target is not one of the targets listed above. An INVALID_OPERATION error is generated if f irst + count is greater than the number of target-specific indexed binding points, as described in section 6.7.1. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if any value in buffers is not zero or the name of an existing buffer object. An INVALID_VALUE error is generated by BindBuffersRange if any value in offsets is less than zero (per binding). An INVALID_VALUE error is generated by BindBuffersRange if any value in sizes is less than or equal to zero (per binding). An INVALID_VALUE error is generated by BindBuffersRange if any pair of values in offsets and sizes does not respectively satisfy the constraints described for those parameters for the specified target, as described in section 6.7.1 (per binding).

6.2

Creating and Modifying Buffer Object Data Stores

The data store of a buffer object is created by calling one of void BufferStorage( enum target, sizeiptr size, const void *data, bitfield flags ); void NamedBufferStorage( uint buffer, sizeiptr size, const void *data, bitfield flags ); For BufferStorage, the buffer object is that bound to target, which must be one of the values listed in table 6.1. For NamedBufferStorage, buffer is the name of the buffer object. size is the size of the data store in basic machine units, and flags containing a bitfield describing the intended usage of the data store. The data store of the buffer object is allocated as a result of these commands, and cannot be de-allocated until the buffer is deleted with a call to DeleteBuffers. Such a store may not be re-allocated through further calls to *BufferStorage or BufferData.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

data specifies the address in client memory of the data that should be used to initialize the buffer object’s data store. If data is NULL, the data store of the buffer object is created, but contains undefined data. Otherwise, data should point to an array of at least size basic machine units. flags is the bitwise OR of flags describing the intended usage of the buffer object’s data store by the application. Valid flags and their meanings are as follows: DYNAMIC_STORAGE_BIT The contents of the data store may be updated after cre-

ation through calls to BufferSubData. If this bit is not set, the buffer content may not be directly updated by the client. The data argument may be used to specify the initial content of the buffer’s data store regardless of the presence of the DYNAMIC_STORAGE_BIT. Regardless of the presence of this bit, buffers may always be updated with server-side calls such as CopyBufferSubData and ClearBufferSubData. MAP_READ_BIT The data store may be mapped by the client for read access and a

pointer in the client’s address space obtained that may be read from. MAP_WRITE_BIT The data store may be mapped by the client for write access and

a pointer in the client’s address space obtained that may be written to. MAP_PERSISTENT_BIT The client may request that the server read from or write

to the buffer while it is mapped. The client’s pointer to the data store remains valid so long as the data store is mapped, even during execution of drawing or dispatch commands. MAP_COHERENT_BIT Shared access to buffers that are simultaneously mapped

for client access and are used by the server will be coherent, so long as that mapping is performed using MapBufferRange or MapNamedBufferRange. That is, data written to the store by either the client or server will be visible to any subsequently issued GL commands with no further action taken by the application. In particular, • If MAP_COHERENT_BIT is not set and the client performs a write followed by a call to one of the FlushMapped*BufferRange commands with a range including the written range, then in subsequent commands the server will see the writes. • If MAP_COHERENT_BIT is set and the client performs a write, then in subsequent commands the server will see the writes.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

• If MAP_COHERENT_BIT is not set and the server performs a write, the application must call MemoryBarrier with the CLIENT_MAPPED_BUFFER_BARRIER_BIT set and then call FenceSync with SYNC_GPU_COMMANDS_COMPLETE (or Finish). Then the CPU will see the writes after the sync is complete. • If MAP_COHERENT_BIT is set and the server does a write, the application must call FenceSync with SYNC_GPU_COMMANDS_COMPLETE (or Finish). Then the CPU will see the writes after the sync is complete. CLIENT_STORAGE_BIT When all other criteria for the buffer storage allocation

are met, this bit may be used by an implementation to determine whether to use storage that is local to the server or to the client to serve as the backing store for the buffer. If flags contains MAP_PERSISTENT_BIT, it must also contain at least one of MAP_READ_BIT or MAP_WRITE_BIT. It is an error to specify MAP_COHERENT_BIT without also specifying MAP_PERSISTENT_BIT. BufferStorage and NamedBufferStorage delete any existing data store, and set 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. Errors An INVALID_OPERATION error is generated by BufferStorage if zero is bound to target. An INVALID_OPERATION error is generated by NamedBufferStorage if buffer is not the name of an existing buffer object. An INVALID_VALUE error is generated if size is less than or equal to zero. An INVALID_VALUE error is generated if flags has any bits set other than those defined above. An INVALID_VALUE error is generated if flags contains MAP_PERSISTENT_BIT but does not contain at least one of MAP_READ_BIT or MAP_WRITE_BIT. An INVALID_VALUE error is generated if flags contains MAP_COHERENT_BIT, but does not also contain MAP_PERSISTENT_BIT. An INVALID_OPERATION error is generated if the BUFFER_-

OpenGL 4.6 (Core Profile) - May 5, 2022

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

Value for BufferData size usage

Value for *BufferStorage size

READ_WRITE

DYNAMIC_DRAW READ_WRITE

FALSE FALSE NULL

TRUE FALSE NULL

0 0

0 0 flags

MAP_READ_BIT | MAP_WRITE_BIT | DYNAMIC_STORAGE_BIT

Table 6.3: Buffer object state after calling BufferData, BufferStorage, or NamedBufferStorage.

IMMUTABLE_STORAGE flag of the buffer bound to target is TRUE.

A mutable data store may be allocated for a buffer object with the commands void BufferData( enum target, sizeiptr size, const void *data, enum usage ); void NamedBufferData( uint buffer, sizeiptr size, const void *data, enum usage ); For BufferData, the buffer object is that bound to target, which must be one of the targets listed in table 6.1. For NamedBufferData, buffer is the name of the buffer object. size is the size of the data store in basic machine units, data points to the source data in client memory, and usage indicates the expected application usage pattern of the data store. 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. In the following descriptions, a buffer’s data store is sourced when if 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. OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

The values are: STREAM_DRAW The data store contents will be specified once by the application,

and sourced at most a few times. 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 and NamedBufferData delete any existing data store, and set 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. Clients must align data elements consistent 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 . Calling *BufferData is equivalent to calling BufferStorage with target, size and data as specified, and flags set to the logical OR of DYNAMIC_STORAGE_BIT, MAP_READ_BIT and MAP_WRITE_BIT. The GL will use the value of the usage parameter to *BufferData as a hint to further determine the intended use of the buffer. OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

However, BufferStorage allocates immutable storage whereas *BufferData allocates mutable storage. Thus, when a buffer’s data store is allocated through a call to *BufferData, the buffer’s BUFFER_IMMUTABLE_STORAGE flag is set to FALSE. Errors An INVALID_OPERATION error is generated by BufferData if zero is bound to target. An INVALID_OPERATION error is generated by NamedBufferData if buffer is not the name of an existing buffer object. An INVALID_VALUE error is generated if size is negative. An INVALID_ENUM error is generated by BufferData if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated if the BUFFER_IMMUTABLE_STORAGE flag of the buffer object is TRUE. 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 commands void BufferSubData( enum target, intptr offset, sizeiptr size, const void *data ); void NamedBufferSubData( uint buffer, intptr offset, sizeiptr size, const void *data ); For BufferSubData, target specifies the target to which the buffer object is bound, and must be one of the values listed in table 6.1. For NamedBufferSubData, buffer is the name of the buffer object. 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 by BufferSubData if zero is bound to target. An INVALID_OPERATION error is generated by NamedBufferSubData if buffer is not the name of an existing buffer object. An INVALID_ENUM error is generated by BufferSubData if target is not

OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

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 object. An INVALID_OPERATION error is generated if any part of the specified buffer range is mapped with MapBufferRange or MapBuffer (see section 6.3), unless it was mapped with MAP_PERSISTENT_BIT set in the MapBufferRange access flags. An INVALID_OPERATION error is generated if the BUFFER_IMMUTABLE_STORAGE flag of the buffer object is TRUE and the value of BUFFER_STORAGE_FLAGS for the buffer does not have the DYNAMIC_STORAGE_BIT set.

6.2.1

Clearing Buffer Object Data Stores

To fill all or part of a buffer object’s data store with constant values, use the commands void ClearBufferSubData( enum target, enum internalformat, intptr offset, sizeiptr size, enum format, enum type, const void *data ); void ClearNamedBufferSubData( uint buffer, enum internalformat, intptr offset, sizeiptr size, enum format, enum type, const void *data ); For ClearBufferSubData, the buffer object is that bound to target, which must be one of the values listed in table 6.1. For ClearNamedBufferSubData, buffer is the name of the buffer object. internalformat must be set to one of the format tokens listed in table 8.16. format and type specify the format and type of the source data and are interpreted as described in section 8.4.4. Unlike the related ClearTexImageSubData command described in section 8.21, all defined conversions between the clear value’s format and type, and the specified internalformat are allowed, including conversions between floating-point and integer components. offset is the offset, measured in basic machine units, into the buffer object’s data store from which to begin filling, and size is the size, also in basic machine units, of the range to fill. data is a pointer to an array of between one and four components containing the data to be used as the source of the constant fill value. The elements of data are converted by the GL into the format specified by internalformat in the manner described in section 2.2.1, and then used to fill the specified range of the destination OpenGL 4.6 (Core Profile) - May 5, 2022

6.2. CREATING AND MODIFYING BUFFER OBJECT DATA STORES

buffer. If data is NULL, then the pointer is ignored and the sub-range of the buffer is filled with zeros. Errors An INVALID_ENUM error is generated by ClearBufferSubData if target is not one of the targets listed in table 6.1. An INVALID_VALUE error is generated by ClearBufferSubData if zero is bound to target. An INVALID_OPERATION error is generated by ClearNamedBufferData if buffer is not the name of an existing buffer object. An INVALID_ENUM error is generated if internalformat is not one of the format tokens listed in table 8.16. An INVALID_VALUE error is generated if offset or size are not multiples of the number of basic machine units for the internal format specified by internalformat. This value may be computed by multiplying the number of components for internalformat from table 8.16 by the size of the base type from that table. 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 object. An INVALID_OPERATION error is generated if any part of the specified buffer range is mapped with MapBufferRange or MapBuffer (see section 6.3), unless it was mapped with MAP_PERSISTENT_BIT set in the MapBufferRange access flags. An INVALID_VALUE error is generated if type is not one of the types in table 8.2. An INVALID_VALUE error is generated if format is not one of the formats in table 8.3. The commands void ClearBufferData( enum target, enum internalformat, enum format, enum type, const void *data ); void ClearNamedBufferData( uint buffer, enum internalformat, enum format, enum type, const void *data ); are respectively equivalent to ClearBufferSubData(target, internalf ormat, 0, size, f ormat, type, data); and OpenGL 4.6 (Core Profile) - May 5, 2022

6.3. MAPPING AND UNMAPPING BUFFER DATA

ClearNamedBufferSubData(buf f er, internalf ormat, 0, size, f ormat, type, data); where size is the value of BUFFER_SIZE for the destination buffer object.

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 with the commands void *MapBufferRange( enum target, intptr offset, sizeiptr length, bitfield acesss ); void *MapNamedBufferRange( uint buffer, intptr offset, sizeiptr length, bitfield access ); For MapBufferRange, the buffer object is that bound to target, which must be one of the values listed in table 6.1. For MapNamedBufferRange, buffer is the name of the buffer object. 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. • MAP_PERSISTENT_BIT indicates that it is not an error for the GL to read data from or write data to the buffer while it is mapped (see section 6.3.2). If this bit is set, the value of BUFFER_STORAGE_FLAGS for the buffer being mapped must include MAP_PERSISTENT_BIT.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.3. MAPPING AND UNMAPPING BUFFER DATA

• MAP_COHERENT_BIT indicates that the mapping should be performed coherently. That is, such a mapping follows the rules set forth in section 6.2. If this bit is set, the value of BUFFER_STORAGE_FLAGS for the buffer being mapped must include MAP_COHERENT_BIT. If no error occurs, the pointer values returned by Map*BufferRange must reflect an allocation aligned to the value of MIN_MAP_BUFFER_ALIGNMENT basic machine units. Subtracting offset basic machine units from the returned pointer will always produce a multiple of the value of MIN_MAP_BUFFER_ALIGNMENT. The returned pointer values 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. • 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

OpenGL 4.6 (Core Profile) - May 5, 2022

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

Value Depends on access1 access TRUE

pointer to the data store offset length

Table 6.4: Buffer object state set by MapBufferRange and MapNamedBufferRange. 1 BUFFER_ACCESS is set to READ_ONLY, WRITE_ONLY, or READ_WRITE if access & (MAP_READ_BIT|MAP_WRITE_BIT) is respectively MAP_READ_BIT, MAP_WRITE_BIT, or MAP_READ_BIT|MAP_WRITE_BIT.

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 Map*BufferRange. 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. A successful Map*BufferRange sets buffer object state values as shown in table 6.4. Errors If an error occurs, a NULL pointer is returned. An INVALID_ENUM error is generated by MapBufferRange if target is not one of the valid targets listed above. An INVALID_OPERATION error is generated by MapBufferRange if zero is bound to target. An INVALID_OPERATION error is generated by MapNamedBufferRange if buffer is not the name of an existing buffer object.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.3. MAPPING AND UNMAPPING BUFFER DATA

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. • Any of MAP_READ_BIT, MAP_WRITE_BIT, MAP_PERSISTENT_BIT, or MAP_COHERENT_BIT are set, but the same bit is not set in the buffer’s storage flags. 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. The entire data store of a buffer object can be mapped into the client’s address space with the commands void *MapBuffer( enum target, enum access ); void *MapNamedBuffer( uint buffer, enum access ); These commands are respectively equivalent to MapBufferRange(target, 0, length, flags); and MapNamedBufferRange(buf f er, 0, length, flags); where length is equal to the value of BUFFER_SIZE for the target buffer and flags is equal to • MAP_READ_BIT, if access is READ_ONLY OpenGL 4.6 (Core Profile) - May 5, 2022

6.3. MAPPING AND UNMAPPING BUFFER DATA

• MAP_WRITE_BIT, if access is WRITE_ONLY • MAP_READ_BIT | MAP_WRITE_BIT, if access is READ_WRITE. The pointer value returned by MapBuffer and MapNamedBuffer must be aligned to the value of MIN_MAP_BUFFER_ALIGNMENT basic machine units. Errors An INVALID_ENUM error is generated if access is not READ_ONLY, WRITE_ONLY, or READ_WRITE. Other errors are generated as described above for MapBufferRange and MapNamedBufferRange. If a buffer object is mapped with the MAP_FLUSH_EXPLICIT_BIT flag, modifications to the mapped range may be indicated with the commands void FlushMappedBufferRange( enum target, intptr offset, sizeiptr length ); void FlushMappedNamedBufferRange( uint buffer, intptr offset, sizeiptr length ); For FlushMappedBufferRange, the buffer object is that bound to target, which must be one of the targets listed in table 6.1. For FlushMappedNamedBufferRange, buffer is the name of the buffer object. 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 the buffer object. FlushMapped*BufferRange may be called multiple times to indicate distinct subranges of the mapping which require flushing. If a buffer range is mapped with both MAP_PERSISTENT_BIT and MAP_FLUSH_EXPLICIT_BIT set, then FlushMapped*BufferRange may be called to ensure that data written by the client into the flushed region becomes visible to the server. Data written to a coherent store will always become visible to the server after an unspecified period of time. Errors An INVALID_ENUM error is generated by FlushMappedBufferRange if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by FlushMappedBufferRange if zero is bound to target.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.3. MAPPING AND UNMAPPING BUFFER DATA

An INVALID_OPERATION error is generated by FlushMappedNamedBufferRange if buffer is not the name of an existing buffer object. An INVALID_OPERATION error is generated if the buffer object 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 range of a buffer object, and before the data in that range are dereferenced by any GL commands, the mapping must be relinquished with one of the commands boolean UnmapBuffer( enum target ); boolean UnmapNamedBuffer( uint buffer ); For UnmapBuffer, the buffer object is that bound to target, which must be one of the targets listed in table 6.1. For UnmapNamedBuffer, buffer is the name of the buffer object. 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. Unmap*Buffer returns TRUE unless data values in the buffer object’s data store have become corrupted during the period that the buffer object was mapped. Such corruption can be the result of a screen resolution change or other window systemdependent 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 object’s data store is mapped. If such corruption has occurred, Unmap*Buffer return FALSE, and the contents of the 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. If an error is generated, FALSE is returned. Errors An INVALID_ENUM error is generated by UnmapBuffer if target is not

OpenGL 4.6 (Core Profile) - May 5, 2022

6.4. EFFECTS OF ACCESSING OUTSIDE BUFFER BOUNDS one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by UnmapBuffer if zero is bound to target. An INVALID_OPERATION error is generated by UnmapNamedBuffer if buffer is not the name of an existing buffer object. An INVALID_OPERATION error is generated if the buffer object’s data store is already in the unmapped state.

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, unless it was allocated by a call to *BufferStorage with the MAP_PERSISTENT_BIT included in flags. However, only commands which explicitly describe this error are required to do so. If an error is not generated, such commands 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. Robust buffer access can be enabled by creating a context with robust access enabled through the window system binding APIs. When enabled, any command unable to generate a GL error as described above, such as buffer object accesses from the active program, will not read or modify memory outside of the data store of the buffer object and will not result in GL interruption or termination. Outof-bounds reads may return values from within the buffer object or zero values. Out-of-bounds writes may modify values within the buffer object or be discarded. Accesses made through resources attached to binding points are only protected within the buffer object from which the binding point is declared. For example, for an out-of-bounds access to a member variable of a uniform block, the access protection is provided within the uniform buffer object, and not for the bound buffer range for this uniform block.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.5. INVALIDATING BUFFER DATA

6.5

Invalidating Buffer Data

All or part of the data store of a buffer object may be invalidated by calling void InvalidateBufferSubData( uint buffer, intptr offset, sizeiptr length ); with buffer set to the name of the buffer whose data store is being invalidated. offset and length specify the range of the data in the buffer object that is to be invalidated. Data in the specified range have undefined values after calling InvalidateBufferSubData. Errors An INVALID_VALUE error is generated if buffer is zero or is not the name of an existing buffer object. An INVALID_VALUE error is generated if offset or length is negative, or if offset + length is greater than the value of BUFFER_SIZE for buffer. An INVALID_OPERATION error is generated if buffer is currently mapped by MapBuffer or if the invalidate range intersects the range currently mapped by MapBufferRange, unless it was mapped with MAP_PERSISTENT_BIT set in the MapBufferRange access flags. The command void InvalidateBufferData( uint buffer ); is equivalent to calling InvalidateBufferSubData with offset equal to zero and length equal to the value of BUFFER_SIZE for buffer.

6.6

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 with the commands void CopyBufferSubData( enum readTarget, enum writeTarget, intptr readOffset, intptr writeOffset, sizeiptr size ); void CopyNamedBufferSubData( uint readBuffer, uint writeBuffer, intptr readOffset, intptr writeOffset, sizeiptr size );

OpenGL 4.6 (Core Profile) - May 5, 2022

6.7. BUFFER OBJECT QUERIES

For CopyBufferSubData, readTarget and writeTarget are the targets to which the source and destination buffers are bound, and each must be one of the targets listed in table 6.1. For CopyNamedBufferSubData, readBuffer and writeBuffer are the names of the source and destination buffers, respectively. 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 destination buffer object that is to be replaced, in terms of basic machine units. readOffset and size specify the range of data in the source buffer object that is to be copied to the corresponding region of writeTarget. Errors An INVALID_OPERATION error is generated by CopyBufferSubData if zero is bound to readTarget or writeTarget. An INVALID_ENUM error is generated by CopyBufferSubData if readTarget or writeTarget is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by CopyNamedBufferSubData if readBuffer or writeBuffer is not the name of an existing buffer object. An INVALID_VALUE error is generated if any of readOffset, writeOffset, or size are negative, if readOffset + size exceeds the size of the source buffer object, or if writeOffset +size exceeds the size of the destination buffer object. An INVALID_VALUE error is generated if the source and destination are the same buffer object, and the ranges [readOffset, readOffset + size) and [writeOffset, writeOffset + size) overlap. An INVALID_OPERATION error is generated if either the source or destination buffer objects is mapped, unless they were mapped with MAP_PERSISTENT_BIT set in the Map*BufferRange access flags.

6.7

Buffer Object Queries

To query information about a buffer object, use the commands void GetBufferParameteriv( enum target, enum pname, int *data ); void GetBufferParameteri64v( enum target, enum pname, int64 *data ); void GetNamedBufferParameteriv( uint buffer, enum pname, int *data ); OpenGL 4.6 (Core Profile) - May 5, 2022

6.7. BUFFER OBJECT QUERIES

void GetNamedBufferParameteri64v( uint buffer, enum pname, int64 *data ); For GetBufferParameter*, the buffer object is that bound to target, which must be one of the targets listed in table 6.1. For GetNamedBufferParameter*, buffer is the name of the buffer object. 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 by GetBufferParameter* if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by GetBufferParameter* if zero is bound to target. An INVALID_OPERATION error is generated by GetNamedBufferParameter* if buffer is not the name of an existing buffer object. An INVALID_ENUM error is generated if pname is not one of the buffer object parameters other than BUFFER_MAP_POINTER. To query the data store of a buffer object, use the commands void GetBufferSubData( enum target, intptr offset, sizeiptr size, void *data ); void GetNamedBufferSubData( uint buffer, intptr offset, sizeiptr size, void *data ); For GetBufferSubData, target specifies the target to which the source buffer object is bound, and must be one of the values listed in table 6.1. For GetNamedBufferSubData, buffer specifies the name of the source buffer object. offset and size indicate the range of data in the source buffer object that is to be queried, in terms of basic machine units. data specifies a region of client memory, size basic machine units in length, into which the data is to be retrieved. Errors An INVALID_ENUM error is generated by GetBufferSubData if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by GetBufferSubData if zero is bound to target.

OpenGL 4.6 (Core Profile) - May 5, 2022

6.7. BUFFER OBJECT QUERIES

An INVALID_OPERATION error is generated by GetNamedBufferSubData if buffer is not the name of an existing buffer object. 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 source buffer object. An INVALID_OPERATION error is generated if the source buffer object is currently mapped, unless it was mapped with MAP_PERSISTENT_BIT set in the Map*BufferRange access flags. While part or all of the data store of a buffer object is mapped, the pointer to the mapped range of the data store may be queried with the commands void GetBufferPointerv( enum target, enum pname, const void **params ); void GetNamedBufferPointerv( uint buffer, enum pname, const void **params ); For GetBufferPointerv, the buffer object is that bound to target, which must be one of the targets listed in table 6.1. For GetNamedBufferPointerv, buffer is the name of the buffer object. pname must be BUFFER_MAP_POINTER. The single buffer map pointer is returned in params. A NULL pointer value is returned if the buffer object’s data store is not currently mapped; or if the requesting context did not map the buffer object’s data store, and the implementation is unable to support mappings on multiple clients. Errors An INVALID_ENUM error is generated by GetBufferPointerv if target is not one of the targets listed in table 6.1. An INVALID_OPERATION error is generated by GetBufferPointerv if zero is bound to target. An INVALID_OPERATION error is generated by GetNamedBufferPointerv if buffer is not the name of an existing buffer object. An INVALID_ENUM error is generated if pname is not BUFFER_MAP_POINTER.

6.7.1

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 OpenGL 4.6 (Core Profile) - May 5, 2022

6.8. BUFFER OBJECT STATE

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

6.8

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 23.46, 23.47, 23.48, and 23.49 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. 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 4.6 (Core Profile) - May 5, 2022

6.8. BUFFER OBJECT STATE

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

MAX_SHADER_STORAGE_BUFFER_-

BINDINGS

Transform feedback array bindings (see sec. 13.3.2) 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_BUFFERS 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 4.6 (Core Profile) - May 5, 2022

6.8. BUFFER OBJECT STATE

The state of each buffer object consists of a buffer size in basic machine units, a usage parameter, an access parameter, a boolean indicating whether or not buffer storage is immutable, an unsigned integer storing the flags with which it was allocated, 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 4.6 (Core Profile) - 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 15. A shader specifies operations that are meant to occur on data as it moves through different programmable stages of the OpenGL 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 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 can be loaded into a shader object. A SPIR-V module can also be associated with a shader and then specialized. 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

7.1. SHADER OBJECTS

unified program object, or the current program object may be set for each stage 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, and are described in 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 15). 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 19), 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. Subroutine uniform variables (see section 7.10) are similar to uniforms but are context state, rather than program object state. Samplers (see section 7.11) are a special form of uniform used for texturing (see chapter 8). Images (see section 7.12) are a special form of uniform identifying a level of a texture to be accessed using built-in shader functions as described in section 8.26. 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 );

OpenGL 4.6 (Core Profile) - May 5, 2022

7.1. SHADER OBJECTS type VERTEX_SHADER TESS_CONTROL_SHADER TESS_EVALUATION_SHADER GEOMETRY_SHADER FRAGMENT_SHADER COMPUTE_SHADER

89 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 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 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 Shading Language Specification. If shader was previously associated with a SPIR-V module (via the ShaderBinary command), that association is broken. Upon successful completion of this command the SPIR_V_BINARY state of shader is set to FALSE. OpenGL 4.6 (Core Profile) - May 5, 2022

7.1. SHADER OBJECTS

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 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 may be queried with GetShaderiv (see section 7.14). 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 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 may be queried with GetShaderInfoLog to obtain more information about the compilation attempt (see section 7.14). 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_OPERATION error is generated if the SPIR_V_BINARY state of shader is TRUE. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.2. SHADER BINARIES

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 7.14). Shader objects can be deleted with the command 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 may be queried with GetShaderiv (see section 7.14). 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.2. SHADER BINARIES

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. GL defines an execution environment for shaders created from SPIR-V modules. To load a SPIR-V binary into GL, set binaryformat to SHADER_BINARY_FORMAT_SPIR_V. binary should point to the start of a valid SPIR-V module binary and length should contain the length of that binary, in bytes. Upon successful consumption of the SPIR-V module: • each entry of shaders will be associated with that SPIR-V module, • the SPIR_V_BINARY state of each shader is set to TRUE, • the COMPILE_STATUS of each of these shaders is set to FALSE, • any existing source string (specified by ShaderSource) is removed, and • any information about a previous compile is lost. Shaders associated with SPIR-V modules must be finalized by calling SpecializeShader, as described in section 7.2.1. GL also provides a mechanism to obtain token values for such formats provided by extensions. The number of 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. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.2. SHADER BINARIES

An INVALID_OPERATION error is generated if binaryformat is not SHADER_BINARY_FORMAT_SPIR_V and 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 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.

7.2.1

Shader Specialization

Shaders associated with SPIR-V modules must be specialized before they can be linked into a program object. It is not necessary to specialize the shader before it is attached to a program object. Once specialized, a shader may not be specialized again without first re-associating the original SPIR-V module with it, through ShaderBinary. Specialization does two things: • Selects the name of the entry point, for that shader’s stage, from the SPIR-V module. • Sets the values of all, or a subset of, the specialization constants in the SPIRV module. To specialize a shader created from a SPIR-V module, call: void SpecializeShader( uint shader, const char *pEntryPoint, uint numSpecializationConstants, const uint *pConstantIndex, const uint *pConstantValue ); shader is the name of a shader object containing unspecialized SPIR-V as created from a successful call to ShaderBinary to which a SPIR-V module was passed. pEntryPoint is a pointer to a null-terminated UTF-8 string specifying the name of the entry point in the SPIR-V module to use for this shader. numSpecializationConstants is the number of specialization constants whose values to set in this call. pConstantIndex is a pointer to an array of numSpecializationConstants OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS unsigned integers, each holding the index of a specialization constant in the SPIRV module whose value to set. The corresponding entry in pConstantValue is used to set the value of the specialization constant indexed by the entry in pConstantIndex. Although this array is of unsigned integer, each entry is bitcast to the appropriate type for the module, and therefore, floating-point constants may be set by including their IEEE-754 bit representation in the pConstantValue array. Specialization constants not referenced by pConstantIndex retain their default values as specified in the SPIR-V module. On successful shader specialization, the compile status for shader is set to TRUE. On failure, the compile status for shader is set to FALSE and additional information about the cause of the failure may be available in the shader compilation log. Specialization can fail if the SPIR-V module fails to meet the requirements listed in appendix C. 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_OPERATION error is generated if the value of SPIR_V_BINARY for shader is not TRUE, or if the shader has already been specialized. An INVALID_VALUE error is generated if pEntryPoint does not match the Name of any OpEntryPoint declaration in the SPIR-V module associated with shader. An INVALID_OPERATION error is generated if the ExecutionMode of the OpEntryPoint indicated by pEntryPoint does not match the type of shader. An INVALID_VALUE error is generated if any element of pConstantIndex refers to a specialization constant that does not exist in the shader module contained in shader.

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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

To attach a shader object to a program object, use the command void AttachShader( uint program, uint shader ); 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 or specialized. Multiple shader objects of the same type may be attached to a single program object, and 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. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

program. In order to use the shader objects contained in a program object, the program object must be linked. The command 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 may be queried with GetProgramiv (see section 7.14). This status will be set to TRUE if a valid executable is created, and FALSE otherwise. It is possible to link successfully compiled or specialized shader objects built from any version of the OpenGL Shading Language or OpenGL ES Shading Language supported by the implementation, and not all shader objects linked together have to use the same version of the shading language. Linking can fail for a variety of reasons as specified in the shading language specification for source shaders, or if the requirements in appendix C are not met for SPIR-V shaders, 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 or specialized successfully. • More active uniform or active sampler variables are used in program than allowed (see sections 7.6, 7.11, and 11.3.3). • program contains objects to form a tessellation control shader (see section 11.2.1), and – the program is not separable and contains no objects to form a vertex shader; – the output patch vertex count is not specified in any compiled tessellation control shader object; or – the output patch vertex count is specified differently in multiple tessellation control shader objects. • program contains objects to form a tessellation evaluation shader (see section 11.2.3), and – the program is not separable and contains no objects to form a vertex shader; OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

– the tessellation primitive mode is not specified in any compiled tessellation evaluation shader object; or – the tessellation primitive mode, spacing, vertex order, or point mode is specified differently in multiple tessellation evaluation shader objects. • program contains objects to form a geometry shader (see section 11.3), and – the program is not separable and contains no objects to form a vertex shader; – the input primitive type, output primitive type, or maximum output vertex count is not specified in any compiled geometry shader object; or – the input primitive type, output primitive type, or maximum output vertex count is specified differently in multiple geometry shader objects. • program contains objects to form a compute shader (see section 19) and, – program also contains objects to form any other type of shader. • All the shader objects attached to program do not have the same value for the SPIR_V_BINARY state. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

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. Each program object has an information log that is overwritten as a result of a link operation. This information log may be queried with GetProgramInfoLog to obtain more information about the link operation or the validation information (see section 7.14). 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/or fragment processing will 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 dispatches. Errors An INVALID_VALUE error is generated if program is neither zero nor the

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

100

State set with this command does not take effect until after the next time LinkProgram or ProgramBinary is called successfully. 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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

101

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: 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

102

commands.

7.3.1

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, interface blocks, and subroutines 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

103

• PROGRAM_INPUT corresponds to the set of active input variables used by the first shader stage of program. If program includes multiple shader stages, 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. • VERTEX_SUBROUTINE,

TESS_CONTROL_SUBROUTINE, TESS_EVALUATION_SUBROUTINE, GEOMETRY_SUBROUTINE, FRAGMENT_SUBROUTINE, and COMPUTE_SUBROUTINE correspond to the set of active

subroutines for the vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute shader stages of program, respectively (see section 7.10). • VERTEX_SUBROUTINE_UNIFORM, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, and COMPUTE_SUBROUTINE_UNIFORM correspond to the set of active subroutine uniform variables used by the vertex, tessellation control, tessellation evaluation, geometry, fragment, and compute shader stages of program, respectively (see section 7.10). • 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 13.3.3). 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. • TRANSFORM_FEEDBACK_BUFFER corresponds to the set of active buffer binding points to which output variables in the TRANSFORM_FEEDBACK_VARYING interface are written.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

104

• 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, interface block, or subroutine 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]". • 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:

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

105

• 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: • 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. For interfaces enumerating active subroutines, a single entry will be generated for each active subroutine, using the subroutine name from the shader source. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

106

For shaders constructed from SPIR-V binaries (that is with a state of SPIR_V_BINARY equal to TRUE), variables may not have names associated with them, as the OpName and OpMemberName debug instructions are optional and may not be present in a SPIR-V module. When the Op*Name instructions are present, it is implementation-dependent if these are reported via the name reflection APIs. If no name reflection information is available, the name string associated with each active variable is the empty string (""). In this case, any queries that operate with a name as input will return INVALID_INDEX or -1 as appropriate, and any queries that return information about the name of a resource will report a name length of one (for the null character) and return an empty string with a length of zero. The order of the active resource list is implementation-dependent for all interfaces except for TRANSFORM_FEEDBACK_VARYING. If variables in the TRANSFORM_FEEDBACK_VARYING interface were specified using the TransformFeedbackVaryings command, the active resource list will be arranged in the variable order specified in the most recent call to TransformFeedbackVaryings before the last call to LinkProgram. If variables in the TRANSFORM_FEEDBACK_VARYING interface were specified using layout qualifiers in shader code, the order of the active resource list is implementation-dependent. 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. For PROGRAM_INPUT and PROGRAM_OUTPUT interfaces for shaders that recieve or produce patch primitves, the active resource list will include both pervertex and per-patch inputs and outputs. For the TRANSFORM_FEEDBACK_BUFFER interface, the list of active buffer binding points is built by identifying each unique binding point to which one or more active output variables will be written in transform feedback mode. Active transform feedback buffers do not have an associated name string. For the TRANSFORM_FEEDBACK_VARYING interface, the active resource list will include entries for the special variable names gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, and gl_SkipComponents4 (see section 11.1.2.1). These variables are used to control how output values are written to transform feedback buffers. When enumerating the properties of such resources, these variables are considered to have a TYPE of NONE and an ARRAY_SIZE of 0 (gl_NextBuffer), 1, 2, 3, and 4, respectively. When a program is linked successfully, active variables in the UNIFORM, OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

107

PROGRAM_INPUT, PROGRAM_OUTPUT, or any of the subroutine uniform interfaces,

are assigned one or more signed integer locations. These locations can be used by commands to assign values to uniforms and subroutine uniforms, to identify generic vertex attributes associated with vertex shader inputs, or to identify fragment color output numbers and indices 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: • 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

108

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. If pname is MAX_NUM_COMPATIBLE_SUBROUTINES, the value returned is the number of compatible subroutines for the active subroutine uniform in programInterface with the most compatible subroutines. 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, MAX_NUM_ACTIVE_VARIABLES, or MAX_NUM_COMPATIBLE_SUBROUTINES. An INVALID_OPERATION error is generated if pname is MAX_NAME_LENGTH and programInterface is ATOMIC_COUNTER_BUFFER or TRANSFORM_FEEDBACK_BUFFER, since active atomic counter and transform feedback buffer 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, TRANSFORM_FEEDBACK_BUFFER, or UNIFORM_BLOCK. An INVALID_OPERATION error is generated if pname is MAX_NUM_COMPATIBLE_SUBROUTINES and programInterface is not VERTEX_SUBROUTINE_UNIFORM, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, or COMPUTE_SUBROUTINE_UNIFORM.

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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

109

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, other than the special names gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, and gl_SkipComponents4 (see section 11.1.2.1). Otherwise, INVALID_INDEX is returned. • 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. 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 or TRANSFORM_FEEDBACK_BUFFER, since active atomic counter and transform feedback buffer resources are not assigned name strings. If name does not match a resource as described above, the value INVALID_INDEX is returned, but no GL error is generated. The command void GetProgramResourceName( uint program, enum programInterface, uint index, sizei bufSize, sizei *length, char *name ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

110

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, may 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 or TRANSFORM_FEEDBACK_BUFFER, since active atomic counter and transform feedback buffer 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 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

111

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 NUM_COMPATIBLE_SUBROUTINES, COMPATIBLE_SUBROUTINES

IS_PER_PATCH

ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BLOCK, TRANSFORM_FEEDBACK_BUFFER, UNIFORM_BLOCK BUFFER_VARIABLE, COMPUTE_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, PROGRAM_INPUT, PROGRAM_OUTPUT, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, TRANSFORM_FEEDBACK_VARYING, UNIFORM, VERTEX_SUBROUTINE_UNIFORM BUFFER_VARIABLE, UNIFORM UNIFORM ATOMIC_COUNTER_BUFFER, SHADER_STORAGE_BLOCK, UNIFORM_BLOCK COMPUTE_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, VERTEX_SUBROUTINE_UNIFORM PROGRAM_INPUT, PROGRAM_OUTPUT

GetProgramResourceiv properties continued on next page

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

112

GetProgramResourceiv properties continued from previous page Property Supported Interfaces LOCATION

LOCATION_COMPONENT LOCATION_INDEX 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 TRANSFORM_FEEDBACK_BUFFER_INDEX TRANSFORM_FEEDBACK_BUFFER_STRIDE TOP_LEVEL_ARRAY_SIZE, TOP_LEVEL_ARRAY_STRIDE TYPE

COMPUTE_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, PROGRAM_INPUT, PROGRAM_OUTPUT, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, UNIFORM, VERTEX_SUBROUTINE_UNIFORM PROGRAM_INPUT, PROGRAM_OUTPUT PROGRAM_OUTPUT all but ATOMIC_COUNTER_BUFFER and TRANSFORM_FEEDBACK_BUFFER BUFFER_VARIABLE, TRANSFORM_FEEDBACK_VARYING, UNIFORM ATOMIC_COUNTER_BUFFER, BUFFER_VARIABLE, PROGRAM_INPUT, PROGRAM_OUTPUT, SHADER_STORAGE_BLOCK, UNIFORM, UNIFORM_BLOCK

TRANSFORM_FEEDBACK_VARYING TRANSFORM_FEEDBACK_BUFFER BUFFER_VARIABLE 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, shader storage block, OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

113

or transform feedback buffer 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 the member of an interface block, the value -1 is written to params. 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, shader storage block, or transform feedback buffer 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, shader storage block, or atomic counter buffer. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

114

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, output, or subroutine uniform 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 LOCATION_COMPONENT, a single integer indicating the first component of the location assigned to an active input or output variable is written to params. For input and output variables with a component specified by a layout qualifier, the specified component is written. For all other input and output variables, the value zero is written. For the property LOCATION_INDEX, a single integer identifying the fragment color index of an active fragment shader output variable is written to params. If the active variable is not an output for a fragment shader, the value -1 will be 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, interface block, or subroutine 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, shader storage block, or transform feedback buffer is written to params. 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 variables in the TRANSFORM_FEEDBACK_VARYING interface, the value written is the offset in the transform feedback buffer storage assigned to each verOpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

115

tex captured in transform feedback mode where the value of the variable will be stored. Such offsets are specified via the xfb_offset layout qualifier or assigned according to the variables position in the list of strings passed to TransformFeedbackVaryings. Offsets are expressed in basic machine units. For all variables not recorded in transform feedback mode, including the special names gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, and gl_SkipComponents4, -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 the active variable is written to params. If the top-level block member is not declared as an array f 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 TRANSFORM_FEEDBACK_BUFFER_INDEX, a single integer identifying the index of the active transform feedback buffer associated with an active variable is written to params. For variables corresponding to the special names gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, and gl_SkipComponents4, -1 is written to params. For the property TRANSFORM_FEEDBACK_BUFFER_STRIDE, a single integer identifying the stride, in basic machine units, between consecutive vertices written to the transform feedback buffer 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

Type Name Token FLOAT FLOAT_VEC2 FLOAT_VEC3 FLOAT_VEC4 DOUBLE DOUBLE_VEC2 DOUBLE_VEC3 DOUBLE_VEC4 INT INT_VEC2 INT_VEC3 INT_VEC4 UNSIGNED_INT UNSIGNED_INT_VEC2 UNSIGNED_INT_VEC3 UNSIGNED_INT_VEC4 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 DOUBLE_MAT2 DOUBLE_MAT3 DOUBLE_MAT4 DOUBLE_MAT2x3 DOUBLE_MAT2x4 DOUBLE_MAT3x2

116 Keyword

Attrib Xfb

Buffer

float vec2 vec3 vec4 double dvec2 dvec3 dvec4 int ivec2 ivec3 ivec4 uint uvec2 uvec3 uvec4 bool bvec2 bvec3 bvec4 mat2 mat3 mat4 mat2x3 mat2x4 mat3x2 mat3x4 mat4x2 mat4x3 dmat2 dmat3 dmat4 dmat2x3 dmat2x4 dmat3x2

• • • • • • • • • • • • • • • •

• • • • • • • • • • • • • • •

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

(Continued on next page)

OpenGL 4.6 (Core Profile) - May 5, 2022

7.3. PROGRAM OBJECTS

117

OpenGL Shading Language Type Tokens (continued) Type Name Token Keyword Attrib DOUBLE_MAT3x4 dmat3x4 • DOUBLE_MAT4x2 dmat4x2 • DOUBLE_MAT4x3 dmat4x3 • SAMPLER_1D SAMPLER_2D SAMPLER_3D SAMPLER_CUBE SAMPLER_1D_SHADOW SAMPLER_2D_SHADOW SAMPLER_1D_ARRAY SAMPLER_2D_ARRAY SAMPLER_CUBE_MAP_ARRAY SAMPLER_1D_ARRAY_SHADOW SAMPLER_2D_ARRAY_SHADOW SAMPLER_2D_MULTISAMPLE SAMPLER_2D_MULTISAMPLE_ARRAY SAMPLER_CUBE_SHADOW SAMPLER_CUBE_MAP_ARRAY_SHADOW SAMPLER_BUFFER SAMPLER_2D_RECT SAMPLER_2D_RECT_SHADOW INT_SAMPLER_1D INT_SAMPLER_2D INT_SAMPLER_3D INT_SAMPLER_CUBE INT_SAMPLER_1D_ARRAY INT_SAMPLER_2D_ARRAY INT_SAMPLER_CUBE_MAP_ARRAY INT_SAMPLER_2D_MULTISAMPLE INT_SAMPLER_2D_MULTISAMPLE_ARRAY

sampler1D sampler2D sampler3D samplerCube sampler1DShadow sampler2DShadow sampler1DArray sampler2DArray samplerCubeArray sampler1DArrayShadow sampler2DArrayShadow sampler2DMS sampler2DMSArray samplerCubeShadow samplerCubeArrayShadow samplerBuffer sampler2DRect sampler2DRectShadow isampler1D isampler2D isampler3D isamplerCube isampler1DArray isampler2DArray isamplerCubeArray isampler2DMS isampler2DMSArray

(Continued on next page)

OpenGL 4.6 (Core Profile) - May 5, 2022

Xfb • • •

Buffer • • •

7.3. PROGRAM OBJECTS

118

OpenGL Shading Language Type Tokens (continued) Type Name Token Keyword Attrib Xfb INT_SAMPLER_BUFFER INT_SAMPLER_2D_RECT UNSIGNED_INT_SAMPLER_1D UNSIGNED_INT_SAMPLER_2D UNSIGNED_INT_SAMPLER_3D UNSIGNED_INT_SAMPLER_CUBE UNSIGNED_INT_SAMPLER_1D_ARRAY 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 UNSIGNED_INT_SAMPLER_2D_RECT IMAGE_1D IMAGE_2D IMAGE_3D IMAGE_2D_RECT IMAGE_CUBE IMAGE_BUFFER IMAGE_1D_ARRAY IMAGE_2D_ARRAY IMAGE_CUBE_MAP_ARRAY IMAGE_2D_MULTISAMPLE IMAGE_2D_MULTISAMPLE_ARRAY INT_IMAGE_1D INT_IMAGE_2D

isamplerBuffer isampler2DRect usampler1D usampler2D usampler3D usamplerCube usampler1DArray usampler2DArray usamplerCubeArray usampler2DMS usampler2DMSArray usamplerBuffer usampler2DRect image1D image2D image3D image2DRect imageCube imageBuffer image1DArray image2DArray imageCubeArray image2DMS image2DMSArray iimage1D iimage2D

(Continued on next page)

OpenGL 4.6 (Core Profile) - May 5, 2022

Buffer

7.3. PROGRAM OBJECTS

119

OpenGL Shading Language Type Tokens (continued) Type Name Token Keyword Attrib Xfb INT_IMAGE_3D INT_IMAGE_2D_RECT INT_IMAGE_CUBE INT_IMAGE_BUFFER INT_IMAGE_1D_ARRAY INT_IMAGE_2D_ARRAY INT_IMAGE_CUBE_MAP_ARRAY INT_IMAGE_2D_MULTISAMPLE INT_IMAGE_2D_MULTISAMPLE_ARRAY UNSIGNED_INT_IMAGE_1D UNSIGNED_INT_IMAGE_2D UNSIGNED_INT_IMAGE_3D UNSIGNED_INT_IMAGE_2D_RECT UNSIGNED_INT_IMAGE_CUBE UNSIGNED_INT_IMAGE_BUFFER UNSIGNED_INT_IMAGE_1D_ARRAY UNSIGNED_INT_IMAGE_2D_ARRAY UNSIGNED_INT_IMAGE_CUBE_MAP_ARRAY UNSIGNED_INT_IMAGE_2D_MULTISAMPLE UNSIGNED_INT_IMAGE_2D_MULTISAMPLE_ARRAY UNSIGNED_INT_ATOMIC_COUNTER

iimage3D iimage2DRect iimageCube iimageBuffer iimage1DArray iimage2DArray iimageCubeArray iimage2DMS iimage2DMSArray uimage1D uimage2D uimage3D uimage2DRect uimageCube uimageBuffer uimage1DArray uimage2DArray uimageCubeArray uimage2DMS uimage2DMSArray atomic_uint

Table 7.3: OpenGL 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). OpenGL 4.6 (Core Profile) - May 5, 2022

Buffer

7.3. PROGRAM OBJECTS

120

For the property COMPATIBLE_SUBROUTINES, an array of integers is written to params, with each integer specifying the index of an active subroutine that can be assigned to the selected subroutine uniform. The number of values written to params for an active subroutine is given by the value of the property NUM_COMPATIBLE_SUBROUTINES for the resource. 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. 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 commands int GetProgramResourceLocation( uint program, enum programInterface, const char *name ); int GetProgramResourceLocationIndex( uint program, enum programInterface, const char *name ); return the location or the fragment color index, respectively, assigned to the variable named name in interface programInterface of program object program. For GetProgramResourceLocation, programInterface must be one of UNIFORM, PROGRAM_INPUT, PROGRAM_OUTPUT, VERTEX_SUBROUTINE_UNIFORM, TESS_CONTROL_SUBROUTINE_UNIFORM, TESS_EVALUATION_SUBROUTINE_UNIFORM, GEOMETRY_SUBROUTINE_UNIFORM, FRAGMENT_SUBROUTINE_UNIFORM, or COMPUTE_SUBROUTINE_UNIFORM. For GetProgramResourceLocationIndex, programInterface must be PROGRAM_OUTPUT. The value -1 will be returned by either command if an error occurs, if name does not identify an active variable on programInterface, or if name identifies an active variable that does OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

121

not have a valid location assigned, as described above. The locations returned by these commands are the same locations returned when querying the LOCATION and LOCATION_INDEX resource properties. A string provided to GetProgramResourceLocation or GetProgramResourceLocationIndex 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 and GetProgramResourceLocationIndex return the location or fragment color index 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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

122

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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

123

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 23.31. 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. Program pipeline objects may also be created with the command void CreateProgramPipelines( sizei n, uint *pipelines ); CreateProgramPipelines returns n previously unused program pipeline names in pipelines, each representing a new program pipeline object which is a state vector comprising all the state and with the same initial values listed in table 23.31. Errors An INVALID_VALUE error is generated if n is negative. 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 void UseProgramStages( uint pipeline, bitfield stages, uint program ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS where pipeline is the program pipeline object to be updated, stages is the bitwise OR of accepted constants representing shader stages, and program identifies the program 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 VERTEX_SHADER_BIT, COMPUTE_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. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

124

7.4. PROGRAM PIPELINE OBJECTS

125

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. – 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

126

• 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 and component layout qualifiers and match in type and qualification. For the purposes of interface matching, variables declared with a location layout qualifier but without a component layout qualifier are considered to have declared a component layout qualifier of zero. 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 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. If either shader redeclares the built-in arrays gl_ClipDistance[] or gl_CullDistance[], the array must have the same size in both shaders. 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, no GL error is generated, but some or all of the inputs on the interface will be undefined. At an interface between program objects, the set of inputs and outputs are considered to match exactly if and only if:

OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

127

• 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. 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 results in all inputs being undefined except for cases noted below. Even if an input has a corresponding output that matches exactly, mismatches on other inputs or outputs may adversely affect the executable code generated to read or write the matching variable. The inputs and outputs on an interface between programs need not match exactly when input and output location qualifiers (sections 4.4.1(“Input Layout Qualifiers”) and 4.4.2(“Output Layout Qualifiers”) of the OpenGL Shading Language Specification) are used. When using location qualifiers, any input with an input location qualifier will be well-defined as long as the other program writes to a matching output, as described above. The names of variables need not match when matching by location. Additionally, scalar and vector inputs with location layout qualifiers will be well-defined if there is a corresponding output satisfying all of the following conditions: • the input and output match exactly in qualification, including in the location layout qualifier; • the output is a vector with the same basic component type and has more components than the input; and • the common component type of the input and output is int, uint, or float (scalars, vectors, and matrices with double component type are excluded). In this case, the components of the input will be taken from the first components of the matching output, and the extra components of the output will be ignored. To use any built-in input or output in the gl_PerVertex block in separable program objects, shader code must redeclare that block prior to use. A separable program will fail to link if: OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

128

• it contains multiple shaders of a single type with different redeclarations of this built-in block; or • any shader uses a built-in block member not found in the redeclaration of that block. • if both of the shaders are OpenGL ES shading language shaders, all matched input and output variables (in a block or otherwise) have identical precision qualification. • if one of the shaders is an OpenGL ES shading language shader and the other an OpenGL shading language shader, all matched input and output variables in the OpenGL ES shading language shader (in a block or otherwise) have highp precision qualification. There is one exception to this rule described below. 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. Built-in inputs or outputs not found in blocks do not affect interface matching. Any such built-in inputs are well-defined unless they are derived from built-in outputs not written by the previous shader stage.

7.4.2

SPIR-V Shader Interface Matching

SPIR-V shaders must also follow the rules in this section, whether they add to or override those given in section 7.4.1. Most importantly, SPIR-V variables and structure members do not have names and so no interface matching is done by name strings. All variables forming the input or output interfaces of shader stages must be listed as operands to the OpEntryPoint instruction and are declared with the Input or Output Storage Classes, respectively, in the SPIR-V module. Shader built-in variables meeting the following requirements define the built-in interface block. They must: • be explicitly declared (there are no implicit built-ins), • be decorated with the BuiltIn decoration,

OpenGL 4.6 (Core Profile) - May 5, 2022

7.4. PROGRAM PIPELINE OBJECTS

129

• be declared in a block whose top-level members are the built-ins, and • not have any Location or Component decorations. Built-ins only participate in interface matching if they are declared in such a block. There must be no more than one built-in interface block per shader per interface. User-defined interface variables must be decorated with a Location and can also be decorated with a Component. These correspond to the location and component discussed in section 7.4.1. Uniform and shader storage block variables must also be decorated with a Binding. A user-defined output variable is considered to match an input variable in the subsequent stage only if the two variables are declared with the same Location and Component decoration and match in type and decoration, except that interpolation decorations are not required to match. Variables or block members declared as structures are considered to match in type if and only if the structure members match in type, decoration, number, 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 size. At an interface between two non-fragment shader stages, the built-in interface block must match exactly, as described above. At an interface involving the fragment shader inputs, the presence or absence of any built-in output does not affect the interface matching. At an interface between two shader stages, the user-defined variable interface must match exactly. Additionally, scalar and vector inputs are well-defined if there is a corresponding output satisfying all of the following conditions: • the input and output match exactly in decoration, • the output is a vector with the same basic type and has at least as many components as the input, and • the common component type of the input and output is 32-bit integer or floating-point (64-bit component types are excluded). In this case, the components of the input will be taken from the first components of the output, and any extra components of the output will be ignored.

7.4.3

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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.5. PROGRAM BINARIES

130

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 7.14), initially empty. • An integer holding the length of the information log.

7.5

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 may 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.5. PROGRAM BINARIES

131

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 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. 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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.5. PROGRAM BINARIES

132

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 and fragment data location bindings as set by BindAttribLocation and BindFragDataLocation. OpenGL 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. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

7.6

133

Uniform Variables

Shaders can declare named uniform variables, as described in the OpenGL 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, samplers, and subroutine uniforms, can be grouped into uniform blocks. Named uniform blocks, as described in the OpenGL 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, except for subroutine uniforms, 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. 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. 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 subroutine uniforms and atomic counters, in the default uniform block accessed by a shader for a particular shader stage may 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, OpenGL 4.6 (Core Profile) - 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 15.1) Compute (see section 19.1)

134

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.

Shader Stage

pname for querying combined uniform block storage, in components

Vertex Tessellation control Tessellation evaluation Geometry Fragment Compute

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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

135

• 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 • 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 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 zero 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 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, 6.2.1, 6.3, 6.5, and 6.6. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

136

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 GetActiveUniformName( uint program, uint uniformIndex, sizei bufSize, sizei *length, char *uniformName ); is equivalent to GetProgramResourceName(program, UNIFORM, uniformIndex , bufSize, length, uniformName); The command void GetUniformIndices( uint program, sizei uniformCount, const char * const *uniformNames, uint *uniformIndices ); is equivalent (assuming no errors are generated) to:

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

137

for (int i = 0; i < uniformCount; i++) { uniformIndices[i] = GetProgramResourceIndex(program, UNIFORM, uniformNames[i]); } The command void GetActiveUniform( uint program, uint index, sizei bufSize, sizei *length, int *size, enum *type, char *name ); is equivalent (assuming no errors are generated) 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 (assuming no errors are generated) 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

138

pname

prop

UNIFORM_TYPE UNIFORM_SIZE UNIFORM_NAME_LENGTH UNIFORM_BLOCK_INDEX UNIFORM_OFFSET UNIFORM_ARRAY_STRIDE UNIFORM_MATRIX_STRIDE UNIFORM_IS_ROW_MAJOR UNIFORM_ATOMIC_COUNTER_BUFFER_INDEX

TYPE ARRAY_SIZE NAME_LENGTH BLOCK_INDEX OFFSET ARRAY_STRIDE MATRIX_STRIDE IS_ROW_MAJOR ATOMIC_COUNTER_BUFFER_INDEX

Table 7.6: GetProgramResourceiv properties used by GetActiveUniformsiv.

uint GetUniformBlockIndex( uint program, const char *uniformBlockName ); is equivalent to GetProgramResourceIndex(program, UNIFORM_BLOCK, uniformBlockName); 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); OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

139

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 UNIFORM_BLOCK_REFERENCED_BY_TESS_CONTROL_SHADER UNIFORM_BLOCK_REFERENCED_BY_TESS_EVALUATION_SHADER UNIFORM_BLOCK_REFERENCED_BY_GEOMETRY_SHADER UNIFORM_BLOCK_REFERENCED_BY_FRAGMENT_SHADER UNIFORM_BLOCK_REFERENCED_BY_COMPUTE_SHADER

BUFFER_BINDING BUFFER_DATA_SIZE NAME_LENGTH NUM_ACTIVE_VARIABLES ACTIVE_VARIABLES

Table 7.7: Blockiv.

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

GetProgramResourceiv properties used by GetActiveUniform-

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. Additionally, the command void GetActiveAtomicCounterBufferiv( uint program, uint bufferIndex, enum pname, int *params ); can be used to determine properties of active atomic counter buffer bindings used by program and is equivalent to GLenum prop; GetProgramResourceiv(program, ATOMIC_COUNTER_BUFFER, bufferIndex , 1, &prop, maxSize, NULL, params); OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

140

pname

prop

ATOMIC_COUNTER_BUFFER_BINDING ATOMIC_COUNTER_BUFFER_DATA_SIZE ATOMIC_COUNTER_BUFFER_ACTIVE_ATOMIC_COUNTERS ATOMIC_COUNTER_BUFFER_ACTIVE_ATOMIC_COUNTER_INDICES ATOMIC_COUNTER_BUFFER_REFERENCED_BY_VERTEX_SHADER ATOMIC_COUNTER_BUFFER_REFERENCED_BY_TESS_CONTROL_SHADER ATOMIC_COUNTER_BUFFER_REFERENCED_BY_TESS_EVALUATION_SHADER ATOMIC_COUNTER_BUFFER_REFERENCED_BY_GEOMETRY_SHADER ATOMIC_COUNTER_BUFFER_REFERENCED_BY_FRAGMENT_SHADER ATOMIC_COUNTER_BUFFER_REFERENCED_BY_COMPUTE_SHADER

BUFFER_BINDING BUFFER_DATA_SIZE NUM_ACTIVE_VARIABLES ACTIVE_VARIABLES 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

Table 7.8: GetProgramResourceiv properties used by GetActiveAtomicCounterBufferiv.

where the value of prop is taken from table 7.8, 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.

7.6.1

Loading Uniform Variables In The Default Uniform Block

To load values into the uniform variables except for subroutine uniforms and atomic counters, of the default uniform block of the active program object, use the commands void Uniform{1234}{ifd ui}( int location, T value ); void Uniform{1234}{ifd ui}v( int location, sizei count, const T *value ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

141

void UniformMatrix{234}{fd}v( int location, sizei count, boolean transpose, const float *value ); void UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}{fd}v( 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. 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*d{v} commands will load count sets of one to four doubleprecision floating-point values into a uniform defined as a double, a double 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 image, an integer, an integer vector, or an array of any of these types. Only the Uniform1i{v} commands can be used to load sampler and image values (see sections 7.11 and 7.12). 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 and UniformMatrix{234}dv commands will load count 2 × 2, 3 × 3, or 4 × 4 matrices (corresponding to 2, 3, or 4 in the command name) of single- or double-precision floating-point values, respectively, 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 and UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}dv 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 single- or double-precision floating-point values, respectively, 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 single-precision matrix consisting of two columns and four rows. If transpose is FALSE, the matrix is OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

142

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 dmat3, UniformMatrix3dv 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. 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_VALUE error is generated if Uniform1i{v} is used to set an image uniform to a value less than zero or greater than or equal to the value of MAX_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}

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

143

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 or image uniform is loaded with any of the Uniform* commands other than Uniform1i{v}. • 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}{ifd}( uint program, int location, T value ); void ProgramUniform{1234}{ifd}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 ); void ProgramUniformMatrix{234}{fd}v( uint program, int location, sizei count, boolean transpose, const T *value ); void ProgramUniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}{fd}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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

144

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 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, and 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: OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

145

• Members of type bool, int, uint, float, and double are respectively extracted from a buffer object by reading a single uint, int, uint, float, or double value at the specified offset. • Vectors with N elements with basic data types of bool, int, uint, float, or double 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 types dmatCxR and matCxR for double-precision and floating-point components respectively, or simply dmatC and matC respectively if C = R) are treated as an array of C column vectors, each consisting of R double-precision or 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 property using GetProgramResourceiv (see section 7.3.1). • Row-major matrices with C columns and R rows (using the types dmatCxR and matCxR for double-precision and floating-point components respectively, or simply dmatC and matC respectively if C = R) are treated as an array of R row vectors, each consisting of C double-precision or floatingpoint 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 property 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 (the value of UNIFORM_ARRAY_STRIDE) is an implementation-dependent value and may be queried after a program is linked.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES 7.6.2.2

146

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 their 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 . 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).

OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

147

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. 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 qualifier not supported for uniform blocks. When 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. 7.6.2.3

SPIR-V Uniform Offsets and Strides

The SPIR-V decorations GLSLShared or GLSLPacked must not be used. A variable in the Uniform Storage Class decorated as a Block or BufferBlock must be explicitly laid out using the Offset, ArrayStride, and MatrixStride decorations. These must follow the alignment rules listed above, but not necessarily the packing rules. More specifically, explicit SPIR-V offsets and strides must obey the following: Define the base alignment recursively as: • A scalar of size N has a base alignment of N . OpenGL 4.6 (Core Profile) - May 5, 2022

7.6. UNIFORM VARIABLES

148

• A two-component vector, with components of size N , has a base alignment of 2N . • A three- or four-component vector, with components of size N , has a base alignment of 4N . • An array has a base alignment equal to the base alignment of its element type, rounded up to a multiple of 16. • A structure has a base alignment equal to the largest base alignment of any of its members, rounded up to a multiple of 16. • A row-major matrix of C columns has a base alignment equal to the base alignment of a vector of C matrix components. • A column-major matrix has a base alignment equal to the base alignment of the matrix column type. Given this definition, blocks must be laid out such that: • The Offset decoration must be a multiple of its base alignment. • Any ArrayStride or MatrixStride decoration must be an integer multiple of the base alignment of the array or matrix. • The Offset decoration of a member must not place it between the end of a structure or an array and the next multiple of the base alignment of that structure or array. • The numeric order of Offset decorations need not follow member declaration order. With one exception: variables in the Uniform storage class with a decoration of BufferBlock do not need their base alignments rounded up to a multiple of 16.

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 using GetIntegerv with the constant MAX_UNIFORM_BUFFER_BINDINGS. Regions of buffer objects are bound as storage for uniform blocks by calling BindBuffer* commands (see section 6) with target set to UNIFORM_BUFFER. OpenGL 4.6 (Core Profile) - May 5, 2022

7.7. ATOMIC COUNTER BUFFERS

149

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 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 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.7. ATOMIC COUNTER BUFFERS

150

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, (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 may 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.8. SHADER BUFFER VARIABLES AND SHADER STORAGE BLOCKS151 layout qualifier in the shader text, either explicitly or implicitly, as described in

the OpenGL 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 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, 6.2.1, 6.3, 6.5, and 6.6; 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 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,

OpenGL 4.6 (Core Profile) - May 5, 2022

7.8. SHADER BUFFER VARIABLES AND SHADER STORAGE BLOCKS152 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. 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. After a program is linked, the command void ShaderStorageBlockBinding( uint program, uint storageBlockIndex, uint storageBlockBinding ); changes the active shader storage block with an assigned index of storageBlockIndex in program object program. ShaderStorageBlockBinding specifies OpenGL 4.6 (Core Profile) - May 5, 2022

7.9. INVOCATION GROUPS

153

that program will use the data store of the buffer object bound to the binding point storageBlockBinding to read and write the values of the buffer variables in the shader storage block identified by storageBlockIndex. 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 storageBlockIndex is not an active shader storage block index in program, or if storageBlockBinding is greater than or equal to the value of MAX_SHADER_STORAGE_BUFFER_BINDINGS. 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

Invocation Groups

An invocation group (see section 8.18(“Shader Invocation Group Functions”) of the OpenGL Shading Language Specification) for a compute shader is the set of invocations in a single workgroup. For graphics shaders, an invocation group is an implementation-dependent subset1 of the set of shader invocations of a given shader stage which are produced by a single drawing command. For MultiDraw* commands with drawcount greater than one, invocations from separate draws are in distinct invocation groups. 1

Because the partitioning of invocations into invocation groups is implementation-dependent and not observable, applications generally need to assume the worst case of all invocations in a draw belong to a single invocation group.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.10. SUBROUTINE UNIFORM VARIABLES

7.10

154

Subroutine Uniform Variables

Subroutine uniform variables are similar to uniform variables, except they are context state rather than program state, and apply only to a single program stage. Having subroutine uniforms be context state allows them to have different values if the program is used in multiple contexts simultaneously. There is a set of subroutine uniforms for each shader stage. A subroutine uniform may have an explicit location specified in the shader. At link time, all active subroutine uniforms without an explicit location will be assigned a unique location. The value of ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS for a program object is the largest specified or assigned location plus one. An assigned location will never take the location of an explicitly specified location, even if that subroutine uniform is inactive. Between the location zero and the value of ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS minus one there may be unused locations, either because they were not assigned a subroutine uniform or because the subroutine uniform was determined to be inactive by the linker. These locations will be ignored when assigning the subroutine index as described below. There is an implementation-dependent limit on the number of active subroutine uniform locations in each shader stage; a program will fail to link if the number of subroutine uniform locations required is greater than the value of MAX_SUBROUTINE_UNIFORM_LOCATIONS or if an explicit subroutine uniform location is outside this limit. For active subroutine uniforms declared as arrays, the declared array elements are assigned consecutive locations. Each function in a shader associated with a subroutine type is considered an active subroutine, unless the compiler conclusively determines that the function could never be assigned to an active subroutine uniform. The subroutine functions can be assigned an explicit index in the shader between zero and the value of MAX_SUBROUTINES minus one. At link time, all active subroutines without an explicit index will be assigned an index between zero and the value of ACTIVE_SUBROUTINES minus one. An assigned index will never take the same index of an explicitly specified index in the shader, even if that subroutine is inactive. Between index zero and the vaue of ACTIVE_SUBROUTINES minus one there may be unused indices either because they weren’t assigned an index by the linker or because the subroutine was determined to be inactive by the linker. If there are no explicitly defined subroutine indices in the shader the implementation must assign indices between zero and the value of ACTIVE_SUBROUTINES minus one with no index unused. It is recommended, but not required, that the application assigns a range of tightly packed indices starting from zero to avoid indices between zero and the value of ACTIVE_SUBROUTINES minus one being unused. To determine the set of active subroutines and subroutines used by a particOpenGL 4.6 (Core Profile) - May 5, 2022

7.10. SUBROUTINE UNIFORM VARIABLES

155

Interface

Shader Type

VERTEX_SUBROUTINE TESS_CONTROL_SUBROUTINE TESS_EVALUATION_SUBROUTINE GEOMETRY_SUBROUTINE FRAGMENT_SUBROUTINE COMPUTE_SUBROUTINE

VERTEX_SHADER TESS_CONTROL_SHADER TESS_EVALUATION_SHADER GEOMETRY_SHADER FRAGMENT_SHADER COMPUTE_SHADER

Table 7.9: Interfaces for active subroutines for a particular shader type in a program.

Interface

Shader Type

VERTEX_SUBROUTINE_UNIFORM TESS_CONTROL_SUBROUTINE_UNIFORM TESS_EVALUATION_SUBROUTINE_UNIFORM GEOMETRY_SUBROUTINE_UNIFORM FRAGMENT_SUBROUTINE_UNIFORM COMPUTE_SUBROUTINE_UNIFORM

VERTEX_SHADER TESS_CONTROL_SHADER TESS_EVALUATION_SHADER GEOMETRY_SHADER FRAGMENT_SHADER COMPUTE_SHADER

Table 7.10: Interfaces for active subroutine uniforms for a particular shader type in a program.

ular shader stage of a program, applications can query the properties and active resources of the interfaces for the shader type, as listed in tables 7.9 and 7.10. Additionally, dedicated commands are provided to determine properties of active subroutines and active subroutine uniforms. The commands uint GetSubroutineIndex( uint program, enum shadertype, const char *name ); void GetActiveSubroutineName( uint program, enum shadertype, uint index, sizei count, sizei *length, char *name ); are equivalent to GetProgramResourceIndex(program, programInterface, name); and OpenGL 4.6 (Core Profile) - May 5, 2022

7.10. SUBROUTINE UNIFORM VARIABLES

156

GetProgramResourceName(program, programInterface, index, count, length, name); respectively, where programInterface is taken from table 7.9 according to the value of shadertype. The commands int GetSubroutineUniformLocation( uint program, enum shadertype, const char *name ); void GetActiveSubroutineUniformName( uint program, enum shadertype, uint index, sizei count, sizei *length, char *name ); void GetActiveSubroutineUniformiv( uint program, enum shadertype, uint index, enum pname, int *values ); are equivalent to GetProgramResourceLocation(program, programInterface, name); GetProgramResourceName(program, programInterface, index, count, length, name); and GetProgramResourceiv(program, programInterface, index, 1, &pname, maxSize, NULL, values); respectively, where programInterface is taken from table 7.10 according to the value of shadertype. For GetActiveSubroutineUniformiv, pname must be one of NUM_COMPATIBLE_SUBROUTINES or COMPATIBLE_SUBROUTINES, and maxSize is taken to specify a sufficiently large buffer to receive all values that would be written to params. The command void UniformSubroutinesuiv( enum shadertype, sizei count, const uint *indices ); will load all active subroutine uniforms for shader stage shadertype with subroutine indices from indices, storing indices[i] into the uniform at location i. The indices for any locations between zero and the value of ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS minus one which are not used will be ignored.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.11. SAMPLERS

157

Errors An INVALID_ENUM error is generated if shadertype is not one of the values in table 7.1. An INVALID_VALUE error is generated if count is negative, is not equal to the value of ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS for the program currently in use at shader stage shadertype, or if the uniform at location i is used and the value in indices[i] is greater than or equal to the value of ACTIVE_SUBROUTINES for the shader stage. An INVALID_VALUE error is generated if the value of indices[i] for a used uniform location specifies an unused subroutine index. An INVALID_OPERATION error is generated if, for any subroutine index being loaded to a particular uniform location, the function corresponding to the subroutine index was not associated (as defined in section 6.1.2 of the OpenGL Shading Language Specification) with the type of the subroutine variable at that location. An INVALID_OPERATION error is generated if no program is active for the shader stage identified by shadertype. Each subroutine uniform must have at least one subroutine to assign to the uniform. A program will fail to link if any stage has one or more subroutine uniforms that has no subroutine associated with the subroutine type of the uniform. When the active program for a shader stage is re-linked or changed by a call to UseProgram, BindProgramPipeline, or UseProgramStages, subroutine uniforms for that stage are reset to arbitrarily chosen default functions with compatible subroutine types.

7.11

Samplers

Samplers are special uniforms used in the OpenGL 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 value of i may range from zero to the implementationdependent 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.12. IMAGES

158

The location of a sampler is queried with GetUniformLocation, just like any uniform variable. Sampler values must be set by calling Uniform1i{v}. Errors 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.

7.12

Images

Images are special uniforms used in the OpenGL 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 8.26. 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 Shading Language Specification for more details). The location of an image variable needs to be queried with GetUniformLocation, just like any uniform variable. Image values must be set by calling Uniform1i{v}. Unlike samplers, there is no limit on the number of active image variables that may be used by a program or by any particular shader. However, given that there OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

159

is an implementation-dependent limit on the number of unique image units, the actual number of images that may be used by all shaders in a program is limited.

7.13

Shader Memory Access

As described in the OpenGL Shading Language Specification, shaders may perform random-access reads and writes to buffer object memory by reading from, 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.

7.13.1

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, tessellation evaluation, 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 14.9.1), scissor test (see section 14.9.2), or is discarded by any of the multisample fragment operations (see section 14.9.3), the fragment shader will not be executed In addition, if early per-fragment tests are enabled (see section 14.9), the fragment shader will not be executed if the fragment is discarded during the early per-fragment tests.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

160

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 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, then stores, atomics, and atomic counter updates 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 always 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 largely undefined. However, when executing a shader whose inputs are generated from a previous programmable stage, the shader invocations from the previous stage are guaranteed to have executed far enough to generate final values for all next-stage inputs. That implies shader completion for all stages except geometry; geometry shaders are guaranteed only to have executed far enough to emit all vertices used to generate the primitive being processed by the fragment shader. 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. The only case where such a guarantee is made is when the inputs of one shader invocation are generated from the outputs of a shader invocation in a previous stage. 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 built-in 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

161

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 and atomic counter 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.

7.13.2

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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

162

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. • 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, store, and atomic functions issued after the barrier will reflect data written by shaders prior to the barrier. Additionally, image stores and atomics 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, MultiDraw*IndirectCount 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, PARAMETER_BUFFER and DISPATCH_INDIRECT_BUFFER bindings, respectively. • 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

163

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*, ClearTex*Image, CopyTex*, or CompressedTex*, and reads via GetTexImage 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, 6.2.1, 6.3, 6.6, 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. • CLIENT_MAPPED_BUFFER_BARRIER_BIT: Access by the client to persistent mapped regions of buffer objects will reflect data written by shaders prior to the barrier. Note that this may cause additional synchronization operations. • QUERY_BUFFER_BARRIER_BIT: Writes of buffer objects via the QUERY_BUFFER binding (see section 4.2.3) 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. • 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. • 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: Memory accesses using shader atomic counter built-in functions issued after the barrier will reflect data written by shaders prior to the barrier. Additionally, atomic counter function invocations after the barrier will not execute until all memory accesses (e.g., loads, stores, texture fetches, vertex fetches) initiated prior to the barrier complete. • SHADER_STORAGE_BARRIER_BIT: Memory accesses using shader buffer variables issued after the barrier will reflect data written by shaders prior to OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

164

the barrier. Additionally, assignments to and atomic operations performed on shader buffer variables after the barrier will not execute until all memory accesses 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 and 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 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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.13. SHADER MEMORY ACCESS

165

• 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 by one shader invocation and consumed by other shader invocations launched as a result of its execution (dependent invocations) should use coherent variables in the producing shader invocation and call memoryBarrier after the last write. The consuming shader invocation should also use coherent variables. • 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 ); behaves as described above for MemoryBarrier, with two differences: First, it narrows the region under consideration so that only reads and writes of prior fragment shaders that are invoked for a smaller region of the framebuffer will be completed/reflected prior to subsequent reads and writes 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 • SHADER_STORAGE_BARRIER_BIT • TEXTURE_FETCH_BARRIER_BIT

OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

166

• 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 and 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.

7.14

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 or specialized 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 an empty 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. If pname is SPIR_V_BINARY, TRUE is returned if the shader has been successfully associated with a SPIR-V binary module by the ShaderBinary command, and FALSE is returned otherwise. OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

167

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, SHADER_SOURCE_LENGTH, or SPIR_V_BINARY. 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 an empty 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 no name reflection information is available, one is returned. OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

168

If pname is ACTIVE_UNIFORMS, the number of active uniforms is returned. If no active uniforms exist, zero is returned. 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 no name reflection information is available, one 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 no name reflection information is available, one 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 no active uniform blocks exist, zero is returned. If no name reflection information is available, one 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

169

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 19), as specified by its input layout qualifier(s), is returned. 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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

170

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 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 is 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 an empty 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 may 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

OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

171

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 validation 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 may 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 or specialization 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 ei-

OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

172

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 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 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 may 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 OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

173

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 range[1] = blog2 (|max|)c precision points to an integer in which 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 4.7(“Precision and Precision Qualifiers”) of the OpenGL 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 GetUniformdv( uint program, int location, double *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 GetnUniformdv( uint program, int location, sizei bufSize, double *params );

OpenGL 4.6 (Core Profile) - May 5, 2022

7.14. SHADER, PROGRAM, AND PROGRAM PIPELINE QUERIES

174

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. 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_VALUE error is generated by GetnUniform* if bufSize is negative. An INVALID_OPERATION error is generated by GetnUniform* if the buffer size required to store the requested data is greater than bufSize. The command void GetUniformSubroutineuiv( enum shadertype, int location, uint *params ); returns the value of the subroutine uniform at location location for shader stage shadertype of the current program. If location represents an unused location, the value INVALID_INDEX is returned and no error is generated. Errors An INVALID_ENUM error is generated if shadertype is not one of the values in table 7.1. An INVALID_VALUE error is generated if location is greater than or equal to the value of ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS for the shader currently in use at shader stage shadertype.

OpenGL 4.6 (Core Profile) - May 5, 2022

7.15. REQUIRED STATE

175

An INVALID_OPERATION error is generated if no program is active. The command void GetProgramStageiv( uint program, enum shadertype, enum pname, int *values ); returns properties of the program object program specific to the programmable stage corresponding to shadertype in values. The parameter value to return is specified by pname. If pname is ACTIVE_SUBROUTINE_UNIFORMS, the number of active subroutine variables in the stage is returned. If pname is ACTIVE_SUBROUTINE_UNIFORM_LOCATIONS, the number of active subroutine variable locations in the stage is returned. If pname is ACTIVE_SUBROUTINES, the number of active subroutines in the stage is returned. If pname is ACTIVE_SUBROUTINE_UNIFORM_MAX_LENGTH or ACTIVE_SUBROUTINE_MAX_LENGTH, the length of the longest subroutine uniform or subroutine name, respectively, for the stage is returned. The returned name length includes space for a null terminator. If there is no shader of type shadertype in program, the values returned will be consistent with a shader with no subroutines or subroutine uniforms. 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 shadertype is not one of the values in table 7.1.

7.15

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. OpenGL 4.6 (Core Profile) - May 5, 2022

7.15. REQUIRED STATE

176

• A boolean holding the SPIR-V binary status, 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. • 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 vertex shaders consists of: OpenGL 4.6 (Core Profile) - May 5, 2022

7.15. REQUIRED STATE

177

• A bit indicating whether or not program point size mode (section 14.4.1) is enabled, initially disabled. 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 23.32-23.42 describe additional program object state specific to program binaries, geometry shaders, tessellation control and evaluation shaders, shader subroutines, and uniform blocks. Table 23.43 describes state related to vertex and geometry shaders that is not program object state.

OpenGL 4.6 (Core Profile) - 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 8.12-8.26 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 8.12-8.13. 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 one-, two-, or three-dimensional arrays of texels (see section 2.6.6). One-, two-, and three-dimensional textures consist respectively of one-, two-, or threedimensional texture images. One- and two-dimensional array textures are arrays of one- or two-dimensional images. Each image consists of one or more layers. Two-dimensional multisample and two-dimensional multisample array textures are special two-dimensional and two-dimensional array textures, respectively, 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

178

8.1. TEXTURE OBJECTS

179

cube map array is a collection of cube map layers stored as a two-dimensional array texture. When accessing a cube map array, 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. Rectangle textures are special two-dimensional textures consisting of only a single image and accessed using unnormalized coordinates. 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 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.1. TEXTURE OBJECTS

180

TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_RECTANGLE, 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 8.22. 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. 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. An INVALID_OPERATION error is generated if texture is not zero or a

OpenGL 4.6 (Core Profile) - May 5, 2022

8.1. TEXTURE OBJECTS

181

name returned from a previous call to GenTextures, or if such a name has since been deleted. The command void BindTextures( uint first, sizei count, const uint *textures ); binds count existing texture objects to texture image units numbered first through f irst + count − 1. If textures is not NULL, it specifies an array of count values, each of which must be zero or the name of an existing texture object. When an entry in textures is the name of an existing texture object, that object is bound to the target, in the corresponding texture unit, that was specified when the object was created. When an entry in textures is zero, each of the targets enumerated at the beginning of this section is reset to its default texture for the corresponding texture image unit. If textures is NULL, each target of each affected texture image unit from first to f irst + count − 1 is reset to its default texture. BindTextures is equivalent (assuming no errors are generated to): for (i = 0; i < count; i++) { uint texture; if (textures == NULL) { texture = 0; } else { texture = textures[i]; } ActiveTexture(TEXTURE0 + f irst + i); if (texture != 0) { enum target = /* target of textures[i] */; BindTexture(target, textures[i]); } else { for (target in all supported targets) { BindTexture(target, 0); } } } except that the active texture selector retains its original value upon completion of the command, and that textures will not be created if they do not exist. The values specified in textures will be checked separately for each texture image unit. When a value for a specific texture image unit is invalid, the state for OpenGL 4.6 (Core Profile) - May 5, 2022

8.1. TEXTURE OBJECTS

182

that texture image unit will be unchanged and an error will be generated. However, state for other texture image units will still be changed if their corresponding values are valid. Errors An INVALID_OPERATION error is generated if f irst + count is greater than the number of texture image units supported by the implementation. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if any value in textures is not zero or the name of an existing texture object (per binding). The command void BindTextureUnit( uint unit, uint texture ); binds an existing texture object to the texture unit numbered unit. texture must be zero or the name of an existing texture object. When texture is the name of an existing texture object, that object is bound to the target, in the corresponding texture unit, that was specified when the object was created. When texture is zero, each of the targets enumerated at the beginning of this section is reset to its default texture for the corresponding texture image unit. Errors An INVALID_OPERATION error is generated by BindTextureUnit if texture is not zero or the name of an existing texture object. Texture objects may also be created with the command void CreateTextures( enum target, sizei n, uint *textures ); CreateTextures returns n previously unused texture names in textures, each representing a new texture object that is a state vector comprising all the state and with the same initial values listed in section 8.22. The new texture objects are and remain textures of the dimensionality and type specified by target until they are deleted. Errors An INVALID_VALUE error is generated if n is negative. Texture objects are deleted by calling OpenGL 4.6 (Core Profile) - May 5, 2022

8.2. SAMPLER OBJECTS

183

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 one-, two-, and three- dimensional, one- and two-dimensional array, rectangle, 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 8.22. A GL texture object includes both categories. The first category represents dimensionality and other image parameters, and the second category OpenGL 4.6 (Core Profile) - May 5, 2022

8.2. SAMPLER OBJECTS

184

represents sampling state. Additionally, a sampler object may be created to encapsulate only 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 23.18. Errors An INVALID_VALUE error is generated if count is negative. Sampler objects may also be created with the command void CreateSamplers( sizei n, uint *samplers ); CreateSamplers returns n previously unused sampler names in samplers, each representing a new sampler object which is a state vector comprising all the state and with the same initial values listed in table 23.181 . Errors An INVALID_VALUE error is generated if n 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 ); 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 4.6 (Core Profile) - May 5, 2022

8.2. SAMPLER OBJECTS

185

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. If state is present in a sampler object bound to a texture unit that would have been rejected by a call to TexParameter* for the texture bound to that unit, the behavior of the implementation is as if the texture were incomplete. For example, if TEXTURE_WRAP_S or TEXTURE_WRAP_T is set to REPEAT, MIRRORED_REPEAT, or MIRROR_CLAMP_TO_EDGE on the sampler object bound to a texture unit and the texture bound to that unit is a rectangle texture, the texture will be considered incomplete. Sampler object state which does not affect sampling for the type of texture bound to a texture unit, such as TEXTURE_WRAP_R for a rectangle texture, does not affect completeness. 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 the sampler object named zero to that unit), the modified state is again replaced with the sampler state associated with the texture object bound to that texture unit. 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 command void BindSamplers( uint first, sizei count, const uint *samplers ); binds count existing sampler objects to texture image units numbered first through f irst + count − 1. If samplers is not NULL, it specifies an array of count values, each of which must be zero or the name of an existing sampler object. If samplers is NULL, each affected texture image unit from first through f irst + count − 1 will be reset to have no bound sampler object. BindSamplers is equivalent (assuming no errors are generated to):

OpenGL 4.6 (Core Profile) - May 5, 2022

8.2. SAMPLER OBJECTS

186

for (i = 0; i < count; i++) { if (samplers == NULL) { BindSampler(f irst + i, 0); } else { BindSampler(f irst + i, samplers[i]); } } The values specified in samplers will be checked separately for each texture image unit. When a value for a specific texture image unit is invalid, the state for that texture image unit will be unchanged and an error will be generated. However, state for other texture image units will still be changed if their corresponding values are valid. Errors An INVALID_OPERATION error is generated if f irst + count is greater than the number of texture image units supported by the implementation. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if any value in samplers is not zero or the name of an existing sampler object (per binding). 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 *param ); 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 param is the new value of that parameter. pname must be one of the sampler state names in table 23.18. Texture state listed in tables 23.16- 23.17 but not listed here and in the sampler state in table 23.18 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.2. SAMPLER OBJECTS

187

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. 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 23.18. An INVALID_ENUM error is generated if SamplerParameter{if} is called for a non-scalar parameter (pname TEXTURE_BORDER_COLOR or TEXTURE_SWIZZLE_RGBA). 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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

8.3. SAMPLER OBJECT QUERIES

188

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 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 23.18. 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 23.18.

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. OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES Parameter Name UNPACK_SWAP_BYTES UNPACK_LSB_FIRST UNPACK_ROW_LENGTH UNPACK_SKIP_ROWS UNPACK_SKIP_PIXELS UNPACK_ALIGNMENT UNPACK_IMAGE_HEIGHT UNPACK_SKIP_IMAGES UNPACK_COMPRESSED_BLOCK_WIDTH UNPACK_COMPRESSED_BLOCK_HEIGHT UNPACK_COMPRESSED_BLOCK_DEPTH UNPACK_COMPRESSED_BLOCK_SIZE

189 Type boolean boolean integer integer integer integer integer integer integer integer integer integer

Initial Value FALSE FALSE

0 0 0 4 0 0 0 0 0 0

Valid Range TRUE/FALSE TRUE/FALSE [0, ∞) [0, ∞) [0, ∞) 1,2,4,8 [0, ∞) [0, ∞) [0, ∞) [0, ∞) [0, ∞) [0, ∞)

Table 8.1: PixelStore* parameters pertaining to one or more of TexImage*D, TexSubImage*D, CompressedTexImage*D and CompressedTexSubImage*D.

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 PixelStore*.

8.4.1

Pixel Storage Modes and Pixel Buffer Objects

Pixel storage modes affect the operation of TexImage*D, TexSubImage*D, CompressedTexImage*D, CompressedTexSubImage*D, and ReadPixels when one of these commands is issued. Pixel storage modes are set with void PixelStore{if}( enum pname, T 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 18.1 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 18.1. An INVALID_VALUE error is generated if param is outside the given range

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

190

for the corresponding pname in table 8.1 or 18.1. 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.

8.4.2 This subsection is only defined in the compatibility profile.

8.4.3 This subsection is only defined in the compatibility profile.

8.4.4

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. 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 8.2. Not all combinations of format and type are valid. Errors An INVALID_OPERATION error is generated if format is one of the INTEGER component formats defined in table 8.3 and type is one of the

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

191

byte, short, int, float, or packed pixel component data stream Unpack

Convert to Float

Pixel Storage Operations

Expansion to RGBA 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

192

floating-point types defined in table 8.2. Some additional constraints on the combinations of format and type values that are accepted are discussed below. Additional restrictions may be imposed by specific commands. 8.4.4.1

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 8.3 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 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 8.2 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). By default the values of each GL data type are interpreted as they would be specified in the language of the client’s GL binding. If UNPACK_SWAP_BYTES is enabled, however, then the values are interpreted with the bit orderings modified as per table 8.4. The modified bit orderings are defined only if the GL data type ubyte has eight bits, and then for each specific GL data type only if that type is represented with 8, 16, or 32 bits. OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

type Parameter Token Name UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT HALF_FLOAT FLOAT UNSIGNED_BYTE_3_3_2 UNSIGNED_BYTE_2_3_3_REV UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_5_6_5_REV UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_4_4_4_4_REV UNSIGNED_SHORT_5_5_5_1 UNSIGNED_SHORT_1_5_5_5_REV UNSIGNED_INT_8_8_8_8 UNSIGNED_INT_8_8_8_8_REV UNSIGNED_INT_10_10_10_2 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

193

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

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

FloatingPoint No No No No No No Yes Yes No No No No No No No No No No No No No Yes Yes No

Table 8.2: 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 8.4.4.2. Floating-point types are incompatible with INTEGER formats as described above.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

Format Name STENCIL_INDEX DEPTH_COMPONENT DEPTH_STENCIL RED GREEN BLUE RG RGB RGBA BGR BGRA RED_INTEGER GREEN_INTEGER BLUE_INTEGER RG_INTEGER RGB_INTEGER RGBA_INTEGER BGR_INTEGER BGRA_INTEGER

194

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

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

Table 8.3: Pixel data formats. The second column gives a description of and the number and order of elements in a group. Unless specified as an index, formats yield components. Components are floating-point unless prefixed with the letter ’i’, which indicates they are integer.

Element Size 8 bit 16 bit 32 bit

Default Bit Ordering [7..0] [15..0] [31..0]

Modified Bit Ordering [7..0] [7..0][15..8] [7..0][15..8][23..16][31..24]

Table 8.4: Bit ordering modification of elements when UNPACK_SWAP_BYTES is enabled. These reorderings are defined only when GL data type ubyte has 8 bits, and then only for GL data types with 8, 16, or 32 bits. Bit 0 is the least significant.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

195

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. 8.4.4.2

Special Interpretations

A type matching one of the types in table 8.5 is a special case in which all the components of each group are packed into a single unsigned byte, unsigned short, 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 index. 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 8.5. Errors An INVALID_OPERATION error is generated by any command processing

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

196

ROW_LENGTH

subimage SKIP_PIXELS

SKIP_ROWS

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

pixel rectangles if a mismatch occurs. Bitfield locations of the first, second, third, and fourth components of each packed pixel type are illustrated in tables 8.6- 8.9. Each bitfield is interpreted as an unsigned integer value. Components are normally packed with the first component in the most significant bits of the bitfield, and successive components 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 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

type Parameter Token Name UNSIGNED_BYTE_3_3_2 UNSIGNED_BYTE_2_3_3_REV UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_5_6_5_REV UNSIGNED_SHORT_4_4_4_4

197

GL Data Type ubyte ubyte ushort ushort ushort

Number of Components 3 3 3 3 4

UNSIGNED_SHORT_4_4_4_4_REV

ushort

UNSIGNED_SHORT_5_5_5_1

ushort

UNSIGNED_SHORT_1_5_5_5_REV

ushort

UNSIGNED_INT_8_8_8_8

uint

UNSIGNED_INT_8_8_8_8_REV

uint

UNSIGNED_INT_10_10_10_2

uint

UNSIGNED_INT_2_10_10_10_REV

uint

UNSIGNED_INT_24_8 UNSIGNED_INT_10F_11F_11F_REV UNSIGNED_INT_5_9_9_9_REV FLOAT_32_UNSIGNED_INT_24_8_REV

uint uint uint n/a

2 3 4 2

Table 8.5: Packed pixel formats.

OpenGL 4.6 (Core Profile) - May 5, 2022

Matching Pixel Formats RGB, RGB_INTEGER RGB, RGB_INTEGER RGB, RGB_INTEGER RGB, RGB_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER RGBA, BGRA, RGBA_INTEGER, BGRA_INTEGER DEPTH_STENCIL RGB RGB DEPTH_STENCIL

8.4. PIXEL RECTANGLES

198

UNSIGNED_BYTE_3_3_2: 7

1st Component

2nd

0 3rd

UNSIGNED_BYTE_2_3_3_REV: 7

6 3rd

4 2nd

1st Component

Table 8.6: UNSIGNED_BYTE formats. Bit numbers are indicated for each component.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.4. PIXEL RECTANGLES

199

UNSIGNED_SHORT_5_6_5: 15

1st Component

2nd

3rd

UNSIGNED_SHORT_5_6_5_REV: 15

3rd

2nd

1st Component

UNSIGNED_SHORT_4_4_4_4: 15

1st Component

2nd

3rd

4th

UNSIGNED_SHORT_4_4_4_4_REV: 15

4th

3rd

2nd

1st Component

UNSIGNED_SHORT_5_5_5_1: 15

1st Component

2nd

3rd

0 4th

UNSIGNED_SHORT_1_5_5_5_REV: 15 4th

3rd

2nd

Table 8.7: UNSIGNED_SHORT formats

OpenGL 4.6 (Core Profile) - May 5, 2022

1st Component

8.4. PIXEL RECTANGLES

200

UNSIGNED_INT_8_8_8_8: 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 1st Component

2nd

3rd

4th

UNSIGNED_INT_8_8_8_8_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

2nd

1st Component

UNSIGNED_INT_10_10_10_2: 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 1st Component

2nd

3rd

0 4th

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 4th

3rd

2nd

1st Component

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

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 3rd

2nd

1st Component

UNSIGNED_INT_5_9_9_9_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

2nd

Table 8.8: UNSIGNED_INT formats

OpenGL 4.6 (Core Profile) - May 5, 2022

1st Component

8.4. PIXEL RECTANGLES

201

FLOAT_32_UNSIGNED_INT_24_8_REV: 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9

1st Component

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

Table 8.9: FLOAT_UNSIGNED_INT formats

OpenGL 4.6 (Core Profile) - May 5, 2022

2nd

8.4. PIXEL RECTANGLES Format RGB RGBA BGRA DEPTH_STENCIL

202

First Component red red blue depth

Second Component green green green stencil

Third Component blue blue red

Fourth Component alpha alpha

Table 8.10: Packed pixel field assignments.

The assignment of components to fields in the packed pixel is as described in table 8.10. Byte swapping, if enabled, is performed before the components are extracted from each pixel. 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 table 8.8. 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 table 8.8. 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).

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION 8.4.4.3

203

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 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. 8.4.4.4

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.

8.4.5 This subsection is only defined in the compatibility profile.

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. Additionally, target may be either PROXY_TEXTURE_3D for a threedimensional proxy texture, PROXY_TEXTURE_2D_ARRAY for a two-dimensional proxy array texture, or PROXY_TEXTURE_CUBE_MAP_ARRAY for a cube map array texture, as discussed in section 8.22. 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 8.4.4.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

204

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 8.4.4. 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 8.4.4. The selected groups are transferred to the GL as described in section 8.4.4 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 8.2), 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]. • Stencil index values are masked by 2n − 1, where n is the number of stencil bits in the internal format resolution (see below). • Otherwise, values are not modified. If the base internal format is DEPTH_STENCIL and format is not DEPTH_STENCIL, then the values of the stencil index texture components are undefined. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

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

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

205 Internal Components D D,S R R,G R,G,B R,G,B,A S

Table 8.11: Conversion from RGBA, depth, and stencil pixel components to internal texture components. Texture components R, G, B, and A are converted back to RGBA colors during filtering as shown in table 15.1.

from) internalformat. Table 8.11 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. internalformat may be specified as one of the internal format symbolic constants listed in table 8.11, as one of the sized internal format symbolic constants listed in tables 8.12- 8.13, as one of the generic compressed internal format symbolic constants listed in table 8.14, or as one of the specific compressed internal format symbolic constants (if listed in table 8.14). 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_1D, TEXTURE_2D, TEXTURE_2D_MULTISAMPLE, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_2D_MULTISAMPLE_ARRAY, TEXTURE_RECTANGLE, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY, PROXY_TEXTURE_1D, PROXY_TEXTURE_2D, PROXY_TEXTURE_2D_MULTISAMPLE, PROXY_TEXTURE_1D_ARRAY, PROXY_TEXTURE_2D_ARRAY, PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY, PROXY_TEXTURE_RECTANGLE, PROXY_TEXTURE_CUBE_MAP, or PROXY_TEXTURE_CUBE_MAP_ARRAY. An INVALID_OPERATION error is generated if these formats are used in conjunction with any other target. Textures with a base internal format of DEPTH_COMPONENT or DEPTH_STENCIL require either depth component data or depth/stencil component data. Textures with other base internal formats require RGBA component data. Textures with integer internal formats (see table 8.12) require integer data. In addition to the specific compressed internal formats listed in table 8.14, the GL provides a mechanism to query token values for specific compressed internal

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

206

formats, suitable for general-purpose2 usage. Formats with restrictions that need to be specifically understood prior to use will not be returned by this query. The number of specific compressed internal formats is obtained by querying the value of NUM_COMPRESSED_TEXTURE_FORMATS. The set of specific compressed internal formats is obtained by querying COMPRESSED_TEXTURE_FORMATS with GetIntegerv, returning an array containing that number of values. Generic compressed internal formats are never used directly as the internal formats of texture images. If internalformat is one of the six generic compressed internal formats, its value is replaced by the symbolic constant for a specific compressed internal format of the GL’s choosing with the same base internal format. If no specific compressed format is available, internalformat is instead replaced by the corresponding base internal format. If internalformat is given as or mapped to a specific compressed internal format, but the GL can not support images compressed in the chosen internal format for any reason (e.g., the compression format might not support 3D textures), internalformat is replaced by the corresponding base internal format and the texture image will not be compressed by the GL. 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, referred to as the effective internal format. The effective internal format chosen may change depending only on the values of format and type, and affects format compatibility for commands such as TextureView (see section 8.18) and CopyImageSubData (see section 18.3.2). 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 8.11; the type (unsigned int, float, etc.) is assigned the same type specified by internalformat; and the memory allocation per texture component is assigned by the GL to match the allocations listed in tables 8.12- 8.13 as closely as possible. (The definition of closely is left up to the implementation. However, a non-zero number of bits must be allocated for each component whose desired allocation in tables 8.12- 8.13 is non-zero, and zero bits must be allocated for all other components).

8.5.1

Required Texture Formats

Implementations are required to support at least one allocation of internal component resolution for each type (unsigned int, float, etc.) for each base internal format. 2 These queries have been deprecated in OpenGL 4.2, because the vagueness of the term “generalpurpose” makes it possible for implementations to choose to return no formats from the query.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

207

In addition, implementations are required to support the following sized and compressed 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 the corresponding table: • Color formats which are checked in the “Req. tex.” column of table 8.12. • All of the specific compressed texture formats in table 8.14. • Depth, depth+stencil, and stencil formats which are checked in the “Req. format” column of table 8.13.

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: 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:

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

208

1 maxs = expp −B−N + 2 2 ( expp , 0 ≤ maxs < 2N exps = expp + 1, maxs = 2N maxc

Finally, three integer values in the range 0 to 2N − 1 are computed: 1 reds = exps −B−N + 2 2 1 greenc greens = exps −B−N + 2 2 bluec 1 blues = exps −B−N + 2 2

redc

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”. Sized Internal Format R8 R8_SNORM R16 R16_SNORM RG8 RG8_SNORM RG16 RG16_SNORM R3_G3_B2 RGB4 RGB5 RGB565 RGB8

Base Internal Format

Bits/component S are shared bits G B A

R S RED 8 RED s8 RED 16 RED s16 RG 8 8 RG s8 s8 RG 16 16 RG s16 s16 RGB 3 3 2 RGB 4 4 4 RGB 5 5 5 RGB 5 6 5 RGB 8 8 8 Sized internal color formats continued on next page OpenGL 4.6 (Core Profile) - May 5, 2022

Req. rend.

Req. tex.

8.5. TEXTURE IMAGE SPECIFICATION

209

Sized internal color formats continued from previous page Sized Base Bits/component CR Req. Internal Internal S are shared bits rend. Format Format R G B A S RGB8_SNORM RGB s8 s8 s8 RGB10 RGB 10 10 10 RGB12 RGB 12 12 12 RGB16 RGB 16 16 16 RGB16_SNORM RGB s16 s16 s16 RGBA2 RGBA 2 2 2 2 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 RGBA12 RGBA 12 12 12 12 RGBA16 RGBA 16 16 16 16 RGBA16_SNORM RGBA s16 s16 s16 s16 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 Sized internal color formats continued on next page

OpenGL 4.6 (Core Profile) - May 5, 2022

Req. tex.

8.5. TEXTURE IMAGE SPECIFICATION

210

Sized internal color formats continued from previous page Sized Base Bits/component CR Req. Internal Internal S are shared bits rend. Format Format R G B A S RG8I RG i8 i8 RG8UI RG ui8 ui8 RG16I RG i16 i16 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 8.12: Correspondence of sized internal color formats to base internal formats, internal data type, and desired component resolutions 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”, “Req. tex.”, and “Req. rend.” columns are described in sections 9.4, 8.5.1, and 9.2.5, respectively.

Req. tex.

If a compressed internal format is specified, the mapping of the R, G, B, and A values to texture components is equivalent to the mapping of the corresponding base internal format’s components, as specified in table 8.11. The specified image is compressed using a (possibly lossy) compression algorithm chosen by the GL. A GL implementation may vary its allocation of internal component resolution OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION Sized Internal Format

Base Internal Format

DEPTH_COMPONENT16 DEPTH_COMPONENT24 DEPTH_COMPONENT32 DEPTH_COMPONENT32F DEPTH24_STENCIL8 DEPTH32F_STENCIL8 STENCIL_INDEX1 STENCIL_INDEX4 STENCIL_INDEX8 STENCIL_INDEX16

DEPTH_COMPONENT DEPTH_COMPONENT DEPTH_COMPONENT DEPTH_COMPONENT DEPTH_STENCIL DEPTH_STENCIL STENCIL_INDEX STENCIL_INDEX STENCIL_INDEX STENCIL_INDEX

211 D bits 16 24 32 f32 24 f32

S bits

ui8 ui8 ui1 ui4 ui8 ui16

Req. format

Table 8.13: Correspondence of sized internal depth and stencil formats to base internal formats, internal data type, and desired component resolutions 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, and no prefix is fixed-point. The “Req. format” column is described in section 8.5.1.

or compressed internal format based on any TexImage3D, TexImage2D (see below), or TexImage1D (see below) parameter (except target), but the allocation and chosen compressed image format must not be a function of any other state and cannot be changed once they are established. In addition, the choice of a compressed image format may not be affected by the data parameter. Allocations must be invariant; the same allocation and compressed image format must be chosen each time a texture image is specified with the same parameter values. These allocation rules also apply to proxy textures, which are described in section 8.22.

8.5.3

Texture Image Structure

The texture 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 8.11. Counting from zero, each resulting nth texel is assigned internal integer coordinates (i, j, k), where

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

212

Compressed Internal Format

Base Internal Format

Type

Border Type

Copyable

COMPRESSED_RED COMPRESSED_RG COMPRESSED_RGB COMPRESSED_RGBA COMPRESSED_SRGB COMPRESSED_SRGB_ALPHA COMPRESSED_RED_RGTC1 COMPRESSED_SIGNED_RED_RGTC1 COMPRESSED_RG_RGTC2 COMPRESSED_SIGNED_RG_RGTC2 COMPRESSED_RGBA_BPTC_UNORM COMPRESSED_SRGB_ALPHA_BPTC_UNORM COMPRESSED_RGB_BPTC_SIGNED_FLOAT COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT 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_R11_EAC COMPRESSED_SIGNED_R11_EAC COMPRESSED_RG11_EAC COMPRESSED_SIGNED_RG11_EAC

RED RG RGB RGBA RGB RGBA RED RED RG RG RGBA RGBA

Generic Generic Generic Generic Generic Generic Specific Specific Specific Specific Specific Specific

unorm unorm unorm unorm unorm unorm unorm snorm unorm snorm unorm unorm

RGB

Specific

float

RGB

Specific

float

RGB RGB RGB

Specific Specific Specific

unorm unorm unorm

RGB

Specific

unorm

RGBA RGBA

Specific Specific

unorm unorm

RED RED RG RG

Specific Specific Specific Specific

unorm snorm unorm snorm

Table 8.14: Generic and specific compressed internal formats. Specific formats are described in appendix D. The “Border Type” field determines how border colors are clamped, as described in section 8.14.2. The “Copyable” field specifies if the format may be used with non-compressed texture image commands.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

213

i = n mod width j n k j= mod height width n mod depth k= 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 and the face is given by f ace = k mod 6. The face number corresponds to the cube map faces as shown in table 9.3. 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-ofdetail number of zero. level must be zero or more. ws , hs , and ds are the specified image width, height and depth respectively. border must be 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 texture image for a threedimensional texture is determined by equation 8.3, where k is log2 of the value of MAX_3D_TEXTURE_SIZE. OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

214

In a similar fashion, the maximum allowable width, or height for twodimensional texture types, of a texture image for a one- or two-dimensional, oneor 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 for one- and two-dimensional array textures (height or depth, respectively), or the maximum number of layer-faces for cube map array textures (depth), must be at least the value of MAX_ARRAY_TEXTURE_LAYERS for all levels. The maximum allowable width or height of a rectangle texture image 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. 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) 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 ws , hs , or ds are negative. An INVALID_VALUE error is generated by TexImage3D if target is TEXTURE_CUBE_MAP_ARRAY or PROXY_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 by TexImage2D if target is one of the cube map face targets from table 8.19, and width and height are not equal.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

215

An INVALID_VALUE error is generated by TexImage2D if target is TEXTURE_RECTANGLE and level is non-zero. 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 values described above. An INVALID_OPERATION error is generated if internalformat is one of the formats in table 8.14 and the “Copyable” field in the table is not . An INVALID_OPERATION error is generated if the internal format is integer and format is not one of the integer formats listed in table 8.3, or if the internal format is not integer and format is an integer format. An INVALID_OPERATION error is generated by TexImage3D if internalformat is one of the EAC, ETC2, or RGTC compressed formats and either border is non-zero, or target is not TEXTURE_2D_ARRAY. An INVALID_OPERATION error is generated by TexImage2D if internalformat is one of the EAC, ETC2, or RGTC compressed formats and either border is non-zero, or target is not TEXTURE_2D or one of the cube map face targets from table 8.19. An INVALID_ENUM error is generated by CompressedTexImage1D if internalformat is one of the specific compressed formats. OpenGL defines no specific one-dimensional compressed formats, but such formats may be provided by extensions. An INVALID_OPERATION error is generated if one of the base internal format and format is DEPTH_COMPONENT or DEPTH_STENCIL, and the other is neither of these values. An INVALID_OPERATION error is generated if either the base internal format or format is STENCIL_INDEX and the other is not also STENCIL_INDEX. 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, TEXTURE_1D_ARRAY for a onedimensional array texture, TEXTURE_RECTANGLE for a rectangle texture, or one OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

216

of the cube map face targets from table 8.19 for a cube map texture. Additionally, target may be either PROXY_TEXTURE_2D for a two-dimensional proxy texture, PROXY_TEXTURE_1D_ARRAY for a one-dimensional proxy array texture, PROXY_TEXTURE_RECTANGLE for a rectangle proxy texture, or PROXY_TEXTURE_CUBE_MAP for a cube map proxy texture in the special case discussed in section 8.22. 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 UNPACK_SKIP_IMAGES is ignored. A two-dimensional or rectangle 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 8.19 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. Finally, the command void TexImage1D( enum target, int level, int internalformat, sizei width, int border, enum format, enum type, const void *data ); is used to specify a one-dimensional texture image. target must be either TEXTURE_1D, or PROXY_TEXTURE_1D in the special case discussed in section 8.22. For the purposes of decoding the texture image, TexImage1D is equivalent to calling TexImage2D with corresponding arguments and height of 1. 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 ws , hs , and ds as defined in section 8.5.3. A two-dimensional or rectangle texture image has depth ds = 1, with height hs and width ws as above. A one-dimensional texture image has depth ds = 1, height hs = 1, and width ws as above. An element (i, j, k) of the texture image is called a texel (for a two-dimensional texture or one-dimensional array texture, k is irrelevant; for a one-dimensional texture, j and k are both irrelevant). The texture value used in texturing a fragment OpenGL 4.6 (Core Profile) - May 5, 2022

8.5. TEXTURE IMAGE SPECIFICATION

217

5.0 4 1.0 3 α

2 t

1 0 0.0 −1 −1.0 −1

−1.0

u 0.0

9.0 1.0

Figure 8.3. A texture image and the coordinates used to access it. This is a twodimensional texture with width 8 and height 4. A one-dimensional texture would consist of a single horizontal strip. α and β, values used in blending adjacent texels to obtain a texture value are also shown.

is determined by sampling the texture in a shader, but may not correspond to any actual texel. See figure 8.3. 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 TexImage1D, TexImage2D, or TexImage3D is NULL, and the pixel unpack buffer object is zero, a one-, 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 treated normally to refer to the beginning of the pixel unpack buffer object’s data.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

8.6

218

Alternate Texture Image Specification Commands

Two-dimensional and one-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, 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, TEXTURE_1D_ARRAY, TEXTURE_RECTANGLE, or one of the cube map face targets from table 8.19. x, y, width, and height correspond precisely to the corresponding arguments to ReadPixels (refer to section 18.2); 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 framebuffer exactly as if these arguments were passed to CopyPixels (see section 18.3) with argument type set to COLOR, DEPTH, DEPTH_STENCIL, or STENCIL_INDEX, depending on internalformat, stopping after conversion of depth values. RGBA data is taken from the current color buffer, while depth component and stencil index data are taken from the depth and stencil buffers, respectively. Subsequent processing is identical to that described for TexImage2D, beginning with clamping of the R, G, B, A, or depth values, and masking of the stencil index values from the resulting pixel groups. Parameters level, internalformat, and border are specified using the same values, with the same meanings, as the corresponding arguments of TexImage2D. The constraints on width, height, and border are exactly those for the corresponding arguments of TexImage2D. Errors An INVALID_ENUM error is generated if target is not TEXTURE_2D, TEXTURE_1D_ARRAY, TEXTURE_RECTANGLE, or one of the cube map face targets from table 8.19. An INVALID_ENUM error is generated if an invalid value is specified for internalformat. An INVALID_OPERATION error is generated if internalformat is one of

OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

219

the formats in table 8.14 and the “Copyable” field in the table is not . An INVALID_VALUE error is generated if the target parameter to CopyTexImage2D 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 depth component data is required and no depth buffer is present • if stencil index data is required and no stencil buffer is present • if integer RGBA data is required and the format of the current color buffer is not integer • if floating- or fixed-point RGBA data is required and the format of the current color buffer is integer • if the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the read buffer (see section 18.2.1) is LINEAR (see section 9.2.3) and internalformat is one of the sRGB formats in table 8.24 • 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 8.24. An INVALID_VALUE error is generated if width or height is negative. An INVALID_FRAMEBUFFER_OPERATION error is generated if the object bound to READ_FRAMEBUFFER_BINDING (see section 9) is not framebuffer complete (as defined in section 9.4.2). An INVALID_OPERATION error is generated if the object bound to READ_FRAMEBUFFER_BINDING is framebuffer complete and its effective value of SAMPLE_BUFFERS (see section 9.2.3.1) is one. The command void CopyTexImage1D( enum target, int level, enum internalformat, int x, int y, sizei width, int border ); defines a one-dimensional texture image in exactly the manner of TexImage1D, except that the image data are taken from the framebuffer, rather than from client memory. Currently, target must be TEXTURE_1D. For the purposes of decoding OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS the texture image, CopyTexImage1D is equivalent to calling CopyTexImage2D with corresponding arguments and height of 1, except that the height of the image is always 1, regardless of the value of border. level, internalformat and border are specified using the same values, with the same meanings, as the corresponding arguments of TexImage1D. The constraints on width and border are exactly those of the corresponding arguments of TexImage1D. 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, 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 TexSubImage1D( enum target, int level, int xoffset, sizei width, 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 ); void CopyTexSubImage1D( enum target, int level, int xoffset, int x, int y, sizei width ); void TextureSubImage3D( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, enum type, const void *pixels ); void TextureSubImage2D( uint texture, int level, int xoffset, int yoffset, sizei width, sizei height, enum format, enum type, const void *pixels ); void TextureSubImage1D( uint texture, int level, int xoffset, sizei width, enum format, enum type, const void *pixels ); void CopyTextureSubImage3D( uint texture, int level, int xoffset, int yoffset, int zoffset, int x, int y, sizei width, sizei height ); OpenGL 4.6 (Core Profile) - May 5, 2022

220

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS Command Names TexSubImage1D, CopyTexSubImage1D, TextureSubImage1D, CopyTextureSubImage1D TexSubImage2D, CopyTexSubImage2D,

TextureSubImage2D, CopyTextureSubImage2D TexSubImage3D, CopyTexSubImage3D,

TextureSubImage3D, CopyTextureSubImage3D

221

Valid targets or effective texture targets TEXTURE_1D

TEXTURE_2D, TEXTURE_1D_ARRAY, TEXTURE_RECTANGLE or one of the

cube map face targets from table 8.19 TEXTURE_2D, TEXTURE_1D_ARRAY or TEXTURE_RECTANGLE TEXTURE_3D, TEXTURE_2D_ARRAY or TEXTURE_CUBE_MAP_ARRAY TEXTURE_3D, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or TEXTURE_CUBE_MAP

Table 8.15: Valid texture target parameters or effective texture targets for texture subimage commands.

void CopyTextureSubImage2D( uint texture, int level, int xoffset, int yoffset, int x, int y, sizei width, sizei height ); void CopyTextureSubImage1D( uint texture, int level, int xoffset, int x, int y, sizei width ); For *TexSubImage*, the texture object is that bound to target, For *TextureSubImage*, texture is the name of the texture object. target or the effective target of texture (the value of TEXTURE_TARGET; see section 8.11.2) must match each command as shown in table 8.15. 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 level parameter of each command specifies the level of the texture image that is modified. OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

222

TexSubImage*D and TextureSubImage*D arguments width, height, depth, format, type, and data match the corresponding arguments to the corresponding3 TexImage*D command (where those arguments exist), meaning that they accept the same values, and have the same meanings. The exception is that a NULL data pointer does not represent unspecified image contents. CopyTex*SubImage3D and CopyTex*SubImage2D arguments x, y, width, and height match the corresponding arguments to CopyTexImage2D. CopyTex*SubImage1D arguments x, y, and width match the corresponding arguments to CopyTexImage1D. Each of these commands interprets and processes pixel groups in exactly the manner of its TexImage counterpart, except that the assignment of R, G, B, A, depth, 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 format argument of these commands and the internalformat of the texture image being respecified as apply to the format and internalformat arguments of their TexImage counterparts. Arguments xoffset, yoffset, and zoffset of Tex*SubImage3D and CopyTex*SubImage3D specify the lower left back texel coordinates of a width-wide by height-high 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 layer-faces to update. For TextureSubImage3D and CopyTextureSubImage3D only, texture may be a cube map texture. In this case, zoffset is interpreted as specifying the cube map face for the corresponding layer in table 9.3 and depth is the number of successive faces to update. The depth argument associated with CopyTex*SubImage3D 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 these commands. The subregion must lie within the bounds of the texture image. If ws , hs , and ds are the specified width, height, and depth of the texture image (as described in section 8.5.3); x, y, and z are the specified xoffset, yoffset, and zoffset values; and w, h, and d are the specified width, height, and depth values; then it is an error if any of the relationships in equations 8.4-8.6 are satisfied.

x < 0,

x + w > ws

(8.4)

y < 0,

y + h > hs

(8.5)

z < 0,

z + d > ds

(8.6)

For example, both TexSubImage2D and TextureSubImage2D correspond to TexImage2D for purposes of this paragraph.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

223

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) j n k j=y+ mod h w n mod d k=z+ w×h Arguments xoffset and yoffset of Tex*SubImage2D and CopyTex*SubImage2D specify the lower left texel coordinates of a width-wide by height-high rectangular subregion of the texture image. The subregion must lie within the bounds of the texture image, as described above for TexSubImage3D. It is an error if any of the relationships in equations 8.4-8.5 are satisfied. Counting from zero, the nth pixel group is assigned to the texel with internal integer coordinates [i, j], where i = x + (n mod w) jnk j =y+( mod h) w The xoffset argument of Tex*SubImage1D and CopyTex*SubImage1D specifies the left texel coordinate of a width-wide subregion of the texture image. The subregion must lie within the bounds of the texture image, as described above for TexSubImage3D. It is an error if any of the relationships in equation 8.4 are satisfied. Counting from zero, the nth pixel group is assigned to the texel with internal integer coordinates [i], where i = x + (n mod w) Texture images with compressed internal formats may be stored in such a way that it is not possible to modify an image with subimage commands without having to decompress and recompress the texture image. Even if the image were modified in this manner, it may not be possible to preserve the contents of some of the texels outside the region being modified. To avoid these complications, the GL does not support arbitrary modifications to texture images with compressed internal formats. Calling any of the Tex*SubImage* or CopyTex*SubImage* commands will generate an INVALID_OPERATION error if xoffset, yoffset, or zoffset is not equal to zero. In addition, the contents of any texel outside the region modified OpenGL 4.6 (Core Profile) - May 5, 2022

8.6. ALTERNATE TEXTURE IMAGE SPECIFICATION COMMANDS

224

by such a call are undefined. These restrictions may be relaxed for specific compressed internal formats whose images are easily modified. If the internal format of the texture image being modified is one of the specific compressed formats described in table 8.14, the texture is stored using the corresponding compressed texture image encoding (see appendix D). Since such images are easily edited along 4 × 4 texel boundaries, the limitations on subimage location and size are relaxed for Tex*SubImage2D, Tex*SubImage3D, CopyTex*SubImage2D, and CopyTex*SubImage3D. These commands will generate an INVALID_OPERATION error if one of the following conditions occurs: • width is not a multiple of four, width + xoffset is not equal to the value of TEXTURE_WIDTH, and either xoffset or yoffset is non-zero. • height is not a multiple of four, height + yoffset is not equal to the value of TEXTURE_HEIGHT, and either xoffset or yoffset is non-zero. • xoffset or yoffset is not a multiple of four. The contents of any 4 × 4 block of texels of such a compressed texture image that does not intersect the area being modified are preserved during valid Tex*SubImage* and Copy*TexSubImage* calls. Errors An INVALID_ENUM error is generated by *TexSubImage* if target does not match the command, as shown in table 8.15. An INVALID_OPERATION error is generated by *TextureSubImage* if texture is not the name of an existing texture object. An INVALID_OPERATION error is generated by *TextureSubImage* if the effective target of texture does not match the command, as shown in table 8.15. An INVALID_OPERATION error is generated by: • *TextureSubImage3D if the effective target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY and the texture object is not cube complete (for TEXTURE_CUBE_MAP or cube array complete (for TEXTURE_CUBE_MAP_ARRAY). An INVALID_VALUE error is generated if width, height, or depth (if each argument is present) is negative.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES An INVALID_VALUE error is generated if level is negative or greater than the log2 of the maximum texture width, height, or depth. An INVALID_VALUE error is generated if the effective target is TEXTURE_RECTANGLE and level is not zero. An INVALID_VALUE error is generated if the specified subregion does not lie within the bounds of the texture image, as described above. An INVALID_FRAMEBUFFER_OPERATION error is generated by CopyTexImage*D, CopyTexSubImage*D and CopyTextureSubImage*D if the object bound to READ_FRAMEBUFFER_BINDING is not framebuffer complete (see section 9.4.2) An INVALID_OPERATION error is generated by CopyTexImage*D, CopyTexSubImage*D and CopyTextureSubImage*D if • the read buffer is NONE, 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. An INVALID_OPERATION error is generated by *TexSubImage2D and *TextureSubImage2D if internalformat is one of the EAC, ETC2, or RGTC formats and the effective target is TEXTURE_RECTANGLE.

8.6.1

Texture Copying Feedback Loops

Calling any of the CopyTex*SubImage* or CopyTexImage* commands will result in undefined behavior if the destination texture image level is also bound to to the selected read buffer (see section 18.2.1) 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 D as well as any additional formats defined by extensions. The commands

OpenGL 4.6 (Core Profile) - May 5, 2022

225

8.7. COMPRESSED TEXTURE IMAGES

226

void CompressedTexImage1D( enum target, int level, enum internalformat, sizei width, int border, sizei imageSize, const void *data ); 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 one-, two-, and three-dimensional texture images, respectively, with incoming data stored in a specific compressed image format. The target, level, internalformat, width, height, depth, and border parameters have the same meaning as in TexImage1D, TexImage2D, and TexImage3D, except that compressed rectangle texture formats are not supported. 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. If the compressed image is not encoded according to the defined image format, the results of the call are undefined. If the compressed data are arranged into fixed-size blocks of texels, the pixel storage modes can be used to select a sub-rectangle from a larger containing rectangle. These pixel storage modes operate in the same way as they do for TexImage*D and as described in section 8.4.4. In the remainder of this section, denote by blocksize, bw , bh , and bd the values of pixel storage modes UNPACK_COMPRESSED_BLOCK_SIZE, UNPACK_COMPRESSED_BLOCK_WIDTH, UNPACK_COMPRESSED_BLOCK_HEIGHT, and UNPACK_COMPRESSED_BLOCK_DEPTH respectively. blocksize is the compressed block size in bytes; bw , bh , and bd are the compressed block width, height, and depth in pixels. By default the pixel storage modes UNPACK_ROW_LENGTH, UNPACK_SKIP_ROWS, UNPACK_SKIP_PIXELS, UNPACK_IMAGE_HEIGHT and UNPACK_SKIP_IMAGES are ignored for compressed images. To enable UNPACK_SKIP_PIXELS and UNPACK_ROW_LENGTH, blocksize and bw must both be non-zero. To also OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

227

enable UNPACK_SKIP_ROWS and UNPACK_IMAGE_HEIGHT, bh must be non-zero. And to also enable UNPACK_SKIP_IMAGES, bd must be non-zero. All parameters must be consistent with the compressed format to produce the desired results. Errors An INVALID_ENUM error is generated if the target parameter to any of the CompressedTexImagenD commands is TEXTURE_RECTANGLE or PROXY_TEXTURE_RECTANGLE. An INVALID_ENUM error is generated if internalformat is not a supported specific compressed internal format from table 8.14. In particular, this error will be generated for any of the generic compressed internal formats. 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 of the following conditions is violated when selecting a sub-rectangle from a compressed image: • the value of UNPACK_SKIP_PIXELS must be a multiple of bw ; • the value of UNPACK_SKIP_ROWS must be a multiple of bh for CompressedTexImage2D and CompressedTexImage3D; • the value of UNPACK_SKIP_IMAGES must be a multiple of bd for CompressedTexImage3D. An INVALID_VALUE error is generated if imageSize does not match the following requirements when pixel storage modes are active: • For CompressedTexImage1D the imageSize parameter must be equal to

width blocksize × bw

• For CompressedTexImage2D the imageSize parameter must be equal to

OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

228

width height blocksize × × bw bh • For CompressedTexImage3D the imageSize parameter must be equal to height depth width × × blocksize × bw bh bd

Based on the definition of unpacking from section 8.4.4 for uncompressed images, unpacking compressed images can be defined where: • n, the number of elements in a group, is 1. • s, the size of an element, is blocksize. • l, the number of groups in a row, is l m  row length , row length > 0 bwm l= l  width , otherwise bw where row length is the value of UNPACK_ROW_LENGTH. • a, the value of UNPACK_ALIGNMENT, is ignored. • k = n × l as is defined for uncompressed images. Before obtaining the first compressed image block from memory, the data pointer is advanced by UNPACK SKIP PIXELS UNPACK SKIP ROWS ×k ×n+ bw bh l m elements. Then width blocks are obtained from contiguous blocks in memory bw (withoutm advancing l l the mpointer), after which the pointer is advanced by k elements. height sets of width blocks are obtained this way. For three-dimensional combh bw IMAGES pressed images the pointer is advanced by UNPACK SKIP times the number of bd elements in one two-dimensional image before obtaining the first group from memory. Then after height rows are obtained, is posil if UNPACK_IMAGE_HEIGHT m UNPACK IMAGE HEIGHT−height tive, the pointer skips over the remaining rows, before bh starting the next two-dimensional image.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

229

Specific compressed internal formats may impose additional format-specific restrictions. For example, a format might be supported only for two-dimensional textures, or might not allow non-zero border values. Any such restrictions will be documented in the extension specification defining the compressed internal format. Any restrictions imposed by specific compressed internal formats will be invariant, meaning that if the GL accepts and stores a texture image in compressed form, providing the same image to CompressedTexImage1D, CompressedTexImage2D, or CompressedTexImage3D will not generate an error if the following restrictions are satisfied: • data points to a compressed texture image returned by GetCompressedTexImage (section 8.11). • target, level, and internalformat match the target, level and format parameters provided to the GetCompressedTexImage call returning data. • width, height, depth, internalformat, and imageSize match the values of TEXTURE_WIDTH, TEXTURE_HEIGHT, TEXTURE_DEPTH, TEXTURE_INTERNAL_FORMAT, and TEXTURE_COMPRESSED_IMAGE_SIZE for image level level in effect at the time of the GetCompressedTexImage call returning data. This guarantee applies not just to images returned by GetCompressedTexImage, but also to any other properly encoded compressed texture image of the same size and format. If internalformat is one of the specific compressed formats described in table 8.14, the compressed image data is stored using the corresponding texture image encoding (see appendix D). The corresponding compression algorithms support only two-dimensional images without borders, though three-dimensional images can be compressed as multiple slices of compressed two-dimensional BPTC images. Errors 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. An INVALID_ENUM error is generated by CompressedTexImage1D if internalformat is one of the specific compressed formats. OpenGL defines no specific one-dimensional compressed formats, but such formats may be provided by extensions.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

230

An INVALID_OPERATION error is generated by CompressedTexImage2D if internalformat is one of the EAC, ETC2, or RGTC formats and border is non-zero. An INVALID_OPERATION error is generated by CompressedTexImage3D if internalformat is one of the EAC, ETC2, or RGTC formats and either border is non-zero, or target is not TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY. An INVALID_OPERATION error is generated by CompressedTexImage2D and CompressedTexImage3D if internalformat is one of the BPTC formats and border is non-zero. If the data argument of CompressedTexImage1D, 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 TexImage1D, 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 CompressedTexSubImage1D( enum target, int level, int xoffset, sizei width, enum format, sizei imageSize, const void *data ); 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 ); void CompressedTextureSubImage1D( uint texture, int level, int xoffset, sizei width, enum format, sizei imageSize, const void *data ); void CompressedTextureSubImage2D( uint texture, int level, int xoffset, int yoffset, sizei width, sizei height, enum format, sizei imageSize, const void *data ); void CompressedTextureSubImage3D( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, sizei imageSize, const void *data );

OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

231

The target, texture, level, xoffset, yoffset, zoffset, width, height, and depth parameters have the same meaning as in the corresponding commands from section 8.6 without the Compressed prefix (where those parameters are present). 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 CompressedTexImage1D, CompressedTexImage2D, and CompressedTexImage3D. Any restrictions imposed by specific compressed internal formats will be invariant, meaning that if the GL accepts and stores a texture image in compressed form, providing the same image to CompressedTex*Image* will not generate an error if the following restrictions are satisfied: • data points to a compressed texture image returned by GetCompressedTexImage (section 8.11). • target, level, and format match the target, level and format parameters provided to the GetCompressedTexImage call returning data. • width, height, depth, format, and imageSize match the values of TEXTURE_WIDTH, TEXTURE_HEIGHT, TEXTURE_DEPTH, TEXTURE_INTERNAL_FORMAT, and TEXTURE_COMPRESSED_IMAGE_SIZE for image level level in effect at the time of the GetCompressedTexImage call returning data. • width, height, depth, and format match the values of TEXTURE_WIDTH, TEXTURE_HEIGHT, TEXTURE_DEPTH, and TEXTURE_INTERNAL_FORMAT currently in effect for image level level. • xoffset, yoffset, and zoffset are all zero. This guarantee applies not just to images returned by GetCompressedTexImage, but also to any other properly encoded compressed texture image of the same size. If the internal format of the image being modified is one of the specific compressed formats described in table 8.14, the texture is stored using the corresponding texture image encoding (see appendix D). Since these specific compressed formats are easily edited along 4 × 4 texel boundaries, the limitations on subimage location and size are relaxed for CompressedTex*SubImage2D and CompressedTex*SubImage3D. The contents of any 4 × 4 block of texels of a compressed texture image in these specific compressed formats that do not intersect the area being modified are preserved during CompressedTex*SubImage* calls. OpenGL 4.6 (Core Profile) - May 5, 2022

8.7. COMPRESSED TEXTURE IMAGES

232

Errors An INVALID_ENUM error is generated by CompressedTexSubImage*D if target is TEXTURE_RECTANGLE or PROXY_TEXTURE_RECTANGLE. An INVALID_OPERATION error is generated by CompressedTextureSubImage*D if texture is not the name of an existing texture object. An INVALID_OPERATION error is generated by CompressedTextureSubImage*D if the effective target is TEXTURE_RECTANGLE. An INVALID_ENUM error is generated if format is one of the generic compressed internal formats. 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 CompressedTex*Image commands. 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 corresponding 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_ENUM error is generated by CompressedTex*SubImage1D if the internal format of the texture is one of the specific compressed formats. An INVALID_OPERATION error is generated by CompressedTex*SubImage2D if the internal format of the texture is one of the EAC, ETC2, or RGTC formats and border is non-zero. An INVALID_OPERATION error is generated by CompressedTex*SubImage3D if the internal format of the texture is one of the EAC, ETC2, or RGTC formats and either border is non-zero, or the effective target for the texture is not TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY. An INVALID_OPERATION error is generated if the internal format of the texture is one of the BPTC formats and border is non-zero.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.8. MULTISAMPLE TEXTURES

233

An INVALID_OPERATION error is generated if any of the following conditions occurs: • width is not a multiple of four, and width + xoffset is not equal to the value of TEXTURE_WIDTH. • height is not a multiple of four, and height + yoffset is not equal to the value of TEXTURE_HEIGHT. • xoffset or yoffset is not a multiple of four.

8.8

Multisample Textures

In addition to the texture types described in previous sections, two additional types of textures are supported. A multisample texture is similar to a two-dimensional or two-dimensional array texture, except it contains multiple samples per texel. Multisample textures do not have multiple image levels. The commands void TexImage2DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, boolean fixedsamplelocations ); void TexImage3DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, sizei depth, boolean fixedsamplelocations ); establish the data storage, format, dimensions, and number of samples of a multisample texture’s image. For TexImage2DMultisample, target must be TEXTURE_2D_MULTISAMPLE or PROXY_TEXTURE_2D_MULTISAMPLE and for TexImage3DMultisample target must be TEXTURE_2D_MULTISAMPLE_ARRAY or PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY. width and height are the dimensions in texels of the texture. 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 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 internal format or size of the image. OpenGL 4.6 (Core Profile) - May 5, 2022

8.9. BUFFER TEXTURES

234

Upon success, TexImage*Multisample deletes any existing image for target and the contents of texels are undefined. 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 14.3.1 determining 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_VALUE error is generated if width, height, or depth is negative. 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 TexImage3DMultisample 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_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 22.3). 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 one-, two-, and three-dimensional, one- and two-dimensional array, and cube map textures described in previous sections, one additional type of texture is supported. A buffer texture is similar to a one-dimensional texture. However,

OpenGL 4.6 (Core Profile) - May 5, 2022

8.9. BUFFER TEXTURES

235

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 commands void TexBufferRange( enum target, enum internalformat, uint buffer, intptr offset, sizeiptr size ); void TextureBufferRange( uint texture, enum internalformat, uint buffer, intptr offset, sizeiptr size ); attach 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 a buffer texture. For TexBufferRange, the buffer texture is that currently bound to target. For TextureBufferRange, texture is the name of the buffer texture. target or the effective target of texture 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 8.16. Errors An INVALID_OPERATION error is generated by TextureBufferRange if texture is not the name of an existing texture object. An INVALID_ENUM error is generated by TexBufferRange if target is not TEXTURE_BUFFER. An INVALID_OPERATION error is generated by TextureBufferRange if the effective target is not TEXTURE_BUFFER. An INVALID_ENUM error is generated if internalformat is not one of the sized internal formats in table 8.16. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

8.9. BUFFER TEXTURES

236

of the value of TEXTURE_BUFFER_OFFSET_ALIGNMENT. The commands void TexBuffer( enum target, enum internalformat, uint buffer ); void TextureBuffer( uint texture, enum internalformat, uint buffer ); are respectively equivalent to TexBufferRange(target, internalf ormat, buf f er, 0, size); and TextureBufferRange(texture, 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 8.16. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.9. BUFFER TEXTURES

237

components of the texture are then converted to a (R, G, B, A) vector according to table 8.16, 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 8.16. Sized Internal Format

Base Type

Components

R8 R16 R16F R32F R8I R16I R32I R8UI R16UI R32UI RG8 RG16 RG16F RG32F RG8I RG16I RG32I RG8UI RG16UI RG32UI RGB32F RGB32I RGB32UI RGBA8 RGBA16 RGBA16F RGBA32F RGBA8I RGBA16I

ubyte 1 ushort 1 half 1 float 1 byte 1 short 1 int 1 ubyte 1 ushort 1 uint 1 ubyte 2 ushort 2 half 2 float 2 byte 2 short 2 int 2 ubyte 2 ushort 2 uint 2 float 3 int 3 uint 3 ubyte 4 ushort 4 half 4 float 4 byte 4 short 4 (Continued on next page)

Norm Yes Yes No No No No No No No No Yes Yes No No No No No No No No No No No Yes Yes No No No No

OpenGL 4.6 (Core Profile) - May 5, 2022

Component 0 1 2 3 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R 0 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G 0 1 R G B 1 R G B 1 R G B 1 R G B A R G B A R G B A R G B A R G B A R G B A

8.10. TEXTURE PARAMETERS

238

Internal formats for buffer textures (continued) Sized Internal Format Base Type Components Norm Component 0 1 2 3 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 8.16: 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.

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( enum target, enum pname, const T *params ); void TextureParameter{if}( uint texture, enum pname, T param ); void TextureParameter{if}v( uint texture, enum pname, const T *params ); OpenGL 4.6 (Core Profile) - May 5, 2022

8.10. TEXTURE PARAMETERS

239

void TextureParameterI{i ui}v( uint texture, enum pname, const T *params ); For TexParameter*, the texture object is that bound to target. For TextureParameter*, texture is the name of the texture object. target or the effective target of texture must be one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY. TEXTURE_RECTANGLE, 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 8.17. 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. If pname is TEXTURE_SWIZZLE_RGBA, params is an array of four enums which respectively set the TEXTURE_SWIZZLE_R, TEXTURE_SWIZZLE_G, TEXTURE_SWIZZLE_B, and TEXTURE_SWIZZLE_A parameters simultaneously. Name DEPTH_STENCIL_TEXTURE_MODE

Type enum

TEXTURE_COMPARE_MODE

int 4 floats, ints, or uints enum

TEXTURE_COMPARE_FUNC

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

Texture parameters continued on next page OpenGL 4.6 (Core Profile) - May 5, 2022

8.10. TEXTURE PARAMETERS

240

Texture parameters continued from previous page Name Type Legal Values TEXTURE_LOD_BIAS float any value TEXTURE_MAG_FILTER enum NEAREST, LINEAR TEXTURE_MAX_ANISOTROPY float greater or equal to 1.0 TEXTURE_MAX_LEVEL int any non-negative integer TEXTURE_MAX_LOD float any value TEXTURE_MIN_FILTER enum 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_SWIZZLE_RGBA

4 enums

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 RED, GREEN, BLUE, ALPHA, ZERO, ONE CLAMP_TO_EDGE, REPEAT, CLAMP_TO_BORDER, MIRRORED_REPEAT, MIRROR_CLAMP_TO_EDGE CLAMP_TO_EDGE, REPEAT, CLAMP_TO_BORDER, MIRRORED_REPEAT, MIRROR_CLAMP_TO_EDGE CLAMP_TO_EDGE, REPEAT, CLAMP_TO_BORDER, MIRRORED_REPEAT, MIRROR_CLAMP_TO_EDGE

Table 8.17: Texture parameters and their values.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.10. TEXTURE PARAMETERS

241

In the remainder of chapter 8, denote by lodmin , lodmax , levelbase , and levelmax the values of the texture parameters TEXTURE_MIN_LOD, TEXTURE_MAX_LOD, TEXTURE_BASE_LEVEL, and TEXTURE_MAX_LEVEL respectively. If the texture was created with TextureView, then the TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL parameters are interpreted relative to the view and not relative to the original data store. 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. Errors An INVALID_ENUM error is generated by TexParameter* if target is not one of the valid targets listed above. An INVALID_OPERATION error is generated by TextureParameter* if the effective 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 8.17. An INVALID_OPERATION error is generated by TextureParameter* if texture is not the name of an existing texture object. An INVALID_ENUM error is generated if the type of the parameter specified by pname is enum, and any value specified by param or params is not among the legal values shown in table 8.17. An INVALID_VALUE error is generated if pname is TEXTURE_BASE_LEVEL or TEXTURE_MAX_LEVEL, and any value specified by param or params is negative. An INVALID_VALUE error is generated if pname is TEXTURE_MAX_ANISOTROPY, and any value specified by param or params is less than 1.0. An INVALID_ENUM error is generated if Tex*Parameter{if} is called for a non-scalar parameter (pname TEXTURE_BORDER_COLOR or TEXTURE_SWIZZLE_RGBA). An INVALID_ENUM error is generated by TexParameter* if target is either TEXTURE_2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY, and pname is any sampler state from table 23.18. An INVALID_OPERATION error is generated by TextureParameter* if the effective target is either TEXTURE_2D_MULTISAMPLE or TEXTURE_2D_MULTISAMPLE_ARRAY, and pname is any sampler state from table 23.18. An INVALID_OPERATION error is generated if the effective target is TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY or

OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

242

TEXTURE_RECTANGLE, and pname TEXTURE_BASE_LEVEL is set to a value

other than zero. An INVALID_ENUM error is generated if the effective target is TEXTURE_RECTANGLE and either of pnames TEXTURE_WRAP_S or TEXTURE_WRAP_T is set to either MIRROR_CLAMP_TO_EDGE, MIRRORED_REPEAT or REPEAT. An INVALID_ENUM error is generated if the effective target is TEXTURE_RECTANGLE and pname TEXTURE_MIN_FILTER is set to a value other than NEAREST or LINEAR (no mipmap filtering is permitted).

8.11

Texture Queries

8.11.1

Active Texture

Queries of most texture state variables are qualified by the value of ACTIVE_TEXTURE to determine which server texture state vector is queried. Table 23.12 indicates those state variables which are qualified by ACTIVE_TEXTURE during state queries.

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 ); void GetTextureParameter{if}v( uint texture, enum pname, T *data ); void GetTextureParameterI{i ui}v( uint texture, enum pname, T *data ); For GetTexParameter*, the texture object is that bound to target. For GetTextureParameter*, texture is the name of the texture object. The value of texture parameter pname for the texture is returned in params. target or the effective target of texture must be one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_RECTANGLE, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_2D_MULTISAMPLE, or TEXTURE_2D_MULTISAMPLE_ARRAY, indicating the currently bound one-, two-, or three-dimensional, one- or two-dimensional array, rectangle, cube map, cube map array, two-dimensional multisample, or two-dimensional multisample array texture object. OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

243

pname must be one of IMAGE_FORMAT_COMPATIBILITY_TYPE, TEXTURE_IMMUTABLE_FORMAT, TEXTURE_IMMUTABLE_LEVELS, TEXTURE_TARGET, TEXTURE_VIEW_MIN_LEVEL, TEXTURE_VIEW_NUM_LEVELS, TEXTURE_VIEW_MIN_LAYER, TEXTURE_VIEW_NUM_LAYERS, or one of the symbolic values in table 8.17. Querying pname TEXTURE_BORDER_COLOR with GetTex*ParameterIiv or GetTex*ParameterIuiv 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. Querying pname TEXTURE_TARGET returns the effective target of the texture object. For GetTexParameter*, this is the target parameter. For GetTextureParameter*, it is the target to which the texture was initially bound when it was created, or the value of the target parameter to the call to CreateTextures which created the texture. Errors An INVALID_OPERATION error is generated by GetTextureParameter* if texture is not the name of an existing texture object. An INVALID_ENUM error is generated by GetTexParameter* if target is not one of the texture targets described above. An INVALID_OPERATION error is generated by GetTextureParameter* if the effective target is not one of the texture 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

Parameters of a specified level-of-detail of a texture object may be queried with the commands void GetTexLevelParameter{if}v( enum target, int level, enum pname, T *params ); void GetTextureLevelParameter{if}v( uint texture, int level, enum pname, T *params ); For GetTexLevelParameter*, the texture object is that bound to target. For GetTextureLevelParameter*, texture is the name of the texture object.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

244

The value of texture parameter pname for level-of-detail level of the texture is returned in params. pname must be one of the symbolic values in tables 23.1623.17. The effective target of the texture object must be one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_RECTANGLE, TEXTURE_BUFFER, TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY, PROXY_TEXTURE_1D, PROXY_TEXTURE_2D, PROXY_TEXTURE_3D, PROXY_TEXTURE_1D_ARRAY, PROXY_TEXTURE_2D_ARRAY, PROXY_TEXTURE_CUBE_MAP_ARRAY, PROXY_TEXTURE_RECTANGLE, PROXY_TEXTURE_CUBE_MAP, PROXY_TEXTURE_2D_MULTISAMPLE, or PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY, indicating the one-, two-, or three-dimensional texture, one- or two-dimensional array texture, cube map array texture, rectangle texture, buffer texture, twodimensional multisample texture, two-dimensional multisample array texture; or the one-, two-, three-dimensional, one- or two-dimensional array, cube map array, rectangle, cube map, two-dimensional multisample, or two-dimensional multisample array proxy state vector. For GetTexLevelParameter* only, target may also be one of the cube map face targets from table 8.19, indicating one of the six distinct two-dimensional images making up the cube map texture object. Note that TEXTURE_CUBE_MAP is not a valid target parameter for GetTexLevelParameter*. For GetTextureLevelParameter* only, texture may also be a cube map texture object. In this case the query is always performed for face zero (the TEXTURE_CUBE_MAP_POSITIVE_X face), since there is no way to specify another face. level determines which level-of-detail’s state is returned. The maximum value of level depends on the texture target: • For targets TEXTURE_CUBE_MAP and TEXTURE_CUBE_MAP_ARRAY, 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_RECTANGLE, 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 GetTex*LevelParameter*, the maximum value is log2 of the value of MAX_TEXTURE_SIZE. For texture images with uncompressed internal formats, queries of pname TEXTURE_RED_TYPE, TEXTURE_GREEN_TYPE, TEXTURE_BLUE_TYPE, OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

245

TEXTURE_ALPHA_TYPE, and TEXTURE_DEPTH_TYPE return the data type used 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 texture 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. Querying pname TEXTURE_COMPRESSED_IMAGE_SIZE returns the size (in ubytes) of the compressed texture image that would be returned by GetCompressedTexImage (section 8.11). target must be a compressed texture target. 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 tables 23.16- 23.17 are returned. 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. Errors An INVALID_OPERATION error is generated by GetTextureLevelParameter* if texture is not the name of an existing texture object. An INVALID_ENUM error is generated by GetTexLevelParameter* if target is not one of the targets described above as valid for the corresponding command. An INVALID_OPERATION error is generated by GetTextureLevelParameter* if the effective target is not one of the targets described above as valid for the corresponding command. An INVALID_ENUM error is generated if pname is not one of the symbolic

OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

246

values in tables 23.16- 23.17. An INVALID_VALUE error is generated if level is negative or larger than the maximum allowable level-of-detail for the effective texture target as described above. An INVALID_OPERATION error is generated if pname is TEXTURE_COMPRESSED_IMAGE_SIZE and the effective texture target is a proxy target, or has an uncompressed internal format.

8.11.4

Texture Image Queries

Texture images may be obtained from a texture object with the commands void GetTexImage( enum target, int level, enum format, enum type, void *pixels ); void GetTextureImage( uint texture, int level, enum format, enum type, sizei bufSize, void *pixels ); void GetnTexImage( enum target, int level, enum format, enum type, sizei bufSize, void *pixels ); For Get*TexImage, target specifies the target to which the texture object is bound. target must be one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or TEXTURE_RECTANGLE, indicating a one-, two- or three-dimensional, one- or two-dimensional array, cube map array or rectangle texture, respectively, or one of the targets from table 8.19, indicates the corresponding face of a cube map texture. For GetTextureImage, texture is the name of the texture object. In addition to the types of textures accepted by the Get*TexImage commands, GetTextureImage also accepts cube map texture objects (with effective target TEXTURE_CUBE_MAP). level is a level-of-detail number, format is a pixel format from table 8.3, and type is a pixel type from table 8.2. If present, bufSize is the size of the buffer to receive the retrieved pixel data. GetnTexImage and GetTextureImage do not write more than bufSize bytes into pixels. These commands obtain component groups from a texture image with the indicated level-of-detail. If format is a color format then the components are assigned among R, G, B, and A according to table 8.18, starting with the first group in the first row, and continuing by obtaining groups in order from each row and proceeding from the first row to the last, and from the first image to the last for three-dimensional textures. One- and two-dimensional array and cube map array OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

247

textures are treated as two-, three-, and three-dimensional images, respectively, where the layers are treated as rows or images. Cube map textures are treated as three-dimensional images with a depth of 6, where the cube map faces are ordered as image layers as shown in table 9.3. If format is DEPTH_COMPONENT, DEPTH_STENCIL, or STENCIL_INDEX, then each depth component and/or stencil index is assigned with the same ordering of rows and images. These groups are then packed and placed in client or pixel buffer object memory. If a pixel pack buffer is bound (as indicated by a non-zero value of PIXEL_PACK_BUFFER_BINDING), pixels is an offset into the pixel pack buffer; otherwise, pixels is a pointer to client memory. Pixel storage modes that are applicable to ReadPixels are applied, as described in table 18.1 and section 18.2.9. For three-dimensional, two-dimensional array, cube map array, and cube map textures pixel storage operations are applied as if the image were two-dimensional, except that the additional pixel storage state values PACK_IMAGE_HEIGHT and PACK_SKIP_IMAGES are applied. The correspondence of texels to memory locations is as defined for TexImage3D in section 8.5. The row length, number of rows, image depth, and number of images are determined by the size of the texture image (including any borders). Errors An INVALID_OPERATION error is generated by GetTextureImage if texture is not the name of an existing texture object. An INVALID_ENUM error is generated by GetTexImage and GetnTexImage if target is not one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_RECTANGLE, or one of the targets from table 8.19. An INVALID_OPERATION error is generated by GetTextureImage if the effective target is not one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY, TEXTURE_RECTANGLE, or TEXTURE_CUBE_MAP (for GetTextureImage only). An INVALID_OPERATION error is generated by GetTextureImage if the effective target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture object is not cube complete or cube array complete, respectively. An INVALID_VALUE error is generated if level is negative or larger than the maximum allowable level. An INVALID_VALUE error is generated if level is non-zero and the effec-

OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

248

tive target is TEXTURE_RECTANGLE. An INVALID_OPERATION error is generated if any of the following mismatches between format and the internal format of the texture image exist: • format is a color format (one of the formats in table 8.3 whose target is the color buffer) and the base internal format of the texture image is not a color format. • format is DEPTH_COMPONENT and the base internal format is not DEPTH_COMPONENT or DEPTH_STENCIL. • format is DEPTH_STENCIL and the base internal format is not DEPTH_STENCIL. • format is STENCIL_INDEX and the base internal format is not STENCIL_INDEX or DEPTH_STENCIL. • format is one of the integer formats in table 8.3 and the internal format of the texture image is not integer, or format is not one of the integer formats in table 8.3 and the internal format is integer. An INVALID_OPERATION error is generated if a pixel pack buffer object is bound and packing the texture image into the buffer’s memory would exceed the size of the buffer. An INVALID_OPERATION error is generated if a pixel pack buffer object is bound and pixels is not evenly divisible by the number of basic machine units needed to store in memory the GL data type corresponding to type (see table 8.2). An INVALID_VALUE error is generated if bufSize is negative. An INVALID_OPERATION error is generated by GetTextureImage and GetnTexImage if the buffer size required to store the requested data is greater than bufSize. Sub-regions of a texture image may be obtained from a texture object with the command void GetTextureSubImage( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, enum type, sizei bufSize, void *pixels ); texture is the name of the texture object, and must not be a buffer or multisample texture. The effective target is the value of TEXTURE_TARGET for texture. OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES Base Internal Format RED RG RGB RGBA

249 R Ri Ri Ri Ri

G 0 Gi Gi Gi

B 0 0 Bi Bi

A 1 1 1 Ai

Table 8.18: Texture return values. Ri , Gi , Bi , and Ai are components of the internal format that are assigned to pixel values R, G, B, and A. If a requested pixel value is not present in the internal format, the specified constant value is used. level, format, type and pixels have the same meaning as for GetTexImage. bufSize is the size of the buffer to receive the retrieved pixel data. For cube map textures, the behavior is as though GetTextureImage were called, but only texels from the requested cube map faces (selected by zoffset and depth, as described below) were returned. xoffset, yoffset and zoffset indicate the position of the subregion to return. width, height and depth indicate the size of the region to return. These parameters have the same meaning as for TexSubImage3D, though for one- and two-dimensional textures there are extra restrictions, described in the errors section below. For one-dimensional array textures, yoffset is interpreted as the first layer to access and height is the number of layers to access. For two-dimensional array textures, zoffset is interpreted as the first layer to access and depth is the number of layers to access. Cube map textures are treated as an array of six slices in the zdimension, where the value of zoffset is interpreted as specifying the cube map face for the corresponding layer in table 9.3 and depth is the number of faces to access. For cube map array textures, zoffset is the first layer-face to access, and depth is the number of layer-faces to access. Each layer-face is translated into an array layer and a cube map face as described for layer-face numbers in section 8.5.3. Component groups from the specified sub-region are packed and placed into memory as described for GetTextureImage, starting with the texel at (xoffset, yoffset, zoffset). Errors An INVALID_OPERATION error is generated if texture is not the name of an existing texture object. An INVALID_OPERATION error is generated if texture is the name of a buffer or multisample texture.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

250

An INVALID_OPERATION error is generated if the effective target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture object is not cube complete or cube array complete, respectively. An INVALID_VALUE error is generated if xoffset, yoffset or zoffset are negative. An INVALID_VALUE error is generated if width, height, or depth is negative. An INVALID_VALUE error is generated if xoffset + width is greater than the texture’s width, yoffset + height is greater than the texture’s height, or zoffset + depth is greater than the texture’s depth. An INVALID_VALUE error is generated if the effective target is TEXTURE_1D and either yoffset is not zero, or height is not one. An INVALID_VALUE error is generated if the effective target is TEXTURE_1D, TEXTURE_1D_ARRAY, TEXTURE_2D or TEXTURE_RECTANGLE, and either zoffset is not zero, or depth is not one. An INVALID_VALUE error is generated if bufSize is negative. An INVALID_OPERATION error is generated if the buffer size required to store the requested data is greater than bufSize. Texture images stored in compressed form may be obtained with the commands void GetCompressedTexImage( enum target, int level, void *pixels ); void GetCompressedTextureImage( uint texture, int level, sizei bufSize, void *pixels ); void GetnCompressedTexImage( enum target, int level, sizei bufSize, void *pixels ); For Get*CompressedTexImage, the texture object is that which is bound to target. For GetCompressedTextureImage, texture is the name of the texture object, and the effective target is the value of TEXTURE_TARGET for texture. target, level, bufSize, and pixels are interpreted in the same manner as the corresponding parameters of GetTexImage, GetTextureImage, and GetnTexImage. When called, GetCompressedTexImage writes n ubytes of compressed image data to the pixel pack buffer or client memory pointed to by pixels, while GetCompressedTextureImage and GetnCompressedTexImage write min(n, buf Size) ubytes. n is the value of TEXTURE_COMPRESSED_IMAGE_SIZE for the texture image The compressed image data is formatted according to the definition of the texture’s internal format. By default the pixel storage modes PACK_ROW_LENGTH, PACK_SKIP_ROWS, PACK_SKIP_PIXELS, PACK_IMAGE_HEIGHT and PACK_SKIP_IMAGES are igOpenGL 4.6 (Core Profile) - May 5, 2022

8.11. TEXTURE QUERIES

251

nored for compressed images. To enable PACK_SKIP_PIXELS and PACK_ROW_LENGTH, the values of PACK_COMPRESSED_BLOCK_SIZE and PACK_COMPRESSED_BLOCK_WIDTH must both be non-zero. To also enable PACK_SKIP_ROWS and PACK_IMAGE_HEIGHT, the value of PACK_COMPRESSED_BLOCK_HEIGHT must be non-zero. And to also enable PACK_SKIP_IMAGES, the value of PACK_COMPRESSED_BLOCK_DEPTH must be non-zero. All parameters must be consistent with the compressed format to produce the desired results. When the pixel storage modes are active, the correspondence of texels to memory locations is as defined for CompressedTexImage3D in section 8.7. Errors An INVALID_OPERATION error is generated by GetCompressedTextureImage if texture is not the name of an existing texture object. An INVALID_OPERATION error is generated by GetCompressedTextureImage if the effective target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture object is not cube complete or cube array complete, respectively. An INVALID_VALUE error is generated if level is negative, or greater than the maximum allowable level. An INVALID_OPERATION error is generated if the texture image is stored with an uncompressed internal format. An INVALID_OPERATION error is generated if a pixel pack buffer object is bound and packing the texture image into the buffer’s memory would exceed the size of the buffer. An INVALID_VALUE error is generated if bufSize is negative. An INVALID_OPERATION error is generated by GetCompressedTextureImage and GetnCompressedTexImage if the buffer size required to store the requested data is greater than bufSize. If the compressed data are arranged into fixed-size blocks of texels, the command void GetCompressedTextureSubImage( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, sizei bufSize, void *pixels ); can be used to obtain a sub-region of a compressed texture image instead of the whole image. texture is the name of the texture object, and must not be a buffer or multisample texture. The effective target is the value of TEXTURE_TARGET for OpenGL 4.6 (Core Profile) - May 5, 2022

8.12. DEPTH COMPONENT TEXTURES

252

texture. level and pixels have the same meaning as the corresponding arguments of CompressedTexSubImage3D. bufSize indicates the size of the buffer to receive the retrieved pixel data. For cube map textures, the behavior is as though GetCompressedTexImage were called once for each requested face (selected by zoffset and depth, as described below) with target corresponding to the requested texture cube map face as indicated by table 9.3. pixels is offset appropriately for each successive image. xoffset, yoffset and zoffset indicate the position of the subregion to return. width, height and depth indicate the size of the region to return. These arguments have the same meaning as for CompressedTexSubImage3D, though there are extra restrictions, described in the errors section below. The mapping between the xoffset, yoffset, zoffset, width, height, and depth parameters and the faces, layers, and layer-faces for cube map, array, and cube map array textures is the same as for GetTextureSubImage. The xoffset, yoffset, zoffset offsets and width, height and depth sizes must be multiples of the values of PACK_COMPRESSED_BLOCK_WIDTH, PACK_COMPRESSED_BLOCK_HEIGHT, and PACK_COMPRESSED_BLOCK_DEPTH respectively, unless an offset is zero and the corresponding size is the same as the texture size in that dimension. Pixel storage modes are treated as for GetCompressedTexSubImage. The texel at (xoffset, yoffset, zoffset) will be stored at the location indicated by pixels and the current pixel packing parameters. Errors In addition to the same errors generated by GetTextureSubImage with corresponding parameters: An INVALID_VALUE error is generated if xoffset, yoffset or zoffset is not a multiple of the compressed block width, height or depth respectively. An INVALID_VALUE error is generated if width, height or depth is not a multiple of the compressed block width, height or depth respectively, unless the offset is zero and the size equals the texture image size. An INVALID_OPERATION error is generated if the texture compression format is not based on fixed-size blocks.

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 8.23). The initial

OpenGL 4.6 (Core Profile) - 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

253 sc −rz rz rx rx rx −rx

tc −ry −ry rz −rz −ry −ry

ma rx rx ry ry rz rz

Table 8.19: Selection of cube map images based on major axis direction of texture coordinates.

state for depth and depth/stencil textures treats them as RED textures.

8.13

Cube Map Texture Selection

When a cube map texture is sampled, the s t r texture coordinates are treated as a direction vector rx ry rz emanating from the center of a cube. The q coordinate is ignored. 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 8.19 explains how the major axis direction maps to the two-dimensional image of a particular cube map target. Using the sc , tc , and madetermined by the major axis direction as specified in table 8.19, 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

Seamless cube map filtering is enabled or disabled by calling Enable or Disable with target TEXTURE_CUBE_MAP_SEAMLESS. When seamless cube map filtering is disabled, the new s t is used to find a texture value in the determined face’s two-dimensional image using the rules given OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

254

in sections 8.14 through 8.15. When seamless cube map filtering is enabled, the rules for texel selection in sections 8.14 through 8.15 are modified so that texture wrap modes are ignored. Instead, • If NEAREST filtering is done within a miplevel, always apply wrap mode CLAMP_TO_EDGE. • 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. The required state is one bit indicating whether seamless cube map filtering is enabled or disabled. Initially, it is disabled.

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. When the texture’s value of TEXTURE_MAX_ANISOTROPY is equal to 1.0, the GL uses an isotropic texture filtering approach as described in this section and section 8.15. However, when the texture’s value of TEXTURE_MAX_ANISOTROPY is greater than 1.0, the GL should use a texture filtering scheme that accounts for

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

255

a degree of anisotropy up to the smaller of the texture’s value of TEXTURE_MAX_ANISOTROPY or the implementation-defined value of MAX_TEXTURE_MAX_ANISOTROPY. The particular scheme for anisotropic texture filtering is implementation dependent. Additionally, implementations are free to consider the current texture minification and magnification modes to control the specifics of the anisotropic filtering scheme used. The anisotropic texture filtering scheme may only access mipmap levels if the minification filter is one that requires mipmaps. Additionally, when a minification filter is specified, the anisotropic texture filtering scheme may only access texture mipmap levels between the texture’s values for TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL, inclusive. Implementations should also respect the values of TEXTURE_MAX_LOD and TEXTURE_MIN_LOD to the extent the anisotropic texture filtering scheme permits this. The following describes one particular approach to implementing anisotropic texture filtering for the two-dimensional texturing case: Anisotropic texture filtering substantially changes section 8.14. Previously a single scale factor P was determined based on the pixel’s projection into texture space. Now two scale factors, ρx and ρy , are computed: q 2 ∂v 2 ρx = ∂u + ∂x ∂x q 2 ∂v 2 ρy = ∂u ∂y + ∂y ρmax = max(ρx , ρy ) ρminl= min(ρ m x , ρy ) , maxAniso) N = min( ρρmax min λ0 = log2 ( ρmax N )

where maxAniso is the smaller of the texture’s value of TEXTURE_MAX_ANISOTROPY or the implementation-defined value of MAX_TEXTURE_MAX_ANISOTROPY. It is acceptable for an implementation to round the sampling rate, N , up to the nearest supported sampling rate. For example, an implementation may only support power-of-two sampling rates. It is also acceptable for an implementation to approximate the ideal functions ρx and ρy with functions fx and fy subject to the following constraints: ∂v 1. fx is continuous and monotonically increasing in ∂u ∂x and ∂x . ∂v 2. fy is continuous and monotonically increasing in ∂u ∂y and ∂y .

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION ∂v 3. max( ∂u ∂x , ∂x ) ≤ fx ≤

√

256 ∂u ∂x

∂v + ∂x . √ ∂v ∂v 4. max( ∂u 2 ∂u ∂y , ∂y ) ≤ fy ≤ ∂y + ∂y . 2

Instead of a single sample, τ , at (u, v, λ), N locations in the mipmap at levelof-detail λ are sampled within the texture footprint of the pixel. This sum τaniso is defined using the single sample τ . When the texture’s value of TEXTURE_MAX_ANISOTROPHY is greater than 1.0, use τaniso instead of τ to determine the fragment’s texture value. ( τaniso =

1 i 1 i 1 PN i=1 τ (u(x − 2 + N +1 , y), v(x − 2 + N +1 , y)), N P N 1 1 i 1 i i=1 τ (u(x, y − 2 + N +1 ), v(x, y − 2 + N +1 )), N

Px > Py Py ≥ Px

It is acceptable to approximate the u and v functions with equally spaced samples in texture space at level-of-details λ: ( τaniso =

8.14.1

∂u i 1 ∂v i 1 1 PN i=1 τ (u(x, y) + ∂x ( N +1 − 2 ), v(x, y) + ∂x ( N +1 − 2 )), N P N 1 ∂u i 1 ∂v i 1 i=1 τ (u(x, y) + ∂y ( N +1 − 2 ), v(x, y) + ∂y ( N +1 − 2 )), N

Px > Py Py ≥ P x

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)]

(8.7)

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

(8.8)

  max ,  lod  0 λ, λ= lod  min ,   undef ined,

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

(8.9)

biastexobj is the value of TEXTURE_LOD_BIAS for the bound texture object (as described in section 8.10). 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

257

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. Different implementations have chosen to perform clamping on intermediate and final terms in computing λ0 differently. Care should be taken that intermediate terms do not exceed the implementation-dependent range as different results may otherwise occur. 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 ( s(x, y) + δu , rectangle texture u(x, y) = ws × s(x, y) + δu , otherwise ( t(x, y) + δv , rectangle texture v(x, y) = hs × t(x, y) + δv , otherwise

(8.10)

w(x, y) = ds × r(x, y) + δw where ws , hs , and ds are as defined in section 8.5.3 for the texture image whose level is levelbase . For a one-dimensional or one-dimensional array texture, define v(x, y) = 0 and w(x, y) = 0; for a two-dimensional, two-dimensional array, rectangle, cube map, or cube map array texture, define w(x, y) = 0. (δu , δv , δw ) are the texel offsets specified in the OpenGL 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. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

258

∂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). 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 ,

∂w ∂v ∂v ∂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

√

2 max( ∂u ∂x +

∂u ∂v ∂w + ∂w ∂x , ∂y + ∂y + ∂y )

For a polygon or point, ρ is given at a fragment with window coordinates (x, y) by

ρ = max

s  

∂u ∂x

∂v ∂x

 2 s 2 2 2  ∂u ∂w ∂v ∂w , + + ∂x ∂y ∂y ∂y 

(8.11) where ∂u/∂x indicates the derivative of u with respect to window x, and similarly for the other derivatives. For a line, the formula is s ρ=

∂u ∂u ∆x + ∆y ∂x ∂y

∂v ∂v ∆x + ∆y ∂x ∂y

∂w ∂w ∆x + ∆y ∂x ∂y

2 l,

(8.12) where ∆x = x2 − x1 and ∆y = y2 − y1 withp(x1 , y1 ) and (x2 , y2 ) being the segment’s window coordinate endpoints and l = ∆x2 + ∆y 2 . While it is generally agreed that equations 8.11 and 8.12 give the best results when texturing, they are often 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/∂x|, |∂v/∂y|, |∂w/∂x|, and |∂w/∂y| 2. Let mu = max

∂u ∂u , ∂x ∂y

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

259

∂v ∂v , ∂x ∂y

∂w ∂w , ∂x ∂y

mv = max

mw = max

Then max{mu , mv , mw } ≤ f (x, y) ≤ mu + mv + mw .

8.14.2

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 8.20. For a three-dimensional texture, the texel at location (i, j, k) becomes the texture value. For twodimensional, two-dimensional array, rectangle, or cube map textures, k is irrelevant, and the texel at location (i, j) becomes the texture value. For one-dimensional texture or one-dimensional array textures, j and k are irrelevant, and the texel at location i becomes the texture value. For one- and two-dimensional array textures, the texel is obtained from image layer l, where ( clamp(RN E(t), 0, hs − 1), for one-dimensional array textures 4 l= clamp(RN E(r), 0, ds − 1), for two-dimensional array textures 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 bvalue + 21 c.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION Wrap mode CLAMP_TO_EDGE CLAMP_TO_BORDER REPEAT MIRRORED_REPEAT MIRROR_CLAMP_TO_EDGE

260

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

Table 8.20: Texel location wrap mode application. mirror(a) returns a if a ≥ 0, and −(1 + a) otherwise. The values of mode and size are TEXTURE_WRAP_S and ws , TEXTURE_WRAP_T and hs , and TEXTURE_WRAP_R and ds 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 ≥ ws

j<0

j ≥ hs

k<0

k ≥ ds

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 8.11. 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. If border values are out-of-range with respect to the texture’s internal format, the result is undefined. 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 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

261

1 i0 = wrap(bu0 − c) 2 1 0 j0 = wrap(bv − c) 2 1 k0 = wrap(bw0 − c) 2 1 0 i1 = wrap(bu − c + 1) 2 1 j1 = wrap(bv 0 − c + 1) 2 1 0 k1 = wrap(bw − c + 1) 2 1 α = f rac(u0 − ) 2 1 0 β = f rac(v − ) 2 1 γ = f rac(w0 − ) 2 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

(8.13)

+ (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, rectangle, 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 1 l = clamp( r + , 0, ds − 1). 2

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

262

Figure 8.4. 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 R component of the corresponding texel.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

263

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 15.1 and then swizzled as described in section 15.2.1. 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 , τi1 j1 , τi1 j0 , and τi0 j0 (see figure 8.4). The selected component is identified 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. 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. And for a one-dimensional or one-dimensional array texture, τ = (1 − α)τi0 + ατi1 where τi is the texel at location i in the one-dimensional texture. For onedimensional array textures, both texels are obtained from layer l, where 1 l = clamp( t + , 0, hs − 1). 2 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. 8.14.2.1

Rendering Feedback Loops

If all of the following conditions are satisfied, then the value of the selected τijk , τij , or τi in the above equations is undefined instead of referring to the value of the OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

264

texel at location (i, j, k), (i, j), or (i) 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_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. Rectangle textures do not support mipmapping (it is an error to specify a minification filter that requires mipmapping). 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 ws × hs × ds , then there are blog2 (maxsize)c + 1 levels in the mipmap, where   for 1D and 1D array textures ws , maxsize = max(ws , hs ), for 2D, 2D array, cube map, and cube map array textures   max(ws , hs , ds ), for 3D textures Numbering the levels such that level levelbase is the 0th level, the ith array has dimensions ws hs ds max(1, ) × max(1, ) × max(1, ) wd hd dd where OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

wd = 2i ( 1, hd = 2i , ( 2i , dd = 1,

265

for 1D and 1D array textures otherwise for 3D textures otherwise

until the last array is reached with dimension 1 × 1 × 1. Each array in a mipmap is defined using TexImage3D, TexImage2D, CopyTexImage2D, TexImage1D, or CopyTexImage1D 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 (see section 8.19), levelbase is clamped to the range [0, levelimmut − 1], levelmax is then clamped to the range [levelbase , levelimmut − 1], and p is one less than levelimmut , where levelimmut is the levels parameter passed to TexStorage* for the texture object (the value of TEXTURE_IMMUTABLE_LEVELS; see section 8.19). 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. 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 + 21 (8.14)   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 8.10, with OpenGL 4.6 (Core Profile) - May 5, 2022

8.14. TEXTURE MINIFICATION

266

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

(8.15) (8.16)

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 8.10, 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 8.10, 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 .

8.14.4

Manual Mipmap Generation

Mipmaps can be generated manually for a texture object with the commands void GenerateMipmap( enum target ); void GenerateTextureMipmap( uint texture ); For GenerateMipmap, the texture object is that bound to target. For GenerateTextureMipmap, texture is the name of the texture object. target or the effective target of texture must be one of TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP, or TEXTURE_CUBE_MAP_ARRAY. If target or the effective target of texture is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, then the texture object 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.15. TEXTURE MAGNIFICATION

267

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. The internal formats and border widths of the derived mipmap texture 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 one- and 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 as the default filter. Any synchronization required before performing this reduction will be done within the Generate*Mipmap commands themselves5 . Errors An INVALID_ENUM error is generated by GenerateMipmap if target is not one of the valid targets listed above. An INVALID_OPERATION error is generated by GenerateTextureMipmap if texture is not the name of an existing texture object. An INVALID_OPERATION error is generated by GenerateTextureMipmap if the effective target is not one of the valid targets listed above. An INVALID_OPERATION error is generated by GenerateTextureMipmap if the effective target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture object is not cube complete or cube array complete, respectively.

8.14.5 This subsection is only defined in the compatibility profile.

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 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 4.6 (Core Profile) - May 5, 2022

8.16. COMBINED DEPTH/STENCIL TEXTURES

268

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 8.20. 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. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.17. TEXTURE COMPLETENESS

269

• 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 internal format of the texture is integer (see table 8.12). – 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.

8.17.1

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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.18. TEXTURE VIEWS

270

TEXTURE_MIN_FILTER requiring mipmaps, the texture will be considered incom-

plete 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 15.2.1.

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 image of level one or greater to be created only if a mipmap complete set of images consistent with the requested image can be supported with default values of TEXTURE_BASE_LEVEL and TEXTURE_MAX_LEVEL (see table 23.14). 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

Texture Views

A texture can be created which references the data store of another texture and interprets the data with a different format, and/or selects a subset of the levels and/or layers of the other texture. The data store for such a texture is shared with the data store of the original texture. Updating the shared data store using the original texture affects texture values read using the new texture, and vice versa. A texture data store remains in existence until all textures that reference it are deleted. The command void TextureView( uint texture, enum target, uint origtexture, enum internalformat, uint minlevel, uint numlevels, uint minlayer, uint numlayers ); initializes the texture named texture to the target specified by target. texture’s data store is inherited from the texture named origtexture, but elements of the data store are interpreted according to the internal format specified by internalformat. Additionally, if origtexture is an array or has multiple mipmap levels, the parameters OpenGL 4.6 (Core Profile) - May 5, 2022

8.18. TEXTURE VIEWS

271

minlayer, numlayers, minlevel, and numlevels control which of those slices and levels are considered part of the texture. The minlevel and minlayer parameters are relative to the view of origtexture. If numlayers or numlevels extend beyond origtexture, they are clamped to the maximum extent of the original texture. If the command is successful, the texture parameters in texture are updated as follows: • TEXTURE_IMMUTABLE_FORMAT is set to TRUE. • TEXTURE_IMMUTABLE_LEVELS is set to the value of TEXTURE_IMMUTABLE_LEVELS for origtexture. • TEXTURE_VIEW_MIN_LEVEL is set to minlevel plus the value of TEXTURE_VIEW_MIN_LEVEL for origtexture. • TEXTURE_VIEW_MIN_LAYER is set to minlayer plus the value of TEXTURE_VIEW_MIN_LAYER for origtexture. • TEXTURE_VIEW_NUM_LEVELS is set to the lesser of numlevels and the value of TEXTURE_VIEW_NUM_LEVELS for origtexture minus minlevel. • TEXTURE_VIEW_NUM_LAYERS is set to the lesser of numlayers and the value of TEXTURE_VIEW_NUM_LAYERS for origtexture minus minlayer. The new texture’s target must be compatible with the target of origtexture, as defined by table 8.21. Numerous constraints on numlayers and the texture dimensions depend on target and the target of origtexture. These constraints are summarized below in the errors section. When origtexture’s target is TEXTURE_CUBE_MAP, the layer parameters are interpreted in the same order as if it were a TEXTURE_CUBE_MAP_ARRAY with 6 layer-faces. The two textures’ internal formats must be compatible according to table 8.22 if the internal format exists in that table. The internal formats must be identical if not in that table. If the internal formats are the same but are a base internal format, the implementation’s effective internal format (see the end of section 8.5) for each texture must be the same. 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 8.26.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.18. TEXTURE VIEWS

Original target TEXTURE_1D TEXTURE_2D TEXTURE_3D TEXTURE_CUBE_MAP

TEXTURE_RECTANGLE TEXTURE_BUFFER TEXTURE_1D_ARRAY TEXTURE_2D_ARRAY

TEXTURE_CUBE_MAP_ARRAY TEXTURE_2D_MULTISAMPLE TEXTURE_2D_MULTISAMPLE_ARRAY

272

Valid new targets TEXTURE_1D, TEXTURE_1D_ARRAY TEXTURE_2D, TEXTURE_2D_ARRAY TEXTURE_3D TEXTURE_CUBE_MAP, TEXTURE_2D_ARRAY, MAP_ARRAY TEXTURE_RECTANGLE

TEXTURE_2D, TEXTURE_CUBE_-

none TEXTURE_1D_ARRAY, TEXTURE_1D TEXTURE_2D_ARRAY, TEXTURE_2D, TEXTURE_CUBE_MAP, TEXTURE_CUBE_MAP_ARRAY TEXTURE_CUBE_MAP_ARRAY, TEXTURE_2D_ARRAY, TEXTURE_2D, TEXTURE_CUBE_MAP TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY

Table 8.21: Legal texture targets for TextureView.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.18. TEXTURE VIEWS

Class VIEW_CLASS_128_BITS VIEW_CLASS_96_BITS VIEW_CLASS_64_BITS VIEW_CLASS_48_BITS VIEW_CLASS_32_BITS

VIEW_CLASS_24_BITS VIEW_CLASS_16_BITS VIEW_CLASS_8_BITS VIEW_CLASS_RGTC1_RED VIEW_CLASS_RGTC2_RG VIEW_CLASS_BPTC_UNORM VIEW_CLASS_BPTC_FLOAT

273

Internal formats RGBA32F, RGBA32UI, RGBA32I RGB32F, RGB32UI, RGB32I RGBA16F, RG32F, RGBA16UI, RG32UI, RGBA16I, RG32I, RGBA16, RGBA16_SNORM RGB16, RGB16_SNORM, RGB16F, RGB16UI, RGB16I RG16F, R11F_G11F_B10F, R32F, RGB10_A2UI, RGBA8UI, RG16UI, R32UI, RGBA8I, RG16I, R32I, RGB10_A2, RGBA8, RG16, RGBA8_SNORM, RG16_SNORM, SRGB8_ALPHA8, RGB9_E5 RGB8, RGB8_SNORM, SRGB8, RGB8UI, RGB8I R16F, RG8UI, R16UI, RG8I, R16I, RG8, R16, RG8_SNORM, R16_SNORM R8UI, R8I, R8, R8_SNORM COMPRESSED_RED_RGTC1, COMPRESSED_SIGNED_RED_RGTC1 COMPRESSED_RG_RGTC2, COMPRESSED_SIGNED_RG_RGTC2 COMPRESSED_RGBA_BPTC_UNORM, COMPRESSED_SRGB_ALPHA_BPTC_UNORM COMPRESSED_RGB_BPTC_SIGNED_FLOAT, COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT

Table 8.22: Compatible internal formats for TextureView. Formats in the same row may be cast to each other.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.18. TEXTURE VIEWS

274

Texture commands that take a level or layer parameter, such as TexSubImage2D, interpret that parameter to be relative to the view of the texture. i.e. the mipmap level of the data store that would be updated via TexSubImage2D would be the sum of level and the value of TEXTURE_VIEW_MIN_LEVEL. Errors An INVALID_VALUE error is generated if texture is zero. An INVALID_OPERATION error is generated if texture is not a valid name returned by GenTextures, or if texture has already been bound and given a target. An INVALID_VALUE error is generated if origtexture is not the name of a texture. An INVALID_OPERATION error is generated if the value of TEXTURE_IMMUTABLE_FORMAT for origtexture is not TRUE. An INVALID_OPERATION error is generated if target is not compatible with the target of origtexture, as defined by table 8.21. An INVALID_OPERATION error is generated if the internal format of origtexture exists in table 8.22 and is not compatible with internalformat, as described in that table. An INVALID_OPERATION error is generated if the internal format of origtexture does not exist in table 8.22, and is not identical to internalformat. An INVALID_VALUE error is generated if minlevel or minlayer are larger than the greatest level or layer, respectively, of origtexture. An INVALID_VALUE error is generated if target is TEXTURE_CUBE_MAP and the clamped numlayers is not 6. An INVALID_VALUE error is generated if target is TEXTURE_CUBE_MAP_ARRAY and the clamped numlayers is not a multiple of 6. An INVALID_VALUE error is generated if target is TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_RECTANGLE, or TEXTURE_2D_MULTISAMPLE and numlayers does not equal 1. An INVALID_OPERATION error is generated if target is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the width and height of origtexture’s levels are not equal. An INVALID_OPERATION error is generated if any dimension of origtexture is larger than the maximum supported corresponding dimension of the new target. For example, if origtexture has a TEXTURE_2D_ARRAY target and target is TEXTURE_CUBE_MAP, its width must be no greater than the value of MAX_CUBE_MAP_TEXTURE_SIZE.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

275

An INVALID_OPERATION error is generated if the computed values of TEXTURE_VIEW_NUM_LEVELS or TEXTURE_VIEW_NUM_LAYERS for texture, as described above, are less than or equal to zero.

8.19

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, unless it is a proxy texture (since otherwise it would no longer be possible to use the proxy). 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: • 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. TEXTURE_IMMUTABLE_LEVELS and TEXTURE_VIEW_NUM_LEVELS become levels. If the texture target is TEXTURE_1D_ARRAY then TEXTURE_VIEW_NUM_LAYERS becomes height. If the texture target is TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY, or TEXTURE_2D_MULTISAMPLE_ARRAY then TEXTURE_VIEW_NUM_LAYERS becomes depth. If the texture target is TEXTURE_CUBE_MAP, then TEXTURE_VIEW_NUM_LAYERS becomes 6. For any other texture target, TEXTURE_VIEW_NUM_LAYERS becomes 1. The TexStorage* commands specify properties of the texture object bound to the target parameter of each command. OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

276

The TextureStorage* commands behave similarly to the equivalent TexStorage* commands, but specify properties of the texture object named by the texture parameter of each command. The effective target of texture must be compatible with the target parameter of the equivalent TexStorage* command. 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 by TexStorage* if zero is bound to target. An INVALID_OPERATION error is generated by TextureStorage* if texture is not the name of an existing texture object. An INVALID_VALUE error is generated if width, height, depth or levels are less than 1, for commands with the corresponding parameters. An INVALID_ENUM error is generated if internalformat is one of the unsized base internal formats listed in table 8.11. The commands void TexStorage1D( enum target, sizei levels, enum internalformat, sizei width ); void TextureStorage1D( uint texture, sizei levels, enum internalformat, sizei width ); specify all the levels of a one-dimensional texture (or for TexStorage1D, proxy) at the same time. TexStorage1D is described by the pseudocode below: for (i = 0; i < levels; i++) { TexImage1D(target, i, internalf ormat, width, 0, f ormat, type, widthNULL); width = max(1, ); 2 }

Errors In addition to the generic errors described at the start of this section, An INVALID_ENUM error is generated by TexStorage1D if target is not TEXTURE_1D or PROXY_TEXTURE_1D. An INVALID_OPERATION error is generated by TextureStorage1D if the

OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

277

effective target is not TEXTURE_1D. An INVALID_OPERATION error is generated if levels is greater than blog2 (width)c + 1. The commands void TexStorage2D( enum target, sizei levels, enum internalformat, sizei width, sizei height ); void TextureStorage2D( uint texture, sizei levels, enum internalformat, sizei width, sizei height ); specify all the levels of a two-dimensional, cube map, one-dimensional array or rectangle texture (or for TexStorage2D, proxy) at the same time. TexStorage2D is described by the target-dependent pseudocode below: targets TEXTURE_2D, PROXY_TEXTURE_2D, TEXTURE_RECTANGLE, PROXY_TEXTURE_RECTANGLE, or PROXY_TEXTURE_CUBE_MAP: 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 (each target in table 8.19) { TexImage2D(face, i, internalf ormat, width, height, 0, f ormat, type, NULL); } width = max(1, jwidth ); 2 k height = max(1,

height 2

);

} targets TEXTURE_1D_ARRAY or PROXY_TEXTURE_1D_ARRAY: for (i = 0; i < levels; i++) { TexImage2D(target, i, internalf ormat, width, height, 0, f ormat, type, NULL); ); width = max(1, width 2 } OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

278

Errors In addition to the generic errors described at the start of this section, An INVALID_ENUM error is generated by TexStorage2D if target is not one of those listed above. An INVALID_OPERATION error is generated by TextureStorage2D if the effective target is not one of those listed above. An INVALID_OPERATION error is generated if any of the following conditions hold: • The effective target is TEXTURE_1D_ARRAY or PROXY_TEXTURE_1D_ARRAY, and levels is greater than blog2 (width)c + 1 • The effective target is not TEXTURE_1D_ARRAY or PROXY_TEXTURE_1D_ARRAY, and levels is greater than blog2 (max(width, height))c + 1 An INVALID_OPERATION error is generated by TexStorage2D and TextureStorage2D if internalformat is one of the EAC, ETC2, or RGTC formats and the effective target is TEXTURE_RECTANGLE. The commands void TexStorage3D( enum target, sizei levels, enum internalformat, sizei width, sizei height, sizei depth ); void TextureStorage3D( uint texture, sizei levels, enum internalformat, sizei width, sizei height, sizei depth ); specify all the levels of a three-dimensional, two-dimensional array texture, or cube map array texture (or for TexStorage3D, proxy). TexStorage3D is described by the target-dependent pseudocode below: targets TEXTURE_3D or PROXY_TEXTURE_3D: for (i = 0; i < levels; i++) { TexImage3D(target, i, internalf ormat, width, height, depth, 0, f ormat, type, NULL); width = max(1, jwidth ); 2 k height = max(1, height ); j 2k depth = max(1, depth ); 2

OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

279

} targets TEXTURE_2D_ARRAY, PROXY_TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or PROXY_TEXTURE_CUBE_MAP_ARRAY: for (i = 0; i < levels; i++) { TexImage3D(target, i, internalf ormat, width, height, depth, 0, f ormat, type, NULL); width = max(1, jwidth ); 2 k height = max(1,

height 2

);

}

Errors In addition to the generic errors described at the start of this section, An INVALID_ENUM error is generated by TexStorage3D if target is not one of those listed above. An INVALID_OPERATION error is generated by TextureStorage3D if the effective target is not one of those listed above. An INVALID_OPERATION error is generated if any of the following conditions hold: • The effective target is TEXTURE_3D or PROXY_TEXTURE_3D and levels is greater than blog2 (max(width, height, depth)))c + 1 • The effective target is TEXTURE_2D_ARRAY, PROXY_TEXTURE_2D_ARRAY, TEXTURE_CUBE_MAP_ARRAY or PROXY_TEXTURE_CUBE_MAP_ARRAY and levels is greater than blog2 (max(width, height))c + 1 • The commands void TexStorage2DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, boolean fixedsamplelocations ); void TextureStorage2DMultisample( uint texture, sizei samples, enum internalformat, sizei width, sizei height, boolean fixedsamplelocations );

OpenGL 4.6 (Core Profile) - May 5, 2022

8.19. IMMUTABLE-FORMAT TEXTURE IMAGES

280

specify a two-dimensional multisample texture (or for TexStorage2DMultisample, proxy). For TexStorage2DMultisample, target must be TEXTURE_2D_MULTISAMPLE or PROXY_TEXTURE_2D_MULTISAMPLE. Calling TexStorage2DMultisample is equivalent to calling TexImage2DMultisample with the equivalently named parameters set to the same values, except that the resulting texture has immutable format. Errors In addition to the generic errors described at the start of this section, An INVALID_ENUM error is generated by TexStorage2DMultisample if target is not TEXTURE_2D_MULTISAMPLE or PROXY_TEXTURE_2D_MULTISAMPLE. An INVALID_OPERATION error is generated by TextureStorage2DMultisample if the effective target is not TEXTURE_2D_MULTISAMPLE. The commands void TexStorage3DMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height, sizei depth, boolean fixedsamplelocations ); void TextureStorage3DMultisample( uint texture, sizei samples, enum internalformat, sizei width, sizei height, sizei depth, boolean fixedsamplelocations ); specify a two-dimensional multisample array texture (or, for TexStorage3DMultisample, proxy). For TexStorage3DMultisample, target must be TEXTURE_2D_MULTISAMPLE_ARRAY or PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY. Calling TexStorage3DMultisample is equivalent to calling TexImage3DMultisample with the equivalently named parameters set to the same values, except that the resulting texture has immutable format. Errors In addition to the generic errors described at the start of this section, An INVALID_ENUM error is generated by TexStorage3DMultisample

OpenGL 4.6 (Core Profile) - May 5, 2022

8.20. INVALIDATING TEXTURE IMAGE DATA

281

if target is not TEXTURE_2D_MULTISAMPLE_ARRAY or PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY. An INVALID_OPERATION error is generated by TextureStorage3DMultisample if target is not TEXTURE_2D_MULTISAMPLE_ARRAY.

8.19.1

Behavior of Immutable-Format Texture Images

After a successful call to any Tex*Storage* command with a non-proxy target, 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 commands CompressedTexImage*, CopyTexImage*, TexImage*, and TexStorage* with the same texture, even if it does not affect the dimensions or format:

8.20

Invalidating Texture Image Data

All or part of a texture image may be invalidated, effectively leaving those texels undefined, by calling void InvalidateTexSubImage( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth ); with texture and level indicating which texture image is being invalidated. After this command, data in that subregion have undefined values. Arguments xoffset, yoffset, and zoffset specify the lower left back texel coordinates of a width-wide by height-high by depth-deep rectangular subregion of the texture image to invalidate, and are interpreted as described for TexSubImage3D in section 8.6. The subregion must lie within the bounds of the texture image, as described in that section. Cube map textures are treated as an array of six slices in the z-dimension, where a value of zoffset is interpreted as specifying the cube map face for the corresponding layer in table 9.3. For texture types that do not have certain dimensions, InvalidateTexSubImage treats those dimensions as having a size of 1. For example, to invalidate a portion of a two-dimensional texture, use zoffset equal to zero and depth equal to one. OpenGL 4.6 (Core Profile) - May 5, 2022

8.21. CLEARING TEXTURE IMAGE DATA

282

Errors An INVALID_VALUE error is generated if level is negative or greater than log2 of the maximum texture width, height, or depth. An INVALID_OPERATION error is generated if the specified subregion does not lie within the bounds of the texture image, as described for TexSubImage3D in section 8.6. An INVALID_OPERATION error is generated if texture is zero or is not the name of a texture. It is not possible to invalidate a portion of a default texture. An INVALID_VALUE error is generated if the effective target of texture is TEXTURE_RECTANGLE, TEXTURE_BUFFER, TEXTURE_2D_MULTISAMPLE, or TEXTURE_2D_MULTISAMPLE_ARRAY, and level is not zero. An INVALID_VALUE error is generated if width, height, or depth is negative. The command void InvalidateTexImage( uint texture, int level ); is equivalent to calling InvalidateTexSubImage with xoffset, yoffset, and zoffset equal to 0 and width, height, and depth equal to the dimensions of the texture image (or zero and one for dimensions the texture doesn’t have).

8.21

Clearing Texture Image Data

All or part of a texture image may be filled with a constant value with the command void ClearTexSubImage( uint texture, int level, int xoffset, int yoffset, int zoffset, sizei width, sizei height, sizei depth, enum format, enum type, const void *data ); with texture and level indicating which texture image is being cleared. It is an error if texture is zero or not the name of a texture object, if texture is a buffer texture, or if the texture image has a compressed internal format. Arguments xoffset, yoffset, and zoffset specify the lower left back texel coordinates of a width-wide by height-high by depth-deep rectangular subregion of the texture image, and are interpreted as described for TexSubImage3D in section 8.6. The subregion must lie within the bounds of the texture image, as described in that section. OpenGL 4.6 (Core Profile) - May 5, 2022

8.21. CLEARING TEXTURE IMAGE DATA

283

For one-dimensional array textures, yoffset is interpreted as the first layer to be cleared and height is the number of layers to clear. For two-dimensional array textures, zoffset is interpreted as the first layer to be cleared and depth is the number of layers to clear. Cube map textures are treated as an array of six slices in the zdimension, where the value of zoffset is interpreted as specifying the cube map face for the corresponding layer in table 9.3 and depth is the number of faces to clear. For cube map array textures, zoffset is the first layer-face to clear, and depth is the number of layer-faces to clear. Each layer-face is translated into an array layer and a cube map face as described for layer-face numbers in section 8.5.3. For texture types that do not have certain dimensions, ClearTexSubImage treats those dimensions as having a size of 1. For example, to clear a portion of a two-dimensional texture, use zoffset equal to zero and depth equal to one. format and type specify the format and type of the source data and are interpreted as they are for TexImage3D, as described in section 8.4.4. Textures with a base internal format of DEPTH_COMPONENT, STENCIL_INDEX, DEPTH_STENCIL require depth component, stencil, or depth/stencil component data respectively. Textures with other base internal formats require RGBA formats. Textures with integer internal formats (see table 8.12) require integer data. data is a pointer to an array of between one and four components of texel data that will be used as the source for the constant fill value. The elements of data are converted by the GL into the internalformat of the texture image (that was specified when the level was defined by any of the TexImage, TexStorage or CopyTexImage commands) in the manner described in section 8.4.4, and then used to fill the specified range of the destination texture level. If data is NULL, then the pointer is ignored and the sub-range of the texture image is filled with zeros. If texture is a multisample texture, all the samples in a texel are cleared to the value specified by data. Errors An INVALID_OPERATION error is generated if texture is zero or not the name of a texture object. An INVALID_OPERATION error is generated if texture is a buffer texture. An INVALID_OPERATION error is generated if texture has a compressed internal format. An INVALID_VALUE error is generated if width, height, or depth is negative. An INVALID_OPERATION error is generated if the base internal format is DEPTH_COMPONENT and format is not DEPTH_COMPONENT.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.22. TEXTURE STATE AND PROXY STATE

284

An INVALID_OPERATION error is generated if the base internal format is DEPTH_STENCIL and format is not DEPTH_STENCIL. An INVALID_OPERATION error is generated if the base internal format is STENCIL_INDEX and format is not STENCIL_INDEX. An INVALID_OPERATION error is generated if the base internal format is RGBA and the format is DEPTH_COMPONENT, STENCIL_INDEX, or DEPTH_STENCIL. An INVALID_OPERATION error is generated if the internal format is integer and format does not specify integer data. An INVALID_OPERATION error is generated if the internal format is not integer and format does specify integer data. An INVALID_OPERATION error is generated if the specified subregion does not lie within the bounds of the texture image, as described for TexSubImage3D in section 8.6. The command void ClearTexImage( uint texture, int level, enum format, enum type, const void * data ); is equivalent to calling ClearTexSubImage with xoffset, yoffset, and zoffset equal to 0, and width, height, and depth equal to the dimensions of the texture image (or zero and one for dimensions the texture doesn’t have). Errors In addition to the errors generated by ClearTexSubImage: An INVALID_OPERATION error is generated if the texture image identified by level has not previously been defined by a TexImage* or TexStorage* command.

8.22

Texture State and Proxy State

The state necessary for texture can be divided into two categories. First, there are the multiple sets of texture images (a single image for the rectangle texture target; one set of mipmap images each for the one-, two-, and three-dimensional and one- 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 array has associated with it a width, height (two- and three-dimensional, rectangle, one-dimensional array, cube map, and cube map array only), and depth OpenGL 4.6 (Core Profile) - May 5, 2022

8.22. TEXTURE STATE AND PROXY STATE

285

(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, floatingpoint, 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, or R8 for buffer textures, 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 the same 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 one-, two-, 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 (two- and three-dimensional and cube map only), 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 image levels, a boolean flag indicating whether the format and dimensions of the texture are immutable, three integers describing the depth texture mode, compare mode, and compare function, an integer describing the depth stencil texture mode, and four integers describing the red, green, blue, and alpha swizzle modes (see section 15.2.1). In the initial state, the value assigned to TEXTURE_MIN_FILTER is NEAREST_MIPMAP_LINEAR (except for rectangle textures, where the initial value is LINEAR), and the value for TEXTURE_MAG_FILTER is LINEAR. s, t, and r wrap modes are all set to REPEAT (except for rectangle textures, where the initial value is CLAMP_TO_EDGE). 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 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. The values of TEXTURE_OpenGL 4.6 (Core Profile) - May 5, 2022

8.22. TEXTURE STATE AND PROXY STATE

286

IMMUTABLE_LEVELS, TEXTURE_VIEW_MIN_LEVEL, TEXTURE_VIEW_NUM_LEVELS, TEXTURE_VIEW_MIN_LAYER, TEXTURE_VIEW_NUM_LAYERS are each

zero. In addition to texture images for the non-proxy texture targets described above, partially instantiated texture images are maintained for one-, two-, and threedimensional, rectangle, one- and two-dimensional array, and cube map array textures. Additionally, a single proxy image is maintained for the cube map texture. Each proxy image includes width, height, depth, number of samples, and internal format state values, as well as state for the red, green, blue, alpha, depth, and stencil component resolutions and types. Proxy images do not include image data or texture parameters. When TexImage3D is executed with target specified as PROXY_TEXTURE_3D, the three-dimensional proxy state values of the specified level-of-detail are recomputed and updated. If the image would not be supported by TexImage3D called with target set to TEXTURE_3D, no error is generated, but the proxy width, height, depth, number of samples, and component resolutions are set to zero, and the component types are set to NONE. If the image would be supported by such a call to TexImage3D, the proxy state values are set exactly as though the actual image were being specified. No pixel data are transferred or processed in either case. Proxy images for one- and two-dimensional textures, one- and twodimensional array textures, and cube map array textures are operated on in the same way when TexImage1D is executed with target specified as PROXY_TEXTURE_1D, TexImage2D is executed with target specified as PROXY_TEXTURE_2D, PROXY_TEXTURE_1D_ARRAY, or PROXY_TEXTURE_RECTANGLE, or TexImage3D is executed with target specified as PROXY_TEXTURE_2D_ARRAY or PROXY_TEXTURE_CUBE_MAP_ARRAY. Proxy images for two-dimensional multisample and two-dimensional multisample array textures are operated on in the same way when TexImage2DMultisample is called with target specified as PROXY_TEXTURE_2D_MULTISAMPLE, or TexImage3DMultisample is called with target specified as PROXY_TEXTURE_2D_MULTISAMPLE_ARRAY. However, if samples is not supported, then no error is generated. The cube map proxy images are operated on in the same manner when TexImage2D is executed with the target field specified as PROXY_TEXTURE_CUBE_MAP, with the addition that determining that a given cube map texture is supported with PROXY_TEXTURE_CUBE_MAP indicates that all six of the cube map twodimensional images are supported. Likewise, if the specified PROXY_TEXTURE_CUBE_MAP is not supported, none of the six cube map two-dimensional images are supported.

OpenGL 4.6 (Core Profile) - May 5, 2022

8.23. TEXTURE COMPARISON MODES

287

Errors An INVALID_ENUM error is generated by BindTexture, GetTexImage, GetTexParameteriv, and GetTexParameterfv when called with a proxy texture target. There is no image or non-level-related state associated with proxy textures, therefore they may not be used as textures.

8.23

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.

8.23.1

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. 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 . • 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 8.23. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.24. SRGB TEXTURE COLOR CONVERSION Texture Comparison Function LEQUAL

GEQUAL

LESS

GREATER

EQUAL

NOTEQUAL ALWAYS NEVER

288

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 8.23: Depth texture comparison functions.

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.

8.24

sRGB Texture Color Conversion

If the currently bound texture’s internal format is one of the sRGB formats in table 8.24, 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. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

8.25. SHARED EXPONENT TEXTURE COLOR CONVERSION

289

Internal Format SRGB SRGB8 SRGB_ALPHA SRGB8_ALPHA8 COMPRESSED_SRGB COMPRESSED_SRGB8_ETC2 COMPRESSED_SRGB_ALPHA COMPRESSED_SRGB8_ALPHA8_ETC2_EAC COMPRESSED_SRGB8_PUNCHTHROUGH_ALPHA1_ETC2 COMPRESSED_SRGB_ALPHA_BPTC_UNORM

Table 8.24: sRGB texture internal formats.

is as follows. ( cl =

cs 12.92 , cs +0.055 2.4 , 1.055

cs ≤ 0.04045 cs > 0.04045

(8.17)

Assume cs is the sRGB component in the range [0, 1].

8.25

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

8.26

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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

290

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 binding point for each texture target; each image unit may have only one texture bound at a time. A texture may be bound to an image unit for use by image loads and stores with the command 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 one-dimensional array, two-dimensional array, three-dimensional, cube map, cube map array, or two-dimensional multisample 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. When layered is FALSE, the single bound layer is treated as a different texture target for image accesses: • one-dimensional array texture layers are treated as one-dimensional textures; • two-dimensional array, three-dimensional, cube map, and cube map array texture layers are treated as two-dimensional textures; and • two-dimensional multisample array textures are treated as two-dimensional multisample textures. For cube map textures where layered is FALSE, the face is taken by mapping the layer number to a face according to table 9.3. 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 9.3. 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 for layered and layer. OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

291

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. 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. The command void BindImageTextures( uint first, sizei count, const uint *textures ); binds count existing texture objects to image units numbered first through f irst + count − 1. If textures is not NULL, it specifies an array of count values, each of which must be zero or the name of an existing texture object. If textures is NULL, each affected image unit from first through f irst + count − 1 will be reset to have no bound texture object. When binding a non-zero texture object to an image unit, the image unit level, layered, layer, and access parameters are set to zero, TRUE, zero, and READ_WRITE, respectively. The image unit format parameter is taken from the internal format of the texture image at level zero of the texture object identified by textures. For cube map textures, the internal format of the TEXTURE_CUBE_MAP_POSITIVE_X image of level zero is used. For multisample, multisample array, buffer, and rectangle textures, the internal format of the single texture level is used. When unbinding a texture object from an image unit, the image unit parameters level, layered, layer, and format will be reset to their default values of zero, FALSE, 0, and R8, respectively. BindImageTextures is equivalent (assuming no errors are generated) to: OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

292

for (i = 0; i < count; i++) { if (textures == NULL || textures[i] = 0) { BindImageTexture(f irst + i, 0, 0, FALSE, 0, READ_ONLY, R8); } else { BindImageTexture(f irst + i, textures[i], 0, TRUE, 0, READ_WRITE, lookupInternalFormat(textures[i])); } } where lookupInternalFormat returns the internal format of the specified texture object. The values specified in textures will be checked separately for each image unit. When a value for a specific image unit is invalid, the state for that image unit will be unchanged and an error will be generated. However, state for other image units will still be changed if their corresponding values are valid. Errors An INVALID_OPERATION error is generated if f irst + count is greater than the number of image units supported by the implementation. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if any value in textures is not zero or the name of an existing texture object (per binding). An INVALID_OPERATION error is generated if the internal format of the level zero texture image of any texture in textures is not found in table 8.26 (per binding). An INVALID_OPERATION error is generated if the width, height, or depth of the level zero texture image of any texture in textures is zero (per binding). 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 one-, two-, or three-dimensional coordinate. Multisample texture accesses also specify a sample number. A coordinate vector is mapped to an individual texel τi , 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 OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES Texture target TEXTURE_1D TEXTURE_2D TEXTURE_3D TEXTURE_RECTANGLE TEXTURE_CUBE_MAP TEXTURE_BUFFER TEXTURE_1D_ARRAY TEXTURE_2D_ARRAY TEXTURE_CUBE_MAP_ARRAY TEXTURE_2D_MULTISAMPLE TEXTURE_2D_MULTISAMPLE_ARRAY

i x x x x x x x x x x x

293 j y y y y y y y y

k z -

Face / layer z y z z z

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

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 described for 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 zero. 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, as described below; • the texture bound to the image unit has layers, and the selected layer or cube map face doesn’t exist;

OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

294

• the selected texel τi , τij , or τijk doesn’t exist; • the image has more samples than the implementation-dependent value of MAX_IMAGE_SAMPLES. Additionally, there are a number of cases where image load, store, or atomic operations are considered to involve a format mismatch. In such cases, undefined values will be returned by image loads and atomic operations and undefined values will be written by stores 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 • 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. For textures with multiple samples per texel, the sample selected for an image load, store, or atomic is undefined if the sample coordinate is negative or greater than or equal to the number of samples in the texture. If a shader performs an image load, store, or atomic operation using an image variable declared as an array, and if the index used to select an individual element is negative or greater than or equal to the size of the array, the results of the operation are undefined but may not lead to termination. 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 GetTexImage command with format and type as RGBA and FLOAT for vec4 data, as RGBA_INTEGER and INT for OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

295

ivec4 data, or as RGBA_INTEGER and UNSIGNED_INT for uvec4 data, respec-

tively. Unused components are filled in with (0, 0, 0, 1) (where 0 and 1 are either floating-point or integer values, depending on the format). Any image variable used for shader loads or atomic memory 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 above. Image variables used exclusively for image stores need not include a format layout qualifier, but any declared qualifier must match the image unit format to avoid a format mismatch. Image Unit Format

Format Qualifer

RGBA32F RGBA16F RG32F RG16F R11F_G11F_B10F R32F R16F RGBA32UI RGBA16UI RGB10_A2UI RGBA8UI RG32UI RG16UI RG8UI R32UI R16UI R8UI RGBA32I RGBA16I RGBA8I RG32I RG16I RG8I R32I R16I R8I

rgba32f rgba16f rg32f rg16f r11f_g11f_b10f r32f r16f rgba32ui rgba16ui rgb10_a2ui rgba8ui rg32ui rg16ui rg8ui r32ui r16ui r8ui rgba32i rgba16i rgba8i rg32i rg16i rg8i r32i r16i r8i

(Continued on next page) OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

296

Supported image unit formats (continued) Image Unit Format Format Qualifer RGBA16 RGB10_A2 RGBA8 RG16 RG8 R16 R8 RGBA16_SNORM RGBA8_SNORM RG16_SNORM RG8_SNORM R16_SNORM R8_SNORM

rgba16 rgb10_a2 rgba8 rg16 rg8 r16 r8 rgba16_snorm rgba8_snorm rg16_snorm rg8_snorm r16_snorm r8_snorm

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

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 a compatible 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 image atomics is performed as though data OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

297

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 image atomics 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 GetTexImage (see section 8.11), 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 RGBA16F RG32F RG16F R11F_G11F_B10F R32F R16F RGBA32UI RGBA16UI RGB10_A2UI RGBA8UI RG32UI RG16UI RG8UI R32UI R16UI R8UI RGBA32I RGBA16I RGBA8I RG32I

Size 128 64 64 32 32 32 16 128 64 32 32 64 32 16 32 16 8 128 64 32 64

Class 4x32 4x16 2x32 2x16 (a) 1x32 1x16 4x32 4x16 (b) 4x8 2x32 2x16 2x8 1x32 1x16 1x8 4x32 4x16 4x8 2x32

Pixel format

Pixel type

RGBA RGBA RG RG RGB RED RED RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RG_INTEGER RG_INTEGER RG_INTEGER RED_INTEGER RED_INTEGER RED_INTEGER RGBA_INTEGER RGBA_INTEGER RGBA_INTEGER RG_INTEGER

FLOAT HALF_FLOAT FLOAT HALF_FLOAT UNSIGNED_INT_10F_11F_11F_REV FLOAT HALF_FLOAT UNSIGNED_INT UNSIGNED_SHORT UNSIGNED_INT_2_10_10_10_REV UNSIGNED_BYTE UNSIGNED_INT UNSIGNED_SHORT UNSIGNED_BYTE UNSIGNED_INT UNSIGNED_SHORT UNSIGNED_BYTE INT SHORT BYTE INT

(Continued on next page) OpenGL 4.6 (Core Profile) - May 5, 2022

8.26. TEXTURE IMAGE LOADS AND STORES

298

Texel sizes, compatibility classes ... (continued) Image Format Size Class Pixel format Pixel type RG16I 32 2x16 RG_INTEGER SHORT RG8I 16 2x8 RG_INTEGER BYTE R32I 32 1x32 RED_INTEGER INT R16I 16 1x16 RED_INTEGER SHORT R8I 8 1x8 RED_INTEGER BYTE RGBA16 64 4x16 RGBA UNSIGNED_SHORT RGB10_A2 32 (b) RGBA UNSIGNED_INT_2_10_10_10_REV RGBA8 32 4x8 RGBA UNSIGNED_BYTE RG16 32 2x16 RG UNSIGNED_SHORT RG8 16 2x8 RG UNSIGNED_BYTE R16 16 1x16 RED UNSIGNED_SHORT R8 8 1x8 RED UNSIGNED_BYTE RGBA16_SNORM 64 4x16 RGBA SHORT RGBA8_SNORM 32 4x8 RGBA BYTE RG16_SNORM 32 2x16 RG SHORT RG8_SNORM 16 2x8 RG BYTE R16_SNORM 16 1x16 RED SHORT R8_SNORM 8 1x8 RED BYTE Table 8.27: Texel sizes, compatibility classes, and pixel format/type combinations for each image format. Class (a) is for 11/11/10 packed floating-point formats; class (b) is for 10/10/10/2 packed formats.

Implementations may support a limited combined number of image units, shader storage blocks (see section 7.8), and active fragment shader outputs (see section 17.4.1). A link error is 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.

8.26.1

Image Unit Queries

The state required for each image unit is summarized in table 23.45 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 4.6 (Core Profile) - 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 299

9.1. FRAMEBUFFER OVERVIEW

300

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 left buffer, the front right buffer, the back left buffer, and the back right buffer. Typically the contents of the front buffers are displayed on a color monitor while the contents of the back buffers are invisible. (Monoscopic contexts display only the front left buffer; stereoscopic contexts display both the front left and the front right buffers.) All color buffers must have the same number of bitplanes, although an implementation or context may choose not to provide right buffers, or back buffers at all. 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 13.8.1 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 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

301

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 may 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 a name returned 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 23.24, as well as one set of the state values listed in table 23.25 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 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

302

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. An INVALID_OPERATION error is generated if framebuffer is not zero or a name returned from a previous call to GenFramebuffers, or if such a name has since been deleted with DeleteFramebuffers. 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 23.73 and 23.24 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. Framebuffer objects may also be created with the command void CreateFramebuffers( sizei n, uint *framebuffers ); CreateFramebuffers returns n previously unused framebuffer names in framebuffers, each representing a new framebuffer object which is a state vector, comprising all the state and with the same initial values listed in table 23.24, as well as one set of the state values listed in table 23.25 for each attachment point of the framebuffer, with the same initial values. Errors An INVALID_VALUE error is generated if n is negative. 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, OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

303

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 objects 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). 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 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

304

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. 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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

305

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 may be queried by calling GetIntegerv with pnames 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 commands void FramebufferParameteri( enum target, enum pname, int param ); void NamedFramebufferParameteri( uint framebuffer, enum pname, int param ); For FramebufferParameteri, the framebuffer object is that bound to target, which must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For NamedFramebufferParameteri, framebuffer is the name of the framebuffer object. pname specifies the parameter of the framebuffer object 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_LAYERS, FRAMEBUFFER_DEFAULT_SAMPLES, or 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 TexImage2DMultisample (see section 8.8). OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

306

Errors An INVALID_ENUM error is generated by FramebufferParameteri if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated if the default framebuffer is bound to target. An INVALID_OPERATION error is generated by NamedFramebufferParameteri if framebuffer is not the name of an existing framebuffer object. 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.

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 23.25. There are several types of framebuffer-attachable images: • The image of a renderbuffer object, which is always two-dimensional. • A single level of a one-dimensional texture, which is treated as a twodimensional image with a height of one.

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

307

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

9.2.3

Framebuffer Object Queries

Parameters of a framebuffer object may be queried with the commands void GetFramebufferParameteriv( enum target, enum pname, int *params ); void GetNamedFramebufferParameteriv( uint framebuffer, enum pname, int *params ); For GetFramebufferParameteriv, the framebuffer object is that bound to target, which must be must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For GetNamedFramebufferParameteriv, framebuffer may be zero, indicating the default draw framebuffer, or the name of the framebuffer object. 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. pname may also be one of DOUBLEBUFFER, IMPLEMENTATION_COLOR_READ_FORMAT, IMPLEMENTATION_COLOR_READ_TYPE, SAMPLES, SAMPLE_BUFFERS, or STEREO, indicating the corresponding framebuffer-dependent state OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

308

from table 23.73. Values of framebuffer-dependent state are identical to those that would be obtained were the framebuffer object bound and queried using the simple state queries in that table. These values may be queried from either a framebuffer object or a default framebuffer. The values of SAMPLES and SAMPLE_BUFFERS are determined as described in section 9.2.3.1. The value of parameter pname for the framebuffer object is returned in params. Errors An INVALID_ENUM error is generated by GetNamedFramebufferParameteriv if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated by GetNamedFramebufferParameteriv if framebuffer is not zero or the name of an existing framebuffer object. An INVALID_ENUM error is generated if pname is not one of the valid values listed above. An INVALID_OPERATION error is generated by GetFramebufferParameteriv if the default framebuffer is bound to target and pname is not one of the accepted values from table 23.73, other than SAMPLE_POSITION. An INVALID_OPERATION error is generated by GetNamedFramebufferParameteriv if framebuffer is zero, and pname is not one of the valid values from table 23.73, other than SAMPLE_POSITION. 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 ); void GetNamedFramebufferAttachmentParameteriv( uint framebuffer, enum attachment, enum pname, int *params ); For GetFramebufferAttachmentParameteriv, the framebuffer object is that bound to target, which must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For GetNamedFramebufferAttachmentParameteriv, framebuffer is zero or the name of a framebuffer object. If framebuffer is zero, then the default draw framebuffer is queried. If a default framebuffer is queried, then attachment must be one of the values listed in table 9.1, selecting a single color, depth or stencil buffer as shown in that OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS attachment FRONT FRONT_LEFT FRONT_RIGHT BACK BACK_LEFT BACK_RIGHT DEPTH STENCIL

309

Buffer Queried Front Left Color Front Left Color Front Right Color Back Left Color Back Left Color Back Right Color Depth buffer Stencil buffer

Table 9.1: Valid attachment parameters when a default framebuffer is queried with Get*FramebufferAttachmentParameteriv, and the buffers they select.

table. Otherwise, attachment must be one of the framebuffer object attachment points listed in table 9.2. 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 Get*FramebufferAttachmentParameteriv, 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: OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

310

• 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, then 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 fixed-point, or unsigned normalized fixed-point components respectively. If no data storage or texture image has been specified for the attachment, then 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, then 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 17.3.6 and 17.3.7. 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 8.24. If attachment is not a color attachment, or no data storage or texture image has been specified for the attachment, then params will contain the value LINEAR. If

the

value RENDERBUFFER, then

FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE

• 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. OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

311

• 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 value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is the name of a cube map texture object, then params will contain the cube map face of the cubemap texture object which contains the attached image. Otherwise params will contain zero. • If pname is FRAMEBUFFER_ATTACHMENT_LAYERED, then params will contain TRUE if an entire level of a three-dimensional texture, cube map texture, or one- or two-dimensional array texture is attached. Otherwise, params will contain FALSE. • If pname is FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER; the value of FRAMEBUFFER_ATTACHMENT_OBJECT_NAME is the name of a threedimensional texture, or a one- or two-dimensional array texture object; 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 by GetFramebufferAttachmentParameteriv if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated by GetNamedFramebufferAttachmentParameteriv if framebuffer is not zero or the name of an existing framebuffer object. 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_ENUM error is generated if the default framebuffer is queried and attachment is not one the values specified in table 9.1. An INVALID_OPERATION error is generated if a framebuffer object is bound to target and attachment is COLOR_ATTACHMENTm where m is greater

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

312

than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if a framebuffer object is queried, attachment is not one of the attachments in table 9.2, 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 attachment is DEPTH_STENCIL_ATTACHMENT and different objects are bound to the depth and stencil attachment points of the framebuffer object. 9.2.3.1

Multisample Queries

The values of SAMPLE_BUFFERS and SAMPLES control whether and how multisampling is performed (see section 14.3.1). They are framebuffer-dependent constants derived from the attachments of a framebuffer object or the buffers of a default framebuffer, and may be determined either by calling GetFramebufferParameteriv and GetNamedFramebufferParameteriv for a specific framebuffer (see section 9.2.3), or by calling GetIntegerv with pname set to SAMPLE_BUFFERS or SAMPLES. If a framebuffer object is not framebuffer complete, as defined in section 9.4.2, then the values of SAMPLE_BUFFERS and SAMPLES are undefined. Otherwise, the 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 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 returned 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

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

313

state and with the same initial values listed in table 23.27. 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. 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 9.2.3. The current RENDERBUFFER binding can be determined by calling GetIntegerv with pname RENDERBUFFER_BINDING. Errors An INVALID_ENUM error is generated if target is not RENDERBUFFER. An INVALID_OPERATION error is generated if renderbuffer is not zero or a name returned from a previous call to GenRenderbuffers, or if such a name has since been deleted with DeleteRenderbuffers. New renderbuffers may also be created with the command void CreateRenderbuffers( sizei n, uint *renderbuffers ); CreateRenderbuffers returns n previously unused renderbuffer names in renderbuffers, each representing a new renderbuffer object which is a state vector comprising all the state and with the initial values listed in table 23.27. The state of each renderbuffer object is as if a name returned from GenRenderbuffers had been bound to the RENDERBUFFER target, except that any existing binding to RENDERBUFFER is not affected. Errors An INVALID_VALUE error is generated if n is negative. The command OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

314

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. 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 data storage, format, dimensions, and number of samples of a renderbuffer object’s image are established with the commands void RenderbufferStorageMultisample( enum target, sizei samples, enum internalformat, sizei width, sizei height ); OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

315

void NamedRenderbufferStorageMultisample( uint renderbuffer, sizei samples, enum internalformat, sizei width, sizei height ); For RenderbufferStorageMultisample, the renderbuffer object is that bound to target, which must be RENDERBUFFER. For NamedRenderbufferStorageMultisample, renderbuffer is the name of the renderbuffer object. internalformat must be color-renderable, depth-renderable, or stencilrenderable (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 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 *RenderbufferStorageMultisample parameter (except target and renderbuffer), 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 by RenderbufferStorageMultisample if target is not RENDERBUFFER. An INVALID_OPERATION error is generated by NamedRenderbufferStorageMultisample if renderbuffer is not the name of an existing renderbuffer object. 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 22.3). An INVALID_ENUM error is generated if internalformat is not one of the

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

316

color-renderable, depth-renderable, or stencil-renderable formats 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 commands void RenderbufferStorage( enum target, enum internalformat, sizei width, sizei height ); void NamedRenderbufferStorage( uint renderbuffer, enum internalformat, sizei width, sizei height ); are equivalent to RenderbufferStorageMultisample(target,0,internalf ormat,width,height); and NamedRenderbufferStorageMultisample(renderbuf f er,0,internalf ormat,width,height); respectively.

9.2.5

Required Renderbuffer Formats

Implementations are required to support at least one allocation of internal component resolution for each type (unsigned int, float, etc.) for each base internal format. In addition, 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 8.12. • Depth, depth+stencil, and stencil formats which are checked in the “Req. format” column of table 8.13. 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 up to the value of MAX_SAMPLES multisamples, with the exception that the signed and unsigned integer formats are required only to support creation of renderbuffers with up to the value of MAX_INTEGER_SAMPLES multisamples, which must be at least one. OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

9.2.6

Renderbuffer Object Queries

Parameters of a renderbuffer object may be queried with the commands void GetRenderbufferParameteriv( enum target, enum pname, int *params ); void GetNamedRenderbufferParameteriv( uint renderbuffer, enum pname, int *params ); For GetRenderbufferParameteriv, the renderbuffer object is that bound to target, which must be RENDERBUFFER. For GetNamedRenderbufferParameteriv, renderbuffer is the name of the renderbuffer object. The value of renderbuffer parameter pname for the renderbuffer object is returned in params. pname must be one of the symbolic values in table 23.27. 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 object. 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 object. Errors An INVALID_ENUM error is generated by GetRenderbufferParameteriv if target is not RENDERBUFFER. An INVALID_OPERATION error is generated by GetRenderbufferParameteriv if the renderbuffer currently bound to target is zero. An INVALID_OPERATION error is generated by GetNamedRenderbufferParameteriv if renderbuffer is not the name of an existing renderbuffer object. An INVALID_ENUM error is generated if pname is not one of the renderbuffer state names in table 23.27.

9.2.7

Attaching Renderbuffer Images to a Framebuffer

A renderbuffer object can be attached as one of the logical buffers of a framebuffer object with the commmands OpenGL 4.6 (Core Profile) - May 5, 2022

317

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

318

void FramebufferRenderbuffer( enum target, enum attachment, enum renderbuffertarget, uint renderbuffer ); void NamedFramebufferRenderbuffer( uint framebuffer, enum attachment, enum renderbuffertarget, uint renderbuffer ); For FramebufferRenderbuffer the framebuffer object is that bound to target, which must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For NamedFramebufferRenderbuffer, framebuffer is the name of the framebuffer object. attachment must be set to one of the attachment points of the framebuffer listed in table 9.2. 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. 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 23.25. No change is made to the state of the renderbuffer object and any previous attachment to the attachment logical buffer of the framebuffer object 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 framebuffer object are set to their default values listed in table 23.25. 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 frameOpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS buffer object. Note that the renderbuffer image is specifically not detached from any non-bound framebuffer objects. Detaching the image from any non-bound framebuffer objects is the responsibility of the application. Name of attachment COLOR_ATTACHMENTi (see caption) DEPTH_ATTACHMENT STENCIL_ATTACHMENT DEPTH_STENCIL_ATTACHMENT

Table 9.2: Framebuffer attachment points. i in COLOR_ATTACHMENTi may range from zero to the value of MAX_COLOR_ATTACHMENTS minus one.

Errors An INVALID_ENUM error is generated by FramebufferRenderbuffer if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated by FramebufferRenderbuffer if zero is bound to target. An INVALID_OPERATION error is generated by NamedFramebufferRenderbuffer if framebuffer is not the name of an existing framebuffer object. 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 9.2, 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.

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.

OpenGL 4.6 (Core Profile) - May 5, 2022

319

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS 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 commands void FramebufferTexture( enum target, enum attachment, uint texture, int level ); void NamedFramebufferTexture( uint framebuffer, enum attachment, uint texture, int level ); For FramebufferTexture, the framebuffer object is that bound to target, which must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For NamedFramebufferTexture, framebuffer is the name of the framebuffer object. attachment must be one of the attachment points of the framebuffer listed in table 9.2. 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 a three-dimensional texture, cube map array texture, cube map texture, one- or two-dimensional array texture, or two-dimensional multisample array texture, the texture level attached to the framebuffer attachment point is an array of images, and the framebuffer attachment is considered layered. Errors An INVALID_ENUM error is generated by FramebufferTexture if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. An INVALID_OPERATION error is generated by FramebufferTexture if zero is bound to target. An INVALID_OPERATION error is generated by NamedFramebufferTexture if framebuffer is not the name of an existing framebuffer object. 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 9.2, 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 commands OpenGL 4.6 (Core Profile) - May 5, 2022

320

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

321

void FramebufferTexture1D( enum target, enum attachment, enum textarget, uint texture, int level ); void FramebufferTexture2D( enum target, enum attachment, enum textarget, uint texture, int level ); void FramebufferTexture3D( enum target, enum attachment, enum textarget, uint texture, int level, int layer ); target specifies the target to which the framebuffer object is bound, and 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 9.2. If texture is not zero, then texture must either name an existing texture object with a target of textarget, or texture must name an existing cube map texture and textarget must be one of the cube map face targets from table 8.19. 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_VIEW_NUM_LEVELS for texture. • If textarget is TEXTURE_RECTANGLE or TEXTURE_2D_MULTISAMPLE, then level must be zero. • If textarget is TEXTURE_3D, 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 textarget is one of the cube map face targets from table 8.19, 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. • For all other values of textarget, level must be greater than or equal to zero and no larger than log2 of the value of MAX_TEXTURE_SIZE. layer specifies the layer of a two-dimensional image within a three-dimensional texture. Errors An INVALID_ENUM error is generated if target is not DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER.

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

322

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 9.2, 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_VALUE error is generated if texture is not zero and layer is larger than the value of MAX_3D_TEXTURE_SIZE minus one. An INVALID_OPERATION error is generated for FramebufferTexture1D if texture is not zero and textarget is not TEXTURE_1D. An INVALID_OPERATION error is generated for FramebufferTexture2D if texture is not zero and textarget is not one of TEXTURE_2D, TEXTURE_2D_MULTISAMPLE, TEXTURE_RECTANGLE, or one of the cube map face targets from table 8.19. An INVALID_OPERATION error is generated for FramebufferTexture3D if texture is not zero and textarget is not TEXTURE_3D. 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 commands void FramebufferTextureLayer( enum target, enum attachment, uint texture, int level, int layer ); void NamedFramebufferTextureLayer( uint framebuffer, enum attachment, uint texture, int level, int layer ); These commands operate identically to the equivalent FramebufferTexture and NamedFramebufferTexture commands, respectively, except for the additional layer argument which selects a layer of the texture object to attach. layer specifies the layer of a one- or 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 9.3. For cube map array textures, layer is translated into an array layer and a cube map face as described for layer-face numbers in section 8.5.3. OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

323

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_VIEW_NUM_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 and NamedFramebufferTexture 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 an array texture, and layer is larger than the value of MAX_ARRAY_TEXTURE_LAYERS minus one. An INVALID_VALUE error is generated if texture is a cube map array texture, and layer 6 is larger than the value of MAX_CUBE_MAP_TEXTURE_SIZE 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, oneor two-dimensional array, cube map, or cube map array texture. An INVALID_VALUE error is generated if texture is not zero and level is

OpenGL 4.6 (Core Profile) - May 5, 2022

9.2. BINDING AND MANAGING FRAMEBUFFER OBJECTS

324

not a supported texture level for texture, as described above. 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 the other FramebufferTexture* 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 23.25. 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 or *FramebufferTexture3D 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 a threedimensional, cube map, two-dimensional multisample array, or one- or two-dimensional array texture, 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 23.25. No change is made to the state of the OpenGL 4.6 (Core Profile) - May 5, 2022

9.3. FEEDBACK LOOPS BETWEEN TEXTURES AND THE FRAMEBUFFER325 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 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 one- or two-dimensional texture level, a face of a cube map texture level, or a layer of a two-dimensional array, three-dimensional texture, or texture view 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 OpenGL 4.6 (Core Profile) - May 5, 2022

9.3. FEEDBACK LOOPS BETWEEN TEXTURES AND THE FRAMEBUFFER326 bound. 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 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 any shader stage fetches texels and the same texels are written via fragment shader outputs, even if the reads and writes are not in the same draw call, unless any of the following exceptions apply: • The reads and writes are from/to disjoint sets of texels (after accounting for texture filtering rules). • There is only a single read and write of each texel, and the read is in the fragment shader invocation that writes the same texel (e.g. using texelFetch2D(sampler, ivec2(gl_FragCoord.xy), 0);). • If a texel has been written, then in order to safely read the result a texel fetch must be in a subsequent draw call separated by the command void TextureBarrier( void ); TextureBarrier will guarantee that writes have completed and caches have been invalidated before subsequent draw calls are executed.

9.3.2

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 (in the same fashion that overlapping copies via BlitFramebuffer are undefined). Specifically, the values of copied texels are undefined if all of the following conditions are true:

OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

327

• 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 18.2.1) 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 RED, RG, RGB, RGBA, or one of the sized internal formats from table 8.12 whose “CR” (color-renderable) column is checked in that table No other formats, including compressed internal formats, are color-renderable. • An internal format is depth-renderable if it is DEPTH_COMPONENT or one of the formats from table 8.13 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 STENCIL_INDEX, DEPTH_STENCIL, or one of the formats from table 8.13 whose base internal format is STENCIL_INDEX or DEPTH_STENCIL. No other formats are stencilrenderable.

OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

9.4.1

328

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, one- or 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, one- or 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. • 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. OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

329

• 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. The framebuffer object bound to target is said to be framebuffer complete if all the following conditions are true: • if the default framebuffer is bound to target, 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. { FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT } • The combination of internal formats of the attached images does not violate an implementation-dependent set of restrictions. { 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 } OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

330

• 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, one- or two-dimensional 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. • 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 1.3.6. OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

331

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 a framebuffer object or default framebuffer may be queried with the commands enum CheckFramebufferStatus( enum target ); enum CheckNamedFramebufferStatus( uint framebuffer, enum target ); For CheckFramebufferStatus, the framebuffer object is that bound to target. For CheckNamedFramebufferStatus, framebuffer is the name of the framebuffer object. target must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER, or FRAMEBUFFER. FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. If framebuffer is zero, then the status of the default read or draw framebuffer (as determined by target) is returned. A value is returned that identifies whether or not the framebuffer object or default framebuffer is complete when treated as a read or draw framebuffer (as determined by target). If the framebuffer object is complete, then FRAMEBUFFER_COMPLETE is returned. Otherwise, the value returned is one of the error codes defined at 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. An INVALID_OPERATION error is generated by CheckNamedFrame-

OpenGL 4.6 (Core Profile) - May 5, 2022

9.4. FRAMEBUFFER COMPLETENESS

332

bufferStatus if framebuffer is not the name of an existing framebuffer object.

9.4.3

Required Framebuffer Formats

Implementations must support framebuffer objects with up to the value of 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 (although implementations are not required to support creation of attachments in all color-renderable formats). 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 are only required to support framebuffer objects where both attachments refer to the same image. Any internal format that returns FULL_SUPPORT from the GetInternalformativ query for pname FRAMEBUFFER_RENDERABLE (see section 22.3.1) can be used in non-layered attachments in any combination with other required formats or formats that offer FULL_SUPPORT. Any internal format that returns FULL_SUPPORT from the query for pname FRAMEBUFFER_RENDERABLE_LAYERED can be used in any combination with other required formats or formats that offer FULL_SUPPORT. There must be at least one default framebuffer format allowing creation of a default framebuffer supporting front-buffered rendering.

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 4.6 (Core Profile) - May 5, 2022

9.5. MAPPING BETWEEN PIXEL AND ELEMENT IN ATTACHED IMAGE333

9.4.5

Effects of Framebuffer State on Framebuffer Dependent Values

The values of the state variables listed in table 23.73 may change when a change is made to the current read or draw framebuffer binding, to the state of a currently bound framebuffer object, or to an image attached to that framebuffer object. Most such state is dependent on the draw framebuffer (the value DRAW_FRAMEBUFFER_BINDING), but IMPLEMENTATION_COLOR_READ_TYPE and IMPLEMENTATION_COLOR_READ_FORMAT are dependent on the read framebuffer (the value of READ_FRAMEBUFFER_BINDING). The values of the state variables listed in table 23.73 may change when a change is made to DRAW_FRAMEBUFFER_BINDING, to the state of the currently bound draw framebuffer object, or to an image attached to that framebuffer object. When DRAW_FRAMEBUFFER_BINDING is zero, the values of the state variables listed in table 23.73 are implementation-defined. When DRAW_FRAMEBUFFER_BINDING is non-zero, if the currently bound draw framebuffer object is not framebuffer complete, then the values of the state variables listed in table 23.73 are undefined. When DRAW_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 23.73 are completely determined by DRAW_FRAMEBUFFER_BINDING, the state of the currently bound draw framebuffer object, and the state of the images attached to that framebuffer object. The actual sizes of the color, depth, or stencil bit planes can be obtained by querying an attachment point using Get*FramebufferAttachmentParameteriv, 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. If the value of FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE at a particular attachment point is TEXTURE, the sizes may be determined by calling GetTexParameter, as described in section 8.11.

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 4.6 (Core Profile) - May 5, 2022

9.6. CONVERSION TO FRAMEBUFFER-ATTACHABLE IMAGE COMPONENTS334 (xw , yw ) correspond 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 8.3 as follows: i = xw j = yw k = layer layer is the value of FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER for the selected logical buffer. For a two-dimensional texture, k and layer are irrelevant; for a one-dimensional texture, j, k, and layer are irrelevant. (xw , yw ) corresponds to a border texel if xw , yw , or layer is less than zero, or if xw , yw , or layer is greater than or equal to the width, height, or depth, respectively, of the texture image.

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 8.11, 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 17.4.2 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 a logical operation or for blending while the draw framebuffer binding is non-zero, components of that color taken from the framebufferattachable image attached to the selected logical buffer are first converted to R, G, B, and A values according to table 15.1 and the internal format of the attached image.

OpenGL 4.6 (Core Profile) - May 5, 2022

9.8. LAYERED FRAMEBUFFERS

9.8

335

Layered Framebuffers

A framebuffer is considered to be layered if it is complete and all of its populated attachments are layered. 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 • 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. 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 17.3.3), depth buffer test (see section 17.3.4), and for blending and color buffer writes (see section 17.3.6). 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 17.4.3 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 9.3. When cube map array texture levels are attached to a layered framebuffer, the layer number corresponds to a layer-face. The layer-face can be translated into an array layer and a cube map face by layer array layer = 6 f ace = layer mod 6 OpenGL 4.6 (Core Profile) - May 5, 2022

9.8. LAYERED FRAMEBUFFERS Layer Number 0 1 2 3 4 5

336

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

. The face number corresponds to the cube map faces as shown in table 9.3.

OpenGL 4.6 (Core Profile) - 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 vertices of a primitive using DrawArrays or one of the other drawing commands defined in section 10.4. 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.

337

338

Coordinates

Vertex Shader Execution

Shaded Vertices

Point, Line Segment, or Triangle (Primitive) Assembly

Point culling, Line Segment or Triangle clipping

Varying Outputs

Generic Vertex Attributes

Primitive type (from DrawArrays or DrawElements mode)

Figure 10.1. Vertex processing and primitive assembly.

OpenGL 4.6 (Core Profile) - May 5, 2022

Rasterization

10.1. PRIMITIVE TYPES

10.1

339

Primitive Types

A sequence of vertices is passed to the GL using DrawArrays or one of the other drawing commands defined in section 10.4. 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, together with any additional state required when assembling primitives from multiple vertices.

10.1.1

Points

A series of individual points are specified with mode POINTS. Each vertex defines a separate point. No state is required for points, since each point is independent of any previous and following points.

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 OpenGL 4.6 (Core Profile) - May 5, 2022

10.1. PRIMITIVE TYPES

340

2 3

6 4 4 5

(a)

5 1

(b)

(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.

but it is used differently: a processed vertex holding the first vertex of the current segment, and a boolean flag indicating whether the current vertex is odd or even (a segment start or end).

10.1.5 This subsection is only defined in the compatibility profile.

10.1.6

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,

OpenGL 4.6 (Core Profile) - May 5, 2022

10.1. PRIMITIVE TYPES

341

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).

10.1.7

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.

10.1.8

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.

10.1.9 This subsection is only defined in the compatibility profile.

10.1.10 This subsection is only defined in the compatibility profile.

10.1.11

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 (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).

OpenGL 4.6 (Core Profile) - May 5, 2022

10.1. PRIMITIVE TYPES

342

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.

OpenGL 4.6 (Core Profile) - May 5, 2022

10.1. PRIMITIVE TYPES

343

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.

10.1.12

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).

10.1.13

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, OpenGL 4.6 (Core Profile) - May 5, 2022

10.1. PRIMITIVE TYPES

344

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.

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).

10.1.14

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.

OpenGL 4.6 (Core Profile) - 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)

345 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.

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.

10.1.15

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, 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.

OpenGL 4.6 (Core Profile) - 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)

346 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).

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.

10.1.16

General Considerations For Polygon Primitives

Depending on the current state of the GL, a polygon primitive generated from a drawing command with mode TRIANGLE_FAN, TRIANGLE_STRIP, TRIANGLES, TRIANGLES_ADJACENCY, or TRIANGLE_STRIP_ADJACENCY may be rendered in one of several ways, such as outlining its border or filling its interior. The order of vertices in such a primitive is significant in polygon rasterization (see section 14.6.1) and fragment shading (see section 15.2.2).

10.1.17 This subsection is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - May 5, 2022

10.2. CURRENT VERTEX ATTRIBUTE VALUES

10.2

347

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}{sfd}( uint index, T values ); void VertexAttrib{123}{sfd}v( uint index, const T *values ); void VertexAttrib4{bsifd ub us ui}v( uint index, const T *values ); void VertexAttrib4Nub( uint index, ubyte x, ubyte y, ubyte z, ubyte w ); void VertexAttrib4N{bsi ub us ui}v( uint index, const T *values ); void VertexAttribI{1234}{i ui}( uint index, T values ); void VertexAttribI{1234}{i ui}v( uint index, const T *values ); void VertexAttribI4{b s ub us}v( uint index, const T *values ); void VertexAttribL{1234}d( uint index, const T values ); void VertexAttribL{1234}dv( uint index, const T *values ); void VertexAttribP{1234}ui(uint index,enum type,boolean normalized,uint value); void VertexAttribP{1234}uiv(uint index,enum type,boolean normalized,const uint *value); The VertexAttrib4N* commands specify fixed-point values that are converted

OpenGL 4.6 (Core Profile) - May 5, 2022

10.2. CURRENT VERTEX ATTRIBUTE VALUES

348

to a normalized [0, 1] or [−1, 1] range as described in equations 2.1 and 2.2, respectively. 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. The VertexAttribL* commands specify double-precision values that will be stored as double-precision values. The VertexAttribP* commands specify up to four attribute component values packed into a single natural type type as described in section 10.3.8. type must be INT_2_10_10_10_REV, UNSIGNED_INT_2_10_10_10_REV, or UNSIGNED_INT_10F_11F_11F_REV, specifying signed, unsigned, or unsigned floating-point data, respectively. The first one (x), two (x, y), three (x, y, z), or four (x, y, z, w) components of the packed data are consumed by VertexAttribP1ui, VertexAttribP2ui, VertexAttribP3ui, and VertexAttribP4ui, respectively. If normalized is TRUE, signed or unsigned components are converted to floating-point by normalizing to [−1, 1] or [0, 1] respectively. If normalized is FALSE, signed and unsigned components are directly cast to floating-point. For floating-point formats, normalized is ignored. The number of components specified must be no greater than the number of components in the packed type. For VertexAttribP*uiv, value contains the address of a single uint containing the packed attribute components. 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. 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]* or VertexAttribP*, for single-precision floating-point OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

349

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 • VertexAttribL*, for double-precision floating-point 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. An INVALID_ENUM error is generated by VertexAttribP4ui and VertexAttribP4uiv if type is UNSIGNED_INT_10F_11F_11F_REV.

10.2.2 This subsection is only defined in the compatibility profile.

10.2.3

Vertex Attribute Queries

Current generic vertex attribute values may be queried using the GetVertexAttrib* commands as described in section 10.5.

10.2.4

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 server’s address space (described in section 10.3.9). Blocks of data in these arrays may then be used to specify multiple geometric primitives through the execution of a single GL command. All of the state required to represent the vertex arrays is stored in a vertex array object.

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

10.3.1

350

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. 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 no vertex array object is bound. 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. 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 tables 23.3 and 23.4. OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

351

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 22). Errors An INVALID_OPERATION error is generated if array is not zero or a name returned from a previous call to CreateVertexArrays or GenVertexArrays, or if such a name has since been deleted with DeleteVertexArrays. An INVALID_OPERATION error is generated by any commands which modify, draw from, or query vertex array state when no vertex array is bound. This occurs in the initial GL state, and may occur as a result of BindVertexArray or a side effect of DeleteVertexArrays. Vertex array objects may also be created with the command void CreateVertexArrays( sizei n, uint *arrays ); CreateVertexArrays returns n previously unused vertex array object names in arrays, each representing a state vector comprising all the state and with the same initial values listed in tables 23.3 and 23.4. Errors An INVALID_VALUE error is generated if n is negative. 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. To bind a buffer object to the element array buffer bind point of a vertex array object, use the command void VertexArrayElementBuffer( uint vaobj, uint buffer );

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

352

vaobj is the name of the vertex array object, and buffer is zero or the name of the buffer object. If buffer is zero, any existing element array buffer binding to vaobj is removed. Errors An INVALID_OPERATION error is generated if vaobj is not the name of an existing vertex array object. An INVALID_OPERATION error is generated if buffer is not zero or the name of an existing buffer object.

10.3.2

Specifying Arrays for Generic Vertex Attributes

To specify the organization of arrays storing generic vertex attributes of a vertex array object, use the commands The commands void VertexAttribFormat( uint attribindex, int size, enum type, boolean normalized, uint relativeoffset ); void VertexAttribIFormat( uint attribindex, int size, enum type, uint relativeoffset ); void VertexAttribLFormat( uint attribindex, int size, enum type, uint relativeoffset ); void VertexArrayAttribFormat( uint vaobj, uint attribindex, int size, enum type, boolean normalized, uint relativeoffset ); void VertexArrayAttribIFormat( uint vaobj, uint attribindex, int size, enum type, uint relativeoffset ); void VertexArrayAttribLFormat( uint vaobj, uint attribindex, int size, enum type, uint relativeoffset ); For VertexAttrib*Format, the vertex array object is that bound to VERTEX_ARRAY_BINDING. For VertexArrayAttrib*Format, vaobj is the name of the vertex array object. attribindex identifies the generic vertex attribute array. size indicates the number of values per vertex that are stored in the array, as well as their component ordering. type specifies the data type of the values stored in the array. Table 10.3 indicates the allowable values for size and type. A type of BYTE, UNSIGNED_BYTE, SHORT, UNSIGNED_SHORT, INT, UNSIGNED_INT, FLOAT, HALF_FLOAT, or DOUBLE indicates the corresponding GL data type shown in

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

353

table 8.2.

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. A type of UNSIGNED_INT_10F_11F_11F_REV indicates two unsigned 11-bit floatingpoint elements and one unsigned 10-bit floating-point element packed into a single uint. Encoding of the unsigned 11- and 10-bit floating-point values is described in sections 2.3.4.3 and 2.3.4.4, respectively. The types INT_2_10_10_10_REV, UNSIGNED_INT_2_10_10_10_REV and UNSIGNED_INT_10F_11F_11F_REV all correspond to the term packed in table 10.3. The components are packed as shown in table 8.8. 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 types are handled. “cast” means that they are converted to floating-point directly. “normalize” means that they 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. “integer” means that they remain as integer values. “flag” means that either “cast” or “normalized” applies, depending on whether the normalized flag to the command is TRUE or FALSE, respectively. The normalized flag is ignored for floating-point data types, including fixed, float, half, double, and any packed types that have floating-point components. If size is BGRA, vertex array values are always normalized, irrespective of the “normalize” table entry. If type is UNSIGNED_INT_10F_11F_11F_REV, vertex array values are never normalized, irrespective of the “normalize” table entry. 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_OPERATION error is generated by VertexAttrib*Format if no vertex array object is currently bound (see section 10.3.1); An INVALID_OPERATION error is generated by VertexArrayAttrib*Format if vaobj is not the name of an existing vertex array object. 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 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

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

354

Command VertexAttribFormat

sizes and Component Ordering 1, 2, 3, 4, BGRA

Integer Handling flag

VertexAttribIFormat

1, 2, 3, 4

integer

VertexAttribLFormat

1, 2, 3, 4

n/a

types byte, ubyte, short, ushort, int, uint, fixed, float, half, double, packed byte, ubyte, short, ushort, int, uint double

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.

token names from table 8.2 corresponding to one of the allowed GL data types for that command as shown in table 10.3. An INVALID_ENUM error is generated by VertexAttribIFormat and VertexAttribLFormat if type is UNSIGNED_INT_10F_11F_11F_REV. An INVALID_OPERATION error is generated under any of the following conditions: • size is BGRA and type is not UNSIGNED_BYTE, INT_2_10_10_10_REV or UNSIGNED_INT_2_10_10_10_REV; • type is INT_2_10_10_10_REV or UNSIGNED_INT_2_10_10_10_REV, and size is neither 4 nor BGRA; • type is UNSIGNED_INT_10F_11F_11F_REV and size is not 3; • size is BGRA and normalized is FALSE. An INVALID_VALUE error is generated if relativeoffset is larger than the value of MAX_VERTEX_ATTRIB_RELATIVE_OFFSET. The source of data for a generic vertex attribute may be determined by attaching a buffer object to a vertex array object with the commands void BindVertexBuffer( uint bindingindex, uint buffer, intptr offset, sizei stride ); void VertexArrayVertexBuffer( uint vaobj, OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

355

uint bindingindex, uint buffer, intptr offset, sizei stride ); For BindVertexBuffer, the vertex array object is the currently bound vertex array object. For VertexArrayVertexBuffer, vaobj is the name of the vertex array object. buffer is either zero or a name returned by GenBuffers or CreateBuffers. If buffer is zero, any buffer object bound to bindingindex is detached. If buffer is not the name of an existing buffer object, the GL first 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. buffer is then attached to the specified bindingindex of the vertex array object. When sourcing vertex data from the buffer object, offset specifies the offset in basic machine units of the first element in the vertex buffer. 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. If the operation is successful no change is made to the state of the newly bound buffer object, and any previous binding to bindingindex is broken. Errors An INVALID_OPERATION error is generated by BindVertexBuffer if no vertex array object is bound. An INVALID_OPERATION error is generated by VertexArrayVertexBuffer if vaobj is not the name of an existing vertex array object. An INVALID_OPERATION error is generated if buffer is not zero or a name returned from a previous call to GenBuffers or CreateBuffers, 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. The source of data for multiple vertex attributes may be determined by attaching multiple existing buffer objects to a vertex array object with the commands void BindVertexBuffers( uint first, sizei count, const uint *buffers, const intptr *offsets, const sizei *strides );

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

356

void VertexArrayVertexBuffers( uint vaobj, uint first, sizei count, const uint *buffers, const intptr *offsets, const sizei *strides ); For BindVertexBuffers, the vertex array object is the currently bound vertex array object. For VertexArrayVertexBuffers, vaobj is the name of the vertex array object. count existing buffer objects are bound to vertex buffer binding points numbered first through f irst + count − 1. If buffers is not NULL, it specifies an array of count values, each of which must be zero or the name of an existing buffer object. offsets and strides specify arrays of count values indicating the offset of the first element and stride between elements in each buffer, respectively. If buffers is NULL, each affected vertex buffer binding point from first through f irst+count−1 will be reset to have no bound buffer object. In this case, the offsets and strides associated with the binding points are set to default values, ignoring offsets and strides. BindVertexBuffers is equivalent (assuming no errors are generated) to: for (i = 0; i < count; i++) { if (buf f ers == NULL) { BindVertexBuffer(f irst + i, 0, 0, 16); } else { BindVertexBuffer(f irst + i, buf f ers[i], of f sets[i], strides[i]); } } except that buffers will not be created if they do not exist. VertexArrayVertexBuffers is equivalent to the pseudocode above, but replacing BindVertexBuffer(args) with VertexArrayVertexBuffers(vaobj, args). The values specified in buffers, offsets, and strides will be checked separately for each vertex buffer binding point. When a value for a specific vertex buffer binding point is invalid, the state for that binding point will be unchanged and an error will be generated. However, state for other vertex buffer binding points will still be changed if their corresponding values are valid. Errors An INVALID_OPERATION error is generated by BindVertexBuffers if no vertex array object is bound.

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS An INVALID_OPERATION error is generated by VertexArrayVertexBuffers if vaobj is not the name of an existing vertex array object. An INVALID_OPERATION error is generated if f irst + count is greater than the value of MAX_VERTEX_ATTRIB_BINDINGS. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if any value in buffers is not zero or the name of an existing buffer object (per binding). An INVALID_VALUE error is generated if any value in offsets or strides is negative, or if any value in strides is greater than the value of MAX_VERTEX_ATTRIB_STRIDE (per binding). 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 ); void VertexArrayAttribBinding( uint vaobj, uint attribindex, uint bindingindex ); For VertexAttribBinding, the vertex array object is the currently bound vertex array object. For VertexArrayAttribBinding, vaobj is the name of the vertex array object. Errors An INVALID_OPERATION error is generated by VertexArrayAttribBinding if vaobj is not the name of an existing vertex array object. 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 no 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. When size is BGRA, it indicates four values. The values within each array element are stored sequentially in memory. However, if size is BGRA, the first, second, third, and fourth values of each array element are taken from the third, second, first, and fourth values in memory respectively. The commands

OpenGL 4.6 (Core Profile) - May 5, 2022

357

10.3. VERTEX ARRAYS 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 ); void VertexAttribLPointer( 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 (assuming no errors are generated) to: VertexAttrib*Format(index, size, type, {normalized, }, 0); VertexAttribBinding(index, index); if (stride != 0) { effectiveStride = stride; } else { compute effectiveStride based on size and type; } VERTEX_ATTRIB_ARRAY_STRIDE[index] = stride; // This sets VERTEX_BINDING_STRIDE to effectiveStride VERTEX_ATTRIB_ARRAY_POINTER[index] = pointer; BindVertexBuffer(index, buffer bound to ARRAY_BUFFER, (char *)pointer - (char *)NULL, effectiveStride); If stride is specified as zero, then array elements are stored sequentially. Errors An INVALID_VALUE error is generated if stride is negative. 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 no buffer is bound to ARRAY_BUFFER, and pointer is not NULL. An INVALID_OPERATION error is generated if no vertex array object is bound. 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 in a vertex array object is enabled with the commands OpenGL 4.6 (Core Profile) - May 5, 2022

358

10.3. VERTEX ARRAYS

359

void EnableVertexAttribArray( uint index ); void EnableVertexArrayAttrib( uint vaobj, uint index ); and is disabled with the commands void DisableVertexAttribArray( uint index ); void DisableVertexArrayAttrib( uint vaobj, uint index ); index identifies the generic vertex attribute array to enable or disable. For EnableVertexAttribArray and DisableVertexAttribArray, the vertex array object is the currently bound vertex array object. For EnableVertexArrayAttrib and DisableVertexArrayAttrib, vaobj is the name of the vertex array object. Errors An INVALID_OPERATION error is generated by EnableVertexAttribArray and DisableVertexAttribArray if no vertex array object is bound. An INVALID_OPERATION error is generated by EnableVertexArrayAttrib and DisableVertexArrayAttrib if vaobj is not the name of an existing vertex array object. An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS.

10.3.3 This subsection is only defined in the compatibility profile.

10.3.4

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 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 divisor value for attributes taken from a vertex array object is set with the commands void VertexBindingDivisor( uint bindingindex, uint divisor );

OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS void VertexArrayBindingDivisor( uint vaobj, uint bindingindex, uint divisor ); For VertexBindingDivisor, the vertex array object is the currently bound vertex array object. For VertexArrayBindingDivisor, vaobj is the name of the vertex array object. These commands set the divisor for the buffer bound to the specified bindingindex of the vertex array object to divisor. Errors An INVALID_OPERATION error is generated by VertexBindingDivisor if no vertex array object is bound. An INVALID_OPERATION error is generated by VertexArrayBindingDivisor if vaobj is not the name of an existing vertex array object. An INVALID_VALUE error is generated if bindingindex is greater than or equal to the value of MAX_VERTEX_ATTRIB_BINDINGS. The command void VertexAttribDivisor( uint index, uint divisor ); is equivalent to (assuming no errors are generated): 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 no vertex array object is bound.

10.3.5

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 4.6 (Core Profile) - May 5, 2022

360

10.3. VERTEX ARRAYS

10.3.6

361

Primitive Restart

Primitive restart is enabled or disabled by calling one of the commands void Enable( enum target ); and void Disable( enum target ); with target PRIMITIVE_RESTART. The command void PrimitiveRestartIndex( uint index ); specifies a vertex array element that is treated specially when primitive restart is enabled. This value is called the primitive restart index. When one of the *DrawElements* commands 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 the primitive restart index, 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 10.4 is used, the primitive restart comparison occurs before the basevertex offset is added to the array index. Primitive restart can also be enabled or disabled with a target of PRIMITIVE_RESTART_FIXED_INDEX. In this case, the primitive restart index is equal to 2N − 1, where N is 8, 16 or 32 if the type is UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT, respectively, and the index value specified by PrimitiveRestartIndex is ignored. If both PRIMITIVE_RESTART and PRIMITIVE_RESTART_FIXED_INDEX are enabled, the index value determined by PRIMITIVE_RESTART_FIXED_INDEX is used. Note that primitive restart is not performed for array elements transferred by any drawing command not taking a type parameter, including all of the *Draw* commands other than *DrawElements*. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

10.3.7

362

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 10.3.10) that reference vertex data that lies outside the enabled attribute’s vertex buffer object result in reading zero. It is not possible to read vertex data from outside the enabled vertex buffer objects or from another GL context, and these accesses do not result in abnormal program termination.

10.3.8

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. Tables 10.4 and 10.5 describe 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

Table 10.4: Packed component layout for non-BGRA formats. Bit numbers are indicated for each component. 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 w

Table 10.5: Packed component layout for BGRA format. Bit numbers are indicated for each component.

Vertex data format UNSIGNED_INT_10F_11F_11F_REV describes a packed, 3-component format that is stored in a single 32-bit word. The first (x), and secOpenGL 4.6 (Core Profile) - May 5, 2022

10.3. VERTEX ARRAYS

363

ond (y) components are represented as 11-bit unsigned floating-point values, and the third (z) component is represented as a 10-bit unsigned floating-point value. Table 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 z

Table 10.6: Packed component layout for UNSIGNED_INT_10F_11F_11F_REV format. Bit numbers are indicated for each component.

10.3.9

Vertex Arrays in Buffer Objects

Blocks of vertex array data are 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 10.4 operate as previously defined, where data for enabled generic attribute arrays are sourced from buffer objects. 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 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. If any enabled array’s buffer binding is zero when DrawArrays or one of the other drawing commands defined in section 10.4 is called, the result is undefined.

10.3.10

Array Indices in Buffer Objects

Blocks of array indices are stored in buffer objects in the formats described by the type parameter of DrawElements (see section 10.4). OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

364

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. DrawElements, DrawRangeElements, and DrawElementsInstanced source their indices from the buffer object whose name is bound to ELEMENT_ARRAY_BUFFER, using their indices parameters as offsets into the buffer object in the same fashion as described in section 10.3.9. 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. Finally, MultiDrawElements and MultiDrawElementsBaseVertex also source their indices from that buffer object, using its indices parameter as a pointer to an array of pointers that represent offsets into the buffer object. If zero is bound to ELEMENT_ARRAY_BUFFER, the result of these drawing commands is undefined. 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.

10.3.11

Indirect Commands in Buffer Objects

Arguments to the indirect commands DrawArraysIndirect, DrawElementsIndirect, MultiDrawArraysIndirect, and MultiDrawElementsIndirect (see section 10.4), and to DispatchComputeIndirect (see section 19) are sourced from the buffer object currently bound to the corresponding indirect buffer target (see table 10.7), using the command’s indirect parameter as an offset into the buffer object in the same fashion as described in section 10.3.9. 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

Drawing Commands Using Vertex Arrays

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

OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS Indirect Command Name DrawArraysIndirect DrawElementsIndirect MultiDrawArraysIndirect MultiDrawElementsIndirect DispatchComputeIndirect

365

Indirect Buffer target DRAW_INDIRECT_BUFFER DRAW_INDIRECT_BUFFER DRAW_INDIRECT_BUFFER DRAW_INDIRECT_BUFFER DISPATCH_INDIRECT_BUFFER

Table 10.7: Indirect commands and corresponding indirect buffer targets.

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 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

OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

366

INVALID_VALUE error is recommended in this case. An INVALID_VALUE error is generated if count is negative. An INVALID_OPERATION error is generated if no vertex array object is

bound (see section 10.3.1), The command void DrawArrays( enum mode, int first, sizei count ); is equivalent to DrawArraysOneInstance(mode, f irst, count, 0, 0); The command void DrawArraysInstancedBaseInstance( enum mode, int first, sizei count, sizei instancecount, uint baseinstance ); 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. Additionally, the first element within those instanced vertex attributes is specified in baseinstance. DrawArraysInstancedBaseInstance is equivalent (assuming no errors are generated) to: if (mode, count, or instancecount is invalid) generate appropriate error else { for (i = 0; i < instancecount; i++) { DrawArraysOneInstance(mode, f irst, count, i, baseinstance); } } The command void DrawArraysInstanced( enum mode, int first, sizei count, sizei instancecount ); is equivalent to OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

367

DrawArraysInstancedBaseInstance(mode, f irst, count, instancecount, 0); The command void DrawArraysIndirect( enum mode, const void *indirect ); is equivalent to typedef struct { uint count; uint instanceCount; uint first; uint baseInstance; } DrawArraysIndirectCommand; DrawArraysIndirectCommand *cmd = (DrawArraysIndirectCommand *)indirect; DrawArraysInstancedBaseInstance(mode, cmd->first, cmd->count, cmd->instanceCount, cmd->baseInstance); Unlike DrawArraysInstanced, first is unsigned and cannot cause an error. Errors An INVALID_OPERATION error is generated if zero is bound to DRAW_INDIRECT_BUFFER. 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. All elements of DrawArraysIndirectCommand are tightly packed 32-bit values. The command void MultiDrawArrays( enum mode, const int *first, const sizei *count, sizei drawcount ); behaves identically to DrawArrays except that drawcount separate ranges of elements are specified instead, all elements are treated as though they are not instanced, and the value of instance remains zero. It is equivalent (assuming no errors are generated) to: OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

368

if (mode or drawcount is invalid) generate appropriate error else { for (i = 0; i < drawcount; i++) { if (count[i] > 0) DrawArraysOneInstance(mode, f irst[i], count[i], 0, 0); } } The index of the draw (i in the above pseudo-code) may be read by a vertex shader as gl_DrawID, as described in section 11.1.3.9. The command void MultiDrawArraysIndirect( enum mode, const void *indirect, sizei drawcount, sizei stride ); behaves identically to DrawArraysIndirect except that indirect is treated as an array of drawcount DrawArraysIndirectCommand structures. indirect contains the offset of the first element of the array within the buffer currently bound to the DRAW_INDIRECT buffer binding. stride specifies the distance, in basic machine units, between the elements of the array. If stride is zero, the array elements are treated as tightly packed. It is equivalent (assuming no errors are generated) to: if (mode is invalid) generate appropriate error else { const ubyte *ptr = (const ubyte *)indirect; for (i = 0; i < drawcount; i++) { DrawArraysIndirect(mode, (DrawArraysIndirectCommand*)ptr); if (stride == 0) { ptr += sizeof(DrawArraysIndirectCommand); } else { ptr += stride; } } } The index of the draw (i in the above pseudo-code) may be read by a vertex shader as gl_DrawID, as described in section 11.1.3.9. OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

369

Errors In addition to errors that would be generated by DrawArraysIndirect: An INVALID_VALUE error is generated if stride is negative. An INVALID_VALUE error is generated if stride is neither zero nor a multiple of four. An INVALID_VALUE error is generated if drawcount is not positive. The command void MultiDrawArraysIndirectCount( enum mode, const void *indirect, intptr drawcount, intptr maxdrawcount, sizei stride ); behaves similarly to MultiDrawArraysIndirect, except that drawcount defines an offset (in bytes) into the buffer object bound to the PARAMETER_BUFFER binding point at which a single sizei typed value is stored, which contains the draw count. maxdrawcount specifies the maximum number of draws that are expected to be stored in the buffer. If the value stored at drawcount into the buffer is greater than maxdrawcount, the implementation stops processing draws after maxdrawcount parameter sets. drawcount must be a multiple of four. Errors In addition to errors that would be generated by MultiDrawArraysIndirect: An INVALID_OPERATION error is generated if no buffer is bound to the PARAMETER_BUFFER binding point. An INVALID_VALUE error is generated if drawcount is not a multiple of four. An INVALID_OPERATION error is generated if reading a sizei typed value from the buffer bound to the PARAMETER_BUFFER target at the offset specified by drawcount would result in an out-of-bounds access. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

370

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. 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. 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 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). GL implementations do not restrict index values; any value representable in a uint may be used. However, for compatibility with OpenGL ES implementations, the maximum representable index vaue may be queried by calling GetInteger64v with pname MAX_ELEMENT_INDEX, and will return 232 − 1. 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_VALUE error is generated if count is negative. An INVALID_ENUM error is generated if type is not UNSIGNED_BYTE, UNSIGNED_SHORT, or UNSIGNED_INT. An INVALID_OPERATION error is generated if no vertex array object is

OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

371

bound (see section 10.3.1), 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 if (mode, count or type is invalid) generate appropriate error else DrawElementsOneInstance(mode, count, type, indices, 0, 0, 0); The command void DrawElementsInstancedBaseInstance( enum mode, sizei count, enum type, const void *indices, sizei instancecount, uint baseinstance ); 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 DrawArraysInstancedBaseInstance, and baseinstance has the same effect. It is equivalent (assuming no errors are generated) to: if (mode, count, type, or instancecount is invalid) generate appropriate error else { for (int i = 0; i < instancecount; i++) { DrawElementsOneInstance(mode, count, type, indices, i, 0, baseinstance); } } The command OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

372

void DrawElementsInstanced( enum mode, sizei count, enum type, const void *indices, sizei instancecount ); behaves identically to DrawElementsInstancedBaseInstance except that baseinstance is zero. It is equivalent to DrawElementsInstancedBaseInstance(mode, count, type, indices, instancecount, 0); The command void MultiDrawElements( enum mode, const sizei *count, enum type, const void * const *indices, sizei drawcount ); behaves identically to DrawElementsInstanced except that drawcount separate sets of elements are specified instead, all elements are treated as though they are not instanced, and the value of instance remains zero. It is equivalent (assuming no errors are generated) to: if (mode, drawcount, or type is invalid) generate appropriate error else { for (int i = 0; i < drawcount; i++) DrawElementsOneInstance(mode, count[i], type, indices[i], 0, 0, 0); } The index of the draw (i in the above pseudo-code) may be read by a vertex shader as gl_DrawID, as described in section 11.1.3.9. 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 pnames MAX_ELEMENTS_VERTICES and MAX_ELEMENTS_INDICES. If end − start + 1 is OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

373

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. Errors An INVALID_VALUE error is generated if end < start. An INVALID_VALUE error is generated if count is negative. 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 ); void DrawElementsInstancedBaseVertexBaseInstance( enum mode, sizei count, enum type, const void *indices, sizei instancecount, int basevertex, uint baseinstance ); 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

OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

374

void DrawElementsIndirect( enum mode, enum type, const void *indirect ); is equivalent to typedef struct { uint count; uint instanceCount; uint firstIndex; int baseVertex; uint baseInstance; } DrawElementsIndirectCommand; if (no element array buffer is bound) { generate appropriate error } else { DrawElementsIndirectCommand *cmd = (DrawElementsIndirectCommand *)indirect; DrawElementsInstancedBaseVertexBaseInstance(mode, cmd->count, type, cmd->firstIndex * size-of-type, cmd->instanceCount, cmd->baseVertex, cmd->baseInstance); }

Errors An INVALID_OPERATION error is generated if zero is bound to DRAW_INDIRECT_BUFFER, or if no element array buffer is bound. 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. All elements of DrawElementsIndirectCommand are tightly packed. The command void MultiDrawElementsIndirect( enum mode, enum type, const void *indirect, sizei drawcount, sizei stride ); OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

375

behaves identically to DrawElementsIndirect except that indirect is treated as an array of drawcount DrawElementsIndirectCommand structures. indirect contains the offset of the first element of the array within the buffer currently bound to the DRAW_INDIRECT buffer binding. stride specifies the distance, in basic machine units, between the elements of the array. If stride is zero, the array elements are treated as tightly packed. It is equivalent (assuming no errors are generated) to: if (mode or type is invalid) generate appropriate error else { const ubyte *ptr = (const ubyte *)indirect; for (i = 0; i < drawcount; i++) { DrawElementsIndirect(mode, type, (DrawElementsIndirectCommand*)ptr); if (stride == 0) { ptr += sizeof(DrawElementsIndirectCommand); } else { ptr += stride; } } } The index of the draw (i in the above pseudo-code) may be read by a vertex shader as gl_DrawID, as described in section 11.1.3.9. Errors In addition to errors that would be generated by DrawElementsIndirect: An INVALID_VALUE error is generated if stride is negative. An INVALID_VALUE error is generated if stride is neither zero nor a multiple of four. An INVALID_VALUE error is generated if drawcount is not positive. The command void MultiDrawElementsIndirectCount( enum mode, enum type, const void *indirect, intptr drawcount, sizei maxdrawcount, sizei stride ); behaves similarly to MultiDrawElementsIndirect, except that drawcount defines an offset (in bytes) into the buffer object bound to the PARAMETER_BUFFER OpenGL 4.6 (Core Profile) - May 5, 2022

10.4. DRAWING COMMANDS USING VERTEX ARRAYS

376

binding point at which a single sizei typed value is stored, which contains the draw count. maxdrawcount specifies the maximum number of draws that are expected to be stored in the buffer. If the value stored at drawcount into the buffer is greater than maxdrawcount, the implementation stops processing draws after maxdrawcount parameter sets. drawcount must be a multiple of four. Errors In addition to errors that would be generated by MultiDrawElementsIndirect: An INVALID_OPERATION error is generated if no buffer is bound to the PARAMETER_BUFFER binding point. An INVALID_VALUE error is generated if drawcount is not a multiple of four. An INVALID_VALUE error is generated if maxdrawcount is negative. An INVALID_OPERATION error is generated if reading a sizei typed value from the buffer bound to the PARAMETER_BUFFER target at the offset specified by drawcount would result in an out-of-bounds access. The command void MultiDrawElementsBaseVertex( enum mode, const sizei *count, enum type, const void * const *indices, sizei drawcount, const int *basevertex ); behaves identically to DrawElementsBaseVertex, except that drawcount separate lists of elements are specified instead. It is equivalent (assuming no errors are generated) to: if (mode or drawcount is invalid) generate appropriate error else { for (int i = 0; i < drawcount; i++) if (count[i] > 0) DrawElementsBaseVertex(mode, count[i], type, indices[i], basevertex[i]); } The index of the draw (i in the above pseudo-code) may be read by a vertex shader as gl_DrawID, as described in section 11.1.3.9. OpenGL 4.6 (Core Profile) - May 5, 2022

10.5. VERTEX ARRAY AND VERTEX ARRAY OBJECT QUERIES

377

Errors In addition to errors that would be generated by DrawElementsBaseVertex: An INVALID_VALUE error is generated if drawcount is negative.

10.4.1 This subsection is only defined in the compatibility profile.

10.5

Vertex Array and Vertex Array Object Queries

To query parameters of a vertex array object, use the command void GetVertexArrayiv( uint vaobj, enum pname, int *param ); vaobj is the name of the vertex array object. The value of parameter pname of vaobj is returned in param. pname must be ELEMENT_ARRAY_BUFFER_BINDING. Errors An INVALID_OPERATION error is generated if vaobj is not the name of an existing vertex array object. An INVALID_ENUM error is generated if pname is not ELEMENT_ARRAY_BUFFER_BINDING. To query parameters of an attribute of a vertex array object, use the commands void GetVertexArrayIndexediv( uint vaobj, uint index, enum pname, int *param ); void GetVertexArrayIndexed64iv( uint vaobj, uint index, enum pname, int64 *param ); vaobj is the name of the vertex array object. The value of parameter pname for attribute index of vaobj is returned in param. For GetVertexArrayIndexediv, pname must be one of 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_LONG, VERTEX_ATTRIB_ARRAY_DIVISOR, VERTEX_ATTRIB_RELATIVE_OpenGL 4.6 (Core Profile) - May 5, 2022

10.5. VERTEX ARRAY AND VERTEX ARRAY OBJECT QUERIES

378

OFFSET, VERTEX_ATTRIB_ARRAY_BUFFER_BINDING, VERTEX_BINDING_STRIDE, VERTEX_BINDING_DIVISOR, or VERTEX_BINDING_BUFFER. For GetVertexArrayIndexed64iv, pname must be VERTEX_BINDING_OFFSET.

Errors An INVALID_OPERATION error is generated if vaobj is not the name of an existing vertex array object. An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS, and pname is one of 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_LONG, VERTEX_ATTRIB_ARRAY_DIVISOR, VERTEX_ATTRIB_RELATIVE_OFFSET, or VERTEX_ATTRIB_ARRAY_BUFFER_BINDING An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIB_BINDINGS, and pname is one of VERTEX_BINDING_OFFSET, VERTEX_BINDING_STRIDE, VERTEX_BINDING_DIVISOR, or VERTEX_BINDING_BUFFER. An INVALID_ENUM error is generated if pname is not one of the valid values listed above for the corresponding command. Queries of vertex array state variables are qualified by the value of VERTEX_ARRAY_BINDING to determine which vertex array object is queried. Tables 23.3 and 23.4 define the set of state stored in a vertex array object. To query parameters of an attribute of the currently bound vertex array object, or current attribute values, use the commands void GetVertexAttribdv( uint index, enum pname, double *params ); void GetVertexAttribfv( uint index, enum pname, float *params ); void GetVertexAttribiv( uint index, enum pname, int *params ); void GetVertexAttribIiv( uint index, enum pname, int *params ); void GetVertexAttribIuiv( uint index, enum pname, uint *params );

OpenGL 4.6 (Core Profile) - May 5, 2022

10.5. VERTEX ARRAY AND VERTEX ARRAY OBJECT QUERIES

379

void GetVertexAttribLdv( uint index, enum pname, double *params ); The value of parameter pname for the attribute numbered index of the currently bound vertex array object is returned in 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_LONG, 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). 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, normalized flag, and unconverted integer flag are set by the commands VertexAttribPointer and VertexAttribIPointer. The normalized flag is always set to FALSE by VertexAttribIPointer. The unconverted integer flag is always set to FALSE by VertexAttribPointer and TRUE by VertexAttribIPointer. The query CURRENT_VERTEX_ATTRIB returns the current value for the generic attribute index. GetVertexAttribdv and GetVertexAttribfv read and return the current attribute values as four floating-point values; GetVertexAttribiv reads them as floating-point values and converts them to four integer values; GetVertexAttribIiv reads and returns them as four signed integers; GetVertexAttribIuiv reads and returns them as four unsigned integers; and GetVertexAttribLdv reads and returns them as four double-precision floating-point values. 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. Errors An INVALID_VALUE error is generated if index is greater than or equal to the value of MAX_VERTEX_ATTRIBS.

OpenGL 4.6 (Core Profile) - May 5, 2022

10.6. REQUIRED STATE

380

An INVALID_OPERATION error is generated if no vertex array object is bound (see section 10.3.1). 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. 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 no vertex array object is bound (see section 10.3.1). Finally, the buffer bound to ELEMENT_ARRAY_BUFFER may be queried by calling GetIntegerv with pname ELEMENT_ARRAY_BUFFER_BINDING.

10.6

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, n boolean values indicating whether the attribute values are double precision, three integers for the current array buffer, current element array buffer, and current vertex array bindings, n unsigned integer vertex attribute binding indices, n unsigned integer relative offsets, k integers representing vertex attribute divisors, k unsigned integer vertex buffer bindings, k 64-bit integer vertex binding offsets, k integer vertex binding strides, an unsigned integer representing the

OpenGL 4.6 (Core Profile) - May 5, 2022

10.7.

381

primitive restart index, and two booleans representing the enable state of primitive restart and primitive restart with a fixed index. 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 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, the vertex binding strides are each 16, the restart index is zero, and the restart enables are both FALSE.

10.7 This section is only defined in the compatibility profile.

10.8 This section is only defined in the compatibility profile.

10.9

Conditional Rendering

Conditional rendering can be used to discard rendering commands based on the result of an occlusion query. Conditional rendering is started and stopped using the commands void BeginConditionalRender( uint id, enum mode ); void EndConditionalRender( void ); id specifies the name of an occlusion query object whose results are used to determine if the rendering commands are discarded. If the result (SAMPLES_PASSED) of the query is zero, or if the result (ANY_SAMPLES_PASSED, ANY_SAMPLES_PASSED_CONSERVATIVE, TRANSFORM_FEEDBACK_OVERFLOW, or TRANSFORM_FEEDBACK_STREAM_OVERFLOW) is FALSE, all rendering commands described in section 2.4 are discarded and have no effect when issued between BeginConditionalRender and the corresponding EndConditionalRender. The effect of commands setting current vertex state, such as VertexAttrib, are undefined. If the result (SAMPLES_PASSED) of the query is non-zero, or if the result (ANY_SAMPLES_PASSED, ANY_SAMPLES_PASSED_CONSERVATIVE, TRANSFORM_FEEDBACK_OVERFLOW, or

OpenGL 4.6 (Core Profile) - May 5, 2022

10.9. CONDITIONAL RENDERING

382

TRANSFORM_FEEDBACK_STREAM_OVERFLOW) is TRUE, such commands are not

discarded. mode specifies how BeginConditionalRender interprets the results of the occlusion query given by id. If mode is QUERY_WAIT, the GL waits for the results of the query to be available and then uses the results to determine if subsquent rendering commands are discarded. If mode is QUERY_NO_WAIT, the GL may choose to unconditionally execute the subsequent rendering commands without waiting for the query to complete. If mode is QUERY_BY_REGION_WAIT, the GL will also wait for occlusion query results and discard rendering commands if the result of the occlusion query is zero. If the query result is non-zero, subsequent rendering commands are executed, but the GL may discard the results of the commands for any region of the framebuffer that did not contribute to the sample count in the specified occlusion query. Any such discarding is done in an implementation-dependent manner, but the rendering command results may not be discarded for any samples that contributed to the occlusion query sample count. If mode is QUERY_BY_REGION_NO_WAIT, the GL operates as in QUERY_BY_REGION_WAIT, but may choose to unconditionally execute the subsequent rendering commands without waiting for the query to complete. If mode is QUERY_WAIT_INVERTED, QUERY_NO_WAIT_INVERTED, QUERY_BY_REGION_WAIT_INVERTED, or QUERY_BY_REGION_NO_WAIT_INVERTED then the condition used to determine whether or not to render subsequent drawing commands is negated with respect to QUERY_WAIT, QUERY_NO_WAIT, QUERY_BY_REGION_WAIT, or QUERY_BY_REGION_NO_WAIT, respectively. If mode is QUERY_NO_WAIT_INVERTED or QUERY_BY_REGION_NO_WAIT_INVERTED, the GL may choose to unconditionally execute subsequent rendering commands without waiting for the query to complete. Errors An INVALID_OPERATION error is generated by BeginConditionalRender if called while conditional rendering is in progress. An INVALID_VALUE error is generated by BeginConditionalRender if id is not the name of an existing query object. An INVALID_OPERATION error is generated by BeginConditionalRender if id is the name of a query object with a target other than SAMPLES_PASSED, ANY_SAMPLES_PASSED, ANY_SAMPLES_PASSED_CONSERVATIVE, TRANSFORM_FEEDBACK_OVERFLOW, or TRANSFORM_-

OpenGL 4.6 (Core Profile) - May 5, 2022

10.10. SUBMISSION QUERIES

383

FEEDBACK_STREAM_OVERFLOW; or if id is the name of a query currently in

progress. An INVALID_ENUM error is generated by BeginConditionalRender if mode is not QUERY_WAIT, QUERY_NO_WAIT, QUERY_BY_REGION_WAIT, QUERY_BY_REGION_NO_WAIT, QUERY_WAIT_INVERTED, QUERY_NO_WAIT_INVERTED, QUERY_BY_REGION_WAIT_INVERTED, or QUERY_BY_REGION_NO_WAIT_INVERTED. An INVALID_OPERATION error is generated by EndConditionalRender if called while conditional rendering is not in progress.

10.10

Submission Queries

Submission queries use query objects to track the number of vertices and primitives that are issued to the GL using draw commands. When BeginQuery is called with a target of VERTICES_SUBMITTED, the submitted vertices count maintained by the GL is set to zero. When a vertices submitted query is active, the submitted vertices count is incremented every time a vertex is transferred to the GL (see sections 10.3.5 and 10.4). In the case of primitive types with adjacency information (see sections 10.1.11, 10.1.12, 10.1.13, and 10.1.14), implementations may or may not count vertices not belonging to the main primitive. In the case of line loop primitives (see section 10.1.3), implementations are allowed to count the first vertex twice for the purposes of VERTICES_SUBMITTED queries. Additionally, vertices corresponding to incomplete primitives may or may not be counted. When BeginQuery is called with a target of PRIMITIVES_SUBMITTED, the submitted primitives count maintained by the GL is set to zero. When a primitives submitted query is active, the submitted primitives count is incremented every time a point, line, triangle, or patch primitive is transferred to the GL (see sections 10.1, 10.3.5, and 10.3.6). Restarting a primitive topology using the primitive restart index has no effect on the issued primitives count. Incomplete primitives may or may not be counted.

OpenGL 4.6 (Core Profile) - 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). Vertices may be further processed by tessellation and geometry shaders (see sections 11.2 and 11.3). The resulting transformed vertices are then processed as described in chapter 13. 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 in the shader text using the location qualifier, in a SPIR-V shader using the Location decoration, 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

384

11.1. VERTEX SHADERS

385

Data type

component layout qualifier

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

0 or unspecified 1 2 3 0 or unspecified 1 2 0 or unspecified 1 0 or unspecified

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.

taken from the components of generic attribute i. The generic attribute components used depend on the type of the variable and value of the component layout qualifier (if any) specified in the variable declaration, as identified in table 11.1. An attribute variable declared using a combination of data type and component layout qualifier not listed in this table is not supported and will result in shader compilation errors. 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. When an attribute variable declared using an array type is bound to generic attribute index i, the active array elements are assigned to consecutive generic attributes beginning with generic attribute i. The number of attributes and components assigned to each element are determined according to the data type of array elements and component layout qualifier (if any) specified in the declaration of the array, as described above. For the 64-bit double precision types listed in table 11.3, no default attribute values are provided if the values of the vertex attribute variable are specified with fewer components than required for the attribute variable. For example, the fourth component of a variable of type dvec4 will be undefined if specified using VertexAttribL3dv, or using a vertex array specified with VertexAttribLPointer and

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

Data type

386

Column vector type layout qualifier

mat2, dmat2 mat2x3, dmat2x3 mat2x4, dmat2x4 mat3x2, dmat3x2 mat3, dmat3 mat3x4, dmat3x4 mat4x2, dmat4x2 mat4x3, dmat4x3 mat4, dmat4

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 int ivec2 ivec3 ivec4 uint uvec2 uvec3 uvec4 float vec2 vec3 vec4 double dvec2 dvec3 dvec4

Command VertexAttribI1i VertexAttribI2i VertexAttribI3i VertexAttribI4i VertexAttribI1ui VertexAttribI2ui VertexAttribI3ui VertexAttribI4ui VertexAttrib1* VertexAttrib2* VertexAttrib3* VertexAttrib4* VertexAttribL1d VertexAttribL2d VertexAttribL3d VertexAttribL4d

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

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

387

a size of three. 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 or SPIR-V binary will automatically be bound to vertex attributes by the GL. Such bindings may be queried using the command GetAttribLocation. LinkProgram will fail if the 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 or SPIR-V binary do not leave enough space to assign a location for an active matrix attribute or an active attribute array, both of which require multiple contiguous generic attributes. If an active attribute has a binding explicitly set within the shader text or SPIR-V binary and a different binding assigned by BindAttribLocation, the assignment in the shader text or SPIR-V binary 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 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. BindAttribLocation has no effect on SPIR-V shaders, since the locations must always be fully specified in the SPIR-V shader as described in section 11.1.1.1. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

388

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 (assuming no errors are generated) to: const enum props[] = { ARRAY_SIZE, TYPE }; GetProgramResourceName(program, PROGRAM_INPUT, index, bufSize, length, name); GetProgramResourceiv(program, PROGRAM_INPUT, 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_BaseInstance, gl_BaseVertex, gl_DrawID, gl_InstanceID, and gl_VertexID. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

389

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); 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. For the purposes of this test, attribute variables of the type dvec3, dvec4, dmat2x3, dmat2x4, dmat3, dmat3x4, dmat4x3, and dmat4 may count as consuming twice as many attributes as equivalent single-precision types. While these types use the same number of generic attributes as their single-precision equivalents, implementations are permitted to consume two single-precision vectors of internal storage for each three- or four-component double-precision vector. 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. It is possible for an application to bind more than one attribute name to the same location. This is referred to as aliasing. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

390

no aliasing is done, and may employ optimizations that work only in the absence of aliasing. 11.1.1.1

SPIR-V Vertex Input Interface

When a SPIR-V vertex stage is present, the vertex shader variables listed by OpEntryPoint with the Input storage class form the vertex input attribute interface. These inputs must be decorated with a Location and can also be decorated with a Component decoration. These correspond to the location and component layout discussed in section 11.1.1. The vertex shader input variables are matched only by the Location and Component decorations, and must have a corresponding attribute and binding in the pipeline.

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 7.11. 11.1.2.1

Output Variables

A vertex shader may define one or more output variables or outputs (see the OpenGL Shading Language Specification). The OpenGL 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 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

391

of the implementation-dependent constant MAX_VERTEX_OUTPUT_COMPONENTS. For the purposes of counting input and output components consumed by a shader, variables declared as vectors, matrices, and arrays will all consume multiple components. Each component of variables declared as double-precision floating-point scalars, vectors, or matrices may be counted as consuming two 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 counts against this limit, except for the components of gl_Position. A program containing only a vertex and fragment shader 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 13.3). 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 • tessellation control shader • vertex shader The set of variables to record can be specified in shader text using the xfb_buffer, xfb_offset, or xfb_stride layout qualifiers. For SPIR-V shaders, these are specified by the XfbBuffer, Offset, and XfbStride decorations, respectively. When a SPIR-V entry point is using any of these for transform feedback, it must declare the Xfb Execution Mode. When recording output variables of each vertex in transform feedback mode, a fixed amount of memory is reserved in the buffer bound to each transform feedback OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

392

buffer binding point. Each output variable recorded is associated with a binding point, specified by the xfb_buffer layout qualifier. Each output variable is written to its associated transform feedback binding point at an offset specified by the xfb_offset layout qualifier, in basic machine units, relative to the base of the memory reserved for its vertex. The amount of memory reserved in each transform feedback binding point for a single vertex can be specified using the xfb_stride layout qualifier. If no xfb_stride qualifier is specified for a binding point, the stride is derived by identifying the variable associated with the binding point having the largest offset, and then adding the offset and the size of the variable, in basic machine units. If any variable associated with the binding point contains double-precision floating-point components, the derived stride is aligned to the next multiple of eight basic machine units. If a binding point has no xfb_stride qualifier and no associated output variables, its stride is zero. When no xfb_buffer, xfb_offset, or xfb_stride layout qualifiers are specified, the set of variables to record is specified with the command 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. The variables in varyings are assigned binding points and offsets sequentially, as though each were specified using the xfb_buffer and xfb_offset layout OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

393

qualifiers. The strides associated with each binding point are derived by adding the offset and size of the last variable associated with that binding point. The first variable in varyings is assigned a binding point and offset of zero. When bufferMode is INTERLEAVED_ATTRIBS, each subsequent variable is assigned to the same binding point as the previous variable and an offset equal to the sum of the offset and size of the previous variable. When bufferMode is SEPARATE_ATTRIBS, each subsequent variable is assigned to the binding point following the binding point of the previous variable with an offset of zero. Several special identifiers are supported when bufferMode is INTERLEAVED_ATTRIBS. These identifiers do not identify output variables captured in transform feedback mode, but can be used to modify the binding point and offsets assigned to subsequent variables. If a string in varyings is gl_NextBuffer, the next variable in varyings will be assigned to the next binding point, with an offset of zero. If a string in varyings is gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, or gl_SkipComponents4, the variable is treated as specifying a one- to four-component floating-point output variable with undefined values. No data will be recorded for such strings, but the offset assigned to the next variable in varyings and the stride of the assigned binding point will be affected. 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. An INVALID_OPERATION error is generated if any pointer in varyings identifies the special names gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, or gl_SkipComponents4 and bufferMode is not INTERLEAVED_ATTRIBS, or if the number of gl_NextBuffer pointers in varyings is greater than or equal to the value of MAX_TRANSFORM_FEEDBACK_BUFFERS. The state set by TransformFeedbackVaryings or using transform feedback layout qualifiers has no effect on the execution of the program until program is

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

394

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 one or more buffer objects. If the set of output variables to record in transform feedback mode is specified by TransformFeedbackVaryings, a program will fail to link if: • the count specified by TransformFeedbackVaryings is non-zero, but the program object has no vertex, tessellation control, tessellation evaluation, or geometry shader; • any variable name specified in the varyings array is not one of gl_NextBuffer, gl_SkipComponents1, gl_SkipComponents2, gl_SkipComponents3, or gl_SkipComponents4, and is not declared as a built-in 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; • 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; or • the set of outputs to capture to any single binding point includes outputs from more than one vertex stream. If the set of output variables to record in transform feedback mode is specified using layout qualifiers, a program will fail to link if: • any pair of variables associated with the same binding point overlap in memory (where the offset of the first variable is less than or equal to the offset of the second, but the sum of the offset and size of the first variable is greater than the offset of the second); • any binding point has a stride declared using the xfb_stride layout qualifier and the sum of the offset and size of any variable associated with that binding point exceeds the value of this stride; OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

395

• any variable containing double-precision floating-point components – has an xfb_offset layout qualifier that is not a multiple of eight; or – is associated with a binding point with an xfb_stride layout qualifier that is not a multiple of eight; • the sum of the offset and size of any variable exceeds the maximum stride supported by the implementation (four times the value of MAX_TRANSFORM_FEEDBACK_INTERLEAVED_COMPONENTS); or • the xfb_stride layout qualifier for any binding point exceeds the maximum stride supported by the implementation. For transform feedback purposes, each component of outputs declared as double-precision floating-point scalars, vectors, or matrices is considered to consume eight basic machine units, and each component of any other type is considered to consume four basic machine units. To determine the set of output variables in a linked program object that will be captured in transform feedback mode and the binding points to which those variables are written, applications can query the properties and active resources of the TRANSFORM_FEEDBACK_VARYING and TRANSFORM_FEEDBACK_BUFFER interfaces. If the shader used to record output variables for transform feedback varyings uses the xfb_buffer, xfb_offset, or xfb_stride layout qualifiers, or its SPIR-V entry point declares the Xfb Execution Mode, the values specified by TransformFeedbackVaryings are ignored, and the set of variables captured for transform feedback is instead derived from the specified layout qualifiers or SPIR-V decorations. Outputs specifying both an XfbBuffer and an Offset are captured, while outputs not specifying both of these are not captured. Values are captured each time the shader writes to such a decorated object. Additionally, the 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 (assuming no errors are generated) to:

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

396

const enum props[] = { ARRAY_SIZE, TYPE }; GetProgramResourceName(program, TRANSFORM_FEEDBACK_VARYING, index, bufSize, length, name); GetProgramResourceiv(program, TRANSFORM_FEEDBACK_VARYING, index, 1, &props[0], 1, NULL, size); GetProgramResourceiv(program, TRANSFORM_FEEDBACK_VARYING, index, 1, &props[1], 1, NULL, (int *)type); Special output names (e.g., gl_NextBuffer, gl_SkipComponents1) passed to TransformFeedbackVaryings in the varyings array are counted as outputs to be recorded for the purposes of determining the value of TRANSFORM_FEEDBACK_VARYINGS and for determining the variable selected by index in GetTransformFeedbackVarying. If index identifies gl_NextBuffer, the values zero and NONE will be written to size and type, respectively. If index is of the form gl_SkipComponentsn, the value NONE will be written to type and the number of components n will be written to size. GetTransformFeedbackVarying may be used to query any transform feedback output variable, not just those specified with TransformFeedbackVarying.

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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

397

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 primitive. However, each geometry shader invocation may emit new vertices, which are arranged into primitives and passed to subsequent pipeline stages. Implementations are allowed to skip the execution of certain shader invocations, and to execute additional shader invocations for any shader type during programmable vertex processing due to implementation dependent reasons, including the execution of shader invocations that do not have an active program object present for the particular shader stage, as long as the results of rendering otherwise remain unchanged. Following shader execution, the fixed-function operations described in chapter 13 are applied. Special considerations for vertex shader execution are described in the following sections. 11.1.3.1

Shader Only Texturing

This section describes texture functionality that is accessible through shaders (of all types). Also refer to chapter 8 and to section 8.9(“Texture Functions”) of the OpenGL Shading Language Specification. 11.1.3.2

Texel Fetches

The OpenGL Shading Language texelFetch built-ins provide the ability to extract a single texel from a specified texture image. Texel fetches cannot access cube maps. 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; OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

398

• 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 8.24; • 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) • 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 defined extents of the computed level-of-detail, where any of i<0

i ≥ ws

j<0

j ≥ hs

k<0

k ≥ ds

and ws , hs , and ds refer to the width, height, and depth of the image, as defined in section 8.5.3. • the texture being accessed is not complete, as defined in section 8.17 • the texture being accessed is not bound. OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

399

In all the above cases, if the context was created with robust buffer access enabled (see section 10.3.7), the result of the texture fetch is zero, or a texture source color of (0, 0, 0, 1) in the case of a texel fetch from an incomplete texture. If robust buffer access is not enabled, 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). 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 functions 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: OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

400

• 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 count are returned. The textureQueryLevels functions provide the ability to query the number of accessible mipmap levels in a texture object associated with a sampler uniform. If the sampler is associated with an immutable-format texture object (see section 8.19), the value returned will be: min(levelimmut − 1, levelmax ) − levelbase + 1. Otherwise, the value returned will be an implementation-dependent value between zero and q−levelbase +1. The value returned in that case must satisfy the following constraints: • if all levels of the texture have zero size, zero must be returned • if the texture is complete, a non-zero value must be returned • if the texture is complete and is accessed with a minification filter requiring mipmaps, q − levelbase + 1 must be 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 implementationdependent 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), OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

401

• MAX_TEXTURE_IMAGE_UNITS (for fragment shaders), and • MAX_COMPUTE_TEXTURE_IMAGE_UNITS (for compute shaders). 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 15.1, followed by application of the texture swizzle as described in section 15.2.1 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 8.9(“Texture Functions”) of the OpenGL 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 8.7). 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 equation 8.11) 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 8.23.1, 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. OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

402

• 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. • 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 base 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 8.23 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 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 implementationdependent constants • MAX_VERTEX_ATOMIC_COUNTERS (for vertex shaders), • MAX_TESS_CONTROL_ATOMIC_COUNTERS shaders),

(for

tessellation

OpenGL 4.6 (Core Profile) - May 5, 2022

control

11.1. VERTEX SHADERS

403

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

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

404

• MAX_VERTEX_SHADER_STORAGE_BLOCKS (for vertex shaders) • MAX_TESS_CONTROL_SHADER_STORAGE_BLOCKS (for tessellation control shaders) • 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) • 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 counts 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_BaseInstance, gl_BaseVertex, gl_DrawID, gl_InstanceID, and gl_VertexID. gl_BaseInstance holds the integer value passed to the baseInstance parameter to the command that resulted in the current shader invocation. In the case where the command has no baseInstance parameter, the value of gl_BaseInstance is zero. gl_BaseVertex holds the integer value passed to the baseVertex parameter to the command that resulted in the current shader invocation. In the case where the command has no baseVertex parameter, the value of gl_BaseVertex is zero. gl_DrawID holds the integer draw number the current draw being processed by the shader invocation. It is dynamically uniform. In MultiDraw* variants, this is the zero-based index of the draw within the list of draws processed by the command. In non-MultiDraw* commands, the value of gl_DrawID is always zero. gl_InstanceID holds the integer instance number of the current primitive in an instanced draw call (see section 10.4). gl_VertexID holds the integer index i implicitly passed by DrawArrays or one of the other drawing commands defined in section 10.4. Section 7.1(“Built-In Variables”) of the OpenGL Shading Language Specification also describes these variables. OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS 11.1.3.10

405

Shader Outputs

A vertex shader can write to user-defined output variables. Interpolation of these values across the primitive is determined by the fragment shader input variable interpolation qualifier. sections 4.3.6(“Output Variables”), 4.5(“Interpolation Qualifiers”), and 7.1(“Built-In Variables”) of the OpenGL 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 variables gl_ClipDistance and gl_CullDistance respectively hold the vertex coordinate, and the clip distance and cull distance used in the clipping stage, as described in section 13.7. If clipping is enabled, gl_ClipDistance should be written. The built-in output gl_PointSize, if written, holds the size of the point to be rasterized, measured in pixels. 11.1.3.11

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 executed1 . 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: • 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. 1

The OpenGL Specification differs from the OpenGL ES Specification in what validation takes place at draw-time. In particular, the OpenGL Specification does not require the implementation to detect mismatched interfaces between the shader stages as this might add extra draw-time overhead. Depending on the implementation, developers can activate more draw-time validation by enabling DEBUG_OUTPUT or using a debug context.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

406

• 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. • 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 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 atomic counters, atomic counter buffers, image uniforms, shader output resources, shader storage blocks, texture image units, and uniform blocks used by the current program objects exceeds the corresponding combined limit (the value of MAX_COMBINED_ATOMIC_COUNTERS, MAX_COMBINED_ATOMIC_COUNTER_BUFFERS, MAX_COMBINED_IMAGE_UNIFORMS, MAX_COMBINED_SHADER_OUTPUT_RESOURCES, MAX_COMBINED_SHADER_STORAGE_BLOCKS, MAX_COMBINED_TEXTURE_IMAGE_UNITS, and MAX_COMBINED_UNIFORM_BLOCKS, respectively). 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 may be queried with GetProgramiv (see section 7.14). 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 is OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

407

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. 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 may be queried with GetProgramPipelineiv (see section 7.14). If validation succeeded, no INVALID_OPERATION validation error is generated if pipeline is bound and no program is 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.1. VERTEX SHADERS

408

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, vector 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. Robust buffer access can be enabled by creating a context with robust access enabled through the window system binding APIs. When enabled, out-of-bounds accesses will be bounded within the working memory of the active program, cannot access memory owned by other GL contexts, and will not result in abnormal program termination. Out-of-bounds access to local and global variables cannot read values from other program invocations. An out-of-bounds read may return another value from the active program’s working memory or zero. An out-of-bounds write may overwrite a value from the active program’s working memory or be discarded. Out-of-bounds accesses to resources backed by buffer objects cannot read or modify data outside of the buffer object. For resources bound to buffer ranges, access is restricted within the buffer object from which the buffer range was created, and not within the buffer range itself. Out-of-bounds reads may return any of the following values: • Values from anywhere within the buffer object. • Zero values, or (0, 0, 0, x) vectors for vector reads where x is a valid value represented in the type of the vector components and may be any of – Zero, one, or the maximum representable positive integer value, for signed or unsigned integer components. – 0.0 or 1.0, for floating-point components. Out-of-bounds writes may modify values within the buffer object or be discarded. Out-of-bounds accesses to arrays of resources, such as an array of textures, can only access the data of bound resources. Reads from unbound resources return

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

409

zero and writes are discarded. It is not possible to access data owned by other GL contexts. Applications that require defined behavior for out-of-bounds accesses should range check all computed indices before dereferencing the array, vector or matrix.

11.1.4

Vertex Shader Queries

Vertex shader queries use query objects to track the number of vertex shader invocations. When BeginQuery is called with a target of VERTEX_SHADER_INVOCATIONS, the vertex shader invocations count maintained by the GL is set to zero. When a vertex shader invocations query is active, the counter is incremented every time the vertex shader is invoked (see section 11.1). The result of vertex shader queries may be implementation dependent due to reasons described in section 11.1.3.

11.2

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 there is an active tessellation control or tessellation evaluation program object. 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, the patch provided is passed through to the tessellation primitive generator stage unmodified. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

410

according to the tessellation levels of the patch and the set of layout declarations specified in the tessellation evaluation shader text. The tessellation levels used to control subdivision are normally written by the tessellation control shader. If no tessellation control shader is active, default tessellation levels are instead used. 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 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 the primitive mode is PATCHES and there is no active tessellation evaluation program. 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 Shading Language SpecificaOpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

411

tion. A program will fail to link if the output patch vertex count is not specified by any tessellation control shader object attached to the program, if it is specified differently by multiple tessellation control shader objects, 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 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 7.11. 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 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 a tessellation evaluation shader (section 11.2.3).

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION 11.2.1.2

412

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.

11.2.1.2.2 Tessellation Control Shader Inputs Section 7.1(“Built-In Variables”) of the OpenGL Shading Language Specification describes the built-in 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 values for a specific vertex of the input patch. The length of gl_in is equal to the implementationdependent 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 members of each element of the gl_in array are gl_Position, gl_PointSize, gl_CullDistance, and gl_ClipDistance. 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

413

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 an array, or inside an input block declared as an array. 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 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 variable 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 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 members of each element of the gl_out array are gl_Position, gl_PointSize, gl_ClipDistance, and gl_CullDistance, and behave identically to equivalently named vertex shader outputs (section 11.1.3). Tessellation control shaders additionally have two built-in per-patch output arrays, gl_TessLevelOuter and gl_TessLevelInner. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

414

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. 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 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 input 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 total number of components of active pervertex 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

415

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 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 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

If a tessellation evaluation shader is present, 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 per-patch tessellation levels written by the tessellation control shader, if present, or taken from default patch parameter values. This subdivision is performed in an implementation-dependent manner. If no tessellation evaluation shader is present, the tessellation primitive generator passes incoming primitives through without modification. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

416

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. When no tessellation control shader is present, the tessellation levels are taken from default patch tessellation levels. These default levels are set by calling void PatchParameterfv( enum pname, const float *values ); If pname is PATCH_DEFAULT_OUTER_LEVEL, values specifies an array of four floating-point values corresponding to the four outer tessellation levels for each subsequent patch. If pname is PATCH_DEFAULT_INNER_LEVEL, values specifies an array of two floating-point values corresponding to the two inner tessellation levels. 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 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

417

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 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

418

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 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. Errors An INVALID_ENUM error is generated if pname is not PATCH_DEFAULT_OUTER_LEVEL or PATCH_DEFAULT_INNER_LEVEL.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION 11.2.2.1

419

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 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 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

420

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.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

421

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. After the area corresponding to any inner triangles is filled, the primitive generator generates triangles to cover area between the outermost triangles and the outermost inner triangles. To do this, the temporary subdivision of the outer triangle edges 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 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 11.2, 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

422

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 edges 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 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 edges 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 edge 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

423

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 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

424

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. 11.2.2.3

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 (if present) or provided by the application and transformed by the vertex shader (if no control shader is used).

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

425

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 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

426

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 (if present) or specified as patch parameters. 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. If a tessellation control shader is present, the number of the vertices in the input patch is fixed and is equal to the tessellation control shader output patch size parameter in effect when the program was last linked. If no tessellation control shader is present, the input patch is provided by the application and can have a variable number of vertices, as specified by PatchParameteri. 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 7.11. Tessellation evaluation shaders can access the transformed attributes of all vertices for their input primitive using input variables. If active, a tessellation control shader writing to output variables generates the values of these input variables. If no tessellation control shader is active, input variables will be obtained from vertex shader outputs. Values for any input variables that are not written by a vertex or 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

427

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 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 written by a previous shader (section 11.1.3). If a tessellation control shader is active, the values of gl_in will be taken from tessellation control shader outputs. Otherwise, they will be taken from vertex shader outputs. 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 members of each element of the gl_in array are gl_Position, gl_PointSize, gl_ClipDistance, and gl_CullDistance. 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

428

• 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. If a tessellation control shader is active, the tessellation levels will be taken from the corresponding outputs of the tessellation control shader. Otherwise, the default levels provided as patch parameters are used. 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. 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 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. If no tessellation control shader is active, all such variables are undefined. 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 tesOpenGL 4.6 (Core Profile) - May 5, 2022

11.2. TESSELLATION

429

sellation 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 number of built-in output variables used to pass values to equivalent built-in input variables read by subsequent shader stages or to subsequent fixed functionality vertex processing pipeline stages. These variables are gl_Position, gl_PointSize, gl_ClipDistance, and gl_CullDistance, and all behave identically to equivalently named vertex shader outputs (see section 11.1.3). A tessellation evaluation shader may also declare user-defined 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 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).

11.2.4

Tessellation Shader Queries

Tessellation shader queries use query objects to track the number of tessellation control shader and tessellation evaluation shader invocations. When BeginQuery is called with a target of TESS_CONTROL_SHADER_PATCHES, the tessellation control shader patches count maintained by the GL is set to zero. When a tessellation control shader patches query is active, the counter is incremented every time a patch is processed by the tessellation control shader stage (see section 11.2.1). When BeginQuery is called with a target of TESS_EVALUATION_SHADER_INVOCATIONS, the tessellation evaluation shader invocations count maintained by the GL is set to zero. When a tessellation evaluation shader invocations query is

OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

430

active, the counter is incremented every time the tessellation evaluation shader is invoked (see section 11.2.3). The result of tessellation shader queries may be implementation dependent due to reasons described in section 11.1.3.

11.3

Geometry Shaders

After vertices are processed, they are arranged into primitives, as described in section 10.1. This section describes optional geometry shaders, an additional pipeline stage defining operations to further process those primitives. Geometry shaders 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 OpenGL 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 (section 13.3). 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.

11.3.1

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. OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

431

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 Shading Language Specification. A program will fail to link if the input primitive type is not specified by any geometry shader object attached to the program, or if it is specified differently by multiple geometry shader objects. The input primitive type may be queried by calling GetProgramiv with pname GEOMETRY_INPUT_TYPE. The supported types and the corresponding OpenGL 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

432

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 10.7. 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 Shading Language Specification. A program will fail to link if either the output primitive type or maximum output vertex count are not specified by any geometry shader object attached to the program, or if they are specified differently by multiple geometry shader objects. The output primitive type and maximum output vertex count of a linked program may be queried by calling GetProgramiv with pnames GEOMETRY_OUTPUT_TYPE and GEOMETRY_VERTICES_OUT, respectively.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

11.3.3

433

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 7.11. 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 shader are undefined. Additionally, a geometry shader has access to a built-in input that holds the ID of the current primitive. This ID is generated by the primitive assembly stage preceding the 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 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. 11.3.4.2

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. OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

434

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 Vertex Streams

Geometry shaders may emit primitives to multiple independent vertex streams. Each vertex emitted by the geometry shader is directed at one of the vertex streams. As vertices are received on each stream, they are arranged into primitives of the type specified by the geometry shader output primitive type. The shading language built-in functions EndPrimitive and EndStreamPrimitive may be used to end the primitive being assembled on a given vertex stream and start a new empty primitive of the same type. If an implementation supports N vertex streams, the individual streams are numbered 0 through N − 1. There is no requirement on the order of the streams to which vertices are emitted, and the number of vertices emitted to each stream may be completely independent, subject only to implementationdependent output limits. The primitives emitted to all vertex streams are passed to the transform feedback stage to be captured and written to buffer objects in the manner specified by the transform feedback state. The primitives emitted to all streams but stream zero are discarded after transform feedback. Primitives emitted to stream zero are passed to subsequent pipeline stages for clipping, rasterization, and subsequent fragment processing. Geometry shaders that emit vertices to multiple vertex streams are currently limited to using only the points output primitive type. A program will fail to link if it includes a geometry shader that calls the EmitStreamVertex built-in function and has any other output primitive type parameter.

OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS 11.3.4.4

435

Geometry Shader Inputs

Section 7.1(“Built-In Variables”) of the OpenGL Shading Language Specification describes the built-in variable array gl_in[] available as input to a geometry shader. gl_in[] receives values from 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 primitive type (see section 11.3.1). The members of each element of the gl_in[] array are: • Structure member gl_ClipDistance[] holds the per-vertex array of clip distances, as written by the upstream shader to the built-in output variable gl_ClipDistance[]. • Structure member gl_CullDistance[] holds the per-vertex array of cull distances, as written by the upstream shader to the built-in output variable gl_CullDistance[]. • Structure member gl_PointSize holds the per-vertex point size written by the upstream shader to the built-in output variable gl_PointSize. If the upstream shader does not write gl_PointSize, the value of gl_PointSize is undefined, regardless of the value of the enable PROGRAM_POINT_SIZE. • Structure member gl_Position holds the per-vertex position written by the upstream shader to the 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 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 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. 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 an array or inside an input block declared as an array. Declaring an array size is optional. If no size is specified, it will be OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

436

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). 11.3.4.5

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 built-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(“Storage Qualifiers”) and 7.1(“Built-In Variables”) of the OpenGL 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 Shading Language Specification. The built-in output gl_Position is intended to hold the homogeneous vertex position. Writing gl_Position is optional. OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

437

The built-in outputs gl_ClipDistance and gl_CullDistance hold the clip distance and cull distance, respectively, used in the clipping stage, as described in section 13.7. The built-in output gl_PointSize, if written, holds the size of the point to be rasterized, measured in pixels. 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 13.6) 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. The built-in output gl_ViewportIndex is used to direct rendering to one of several viewports and is 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). 11.3.4.6

Layer and Viewport Selection

Geometry shaders can be used to render to one of several different layers of cube map textures, three-dimensional textures, or one- or two-dimensional texture arrays. 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). Geometry shaders may also select the destination viewport for each output primitive. The destination viewport for a primitive may be selected in the geometry shader by writing to the built-in output variable gl_ViewportIndex. This functionality allows a geometry shader to direct its output to a different viewport OpenGL 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

438

for each primitive, or to draw multiple versions of a primitive into several different viewports. The specific vertex of a primitive that is used to select the rendering layer or viewport index is implementation-dependent and thus portable applications will assign the same layer and viewport index for all vertices in a primitive. The vertex conventions followed for gl_Layer and gl_ViewportIndex may be determined by calling GetIntegerv with pnames LAYER_PROVOKING_VERTEX and VIEWPORT_INDEX_PROVOKING_VERTEX, respectively. For either query, if the value returned is PROVOKING_VERTEX, then vertex selection follows the convention specified by ProvokingVertex (see section 13.6). 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 13.2. 11.3.4.7

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 4.6 (Core Profile) - May 5, 2022

11.3. GEOMETRY SHADERS

11.3.5

439

Geometry Shader Queries

Geometry shader queries use query objects to track the number of geometry shader invocations and the number of primitives those emitted. When BeginQuery is called with a target of GEOMETRY_SHADER_INVOCATIONS, the geometry shader invocations count maintained by the GL is set to zero. When a geometry shader invocations query is active, the counter is incremented every time the geometry shader is invoked (see section 11.3). In case of instanced geometry shaders (see section 11.3.4.2), the geometry shader invocations count is incremented for each separate instanced invocation. When BeginQuery is called with a target of GEOMETRY_SHADER_PRIMITIVES_EMITTED, the geometry shader output primitives count maintained by the GL is set to zero. When a geometry shader primitives emitted query is active, the counter is incremented every time the geometry shader emits a primitive to a vertex stream. Implementations may or may not count primitives emitted to a vertex stream that is not further processed by the GL (see section 11.3.2). Restarting primitive topology using the shading language built-in functions EndPrimitive or EndStreamPrimitive does not increment the geometry shader output primitives count. The result of geometry shader queries may be implementation dependent due to reasons described in section 11.1.3.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 12

This chapter is only defined in the compatibility profile.

440

Chapter 13

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 13.3). • Primitive queries (see section 13.4). • Flatshading (see section 13.6). • Primitive clipping, including client-defined half-spaces (see section 13.7). • Shader output clipping (see section 13.7.1). • Perspective division on clip coordinates (see section 13.8). • Viewport mapping, including depth range scaling (see section 13.8.1). • Front face determination for polygon primitives (see section 14.6.1). • Generic attribute clipping (see section 13.7.1). Next, rasterization is performed on primitives as described in chapter 14.

13.1 This section is only defined in the compatibility profile.

441

13.2. THE LAST VERTEX PROCESSING STAGE

13.2

442

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: • 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.

13.3

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. 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.

13.3.1

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 ); 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. OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

443

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. 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. 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 23.48. Additionally, the new object is bound to the GL state vector and is used for subsequent transform feedback operations1 . 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 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

444

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 object 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. New transform feedback objects may also be created with the command void CreateTransformFeedbacks( sizei n, uint *ids ); CreateTransformFeedbacks returns n previously unused transform feedback object names in ids, each representing a new state vector, comprising the state and with all the same initial values listed in table 23.48. Errors An INVALID_VALUE error is generated if n is negative.

13.3.2

Transform Feedback Primitive Capture

Transform feedback for the currently bound transform feedback object is started (made active) and finished (made inactive) with the commands OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

445

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, as shown in table 13.1. 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 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK Transform Feedback primitiveMode

Allowed render primitive modes

POINTS LINES TRIANGLES

POINTS LINES, LINE_LOOP, LINE_STRIP TRIANGLES, TRIANGLE_STRIP, TRIANGLE_FAN

446

Table 13.1: Legal combinations of the transform feedback primitive mode, as passed to BeginTransformFeedback, and the current primitive mode.

Errors 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 13.1. 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.

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. Regions of buffer objects are bound as targets of the currently bound transform feedback object by calling one of the BindBuffer* commands (see sections 6.1 and 6.1.1 with target set to TRANSFORM_FEEDBACK_BUFFER. Alternatively, regions of buffer objects may be bound directly to a transform feedback object with the commands void TransformFeedbackBufferRange( uint xfb, uint index, uint buffer, intptr offset, sizeiptr size ); void TransformFeedbackBufferBase( uint xfb, uint index, uint buffer ); OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

447

xfb must be zero, indicating the default transform feedback object, or the name of an existing transform feedback object. buffer must be zero or the name of an existing buffer object. TransformFeedbackBufferRange and TransformFeedbackBufferBase behave similarly to BindBufferRange and BindBufferBase, respectively, except that the target of the operation is xfb, and they do not affect any binding to the generic TRANSFORM_FEEDBACK_BUFFER target. Errors An INVALID_OPERATION error is generated if xfb is not zero or the name of an existing transform feedback object. An INVALID_VALUE error is generated if buffer is not zero or the name of an existing buffer object. An INVALID_VALUE error is generated if index is greater than or equal to the number of binding points for transform feedback, as described in section 6.7.1. An INVALID_VALUE error is generated by TransformFeedbackBufferRange if offset is negative. An INVALID_VALUE error is generated by TransformFeedbackBufferRange if size is less than or equal to zero. An INVALID_VALUE error is generated by TransformFeedbackBufferRange if offset or size do not satisfy the constraints described for those parameters for transform feedback array bindings, as described in section 6.7.1. 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 attributes of the first vertex received after BeginTransformFeedback are written at the starting offsets of the bound buffer objects set by BindBufferRange, and subsequent vertex attributes are appended to the buffer object. When capturing line and triangle primitives, all attributes of the first vertex are written first, followed by attributes of the subsequent vertices. When capturing vertices, the stride associated with each transform feedback binding point indicates the number of basic machine units of storage reserved for each vertex in the bound buffer object. For every vertex captured, each output variable with an assigned transform feedback offset will be written to the storage reserved for the vertex at the associated binding point. When writing output variables that are arrays or structures, individual array elements or structure members are written in order. For vector types, individual components are written in order. OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

448

For matrix types, outputs are written as an array of column vectors. If any component of an output with an assigned transform feedback offset was not written to by its shader, the value recorded for that component is undefined. The results of writing 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 capturing a vertex, any portion of the reserved storage not associated with an output variable with an assigned transform feedback offset will be unmodified. 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. When using a geometry shader that writes vertices to multiple vertex streams, each vertex emitted may trigger a new primitive in the vertex stream to which it was emitted. If transform feedback is active, the outputs of the primitive are written to a transform feedback binding point if and only if the outputs directed at that binding point belong to the vertex stream in question. All outputs assigned to a given binding point are required to come from a single vertex stream. If recording the vertices of a primitive to the buffer objects being used for transform feedback purposes would result in either exceeding the limits of any buffer object’s size, or in exceeding the end position offset + size − 1, as set by BindBufferRange, then no vertices of that primitive are recorded in any buffer object, and the counter corresponding to the asynchronous query target TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN (see section 13.4) is not incremented. For the purposes of this test, gl_SkipComponents variables are counted as recording data to a buffer object. Any transform feedback binding point used for capturing vertices must have buffer objects bound when BeginTransformFeedback is called. A binding point requires a bound buffer object if and only if its associated stride in the program object used for transform feedback primitive capture is non-zero and the transform feedback buffer is associated with output variables, blocks or block members in the program object. Errors An INVALID_OPERATION error is generated by BeginTransformFeedback if any of these binding points does not have a buffer object bound.

OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

449

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 object 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. That program object must be active until the 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; • UseProgramStages if the program pipeline object it refers to is current and the current transform feedback object is active and not paused; • BindProgramPipeline if the current transform feedback object is active and not paused; • 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; • ResumeTransformFeedback if the program object being used by the current transform feedback object is not active; • 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 • 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 OpenGL 4.6 (Core Profile) - May 5, 2022

13.3. TRANSFORM FEEDBACK

450

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, MapBuffer may need to block pending the completion of a previous transform feedback operation.

13.3.3

Transform Feedback Draw Operations

When transform feedback is active, the values of output variables or transformed vertex attributes are captured into the buffer objects attached to the current transform feedback object. After transform feedback is complete, subsequent rendering operations may use the contents of these buffer objects (see section 6). The number of vertices captured from each vertex stream during transform feedback is stored in the corresponding transform feedback object and may be used in conjunction with the commands void DrawTransformFeedback( enum mode, uint id ); void DrawTransformFeedbackInstanced( enum mode, uint id, sizei instancecount ); void DrawTransformFeedbackStream( enum mode, uint id, uint stream ); void DrawTransformFeedbackStreamInstanced( enum mode, uint id, uint stream, sizei instancecount ); to replay the captured vertices. DrawTransformFeedbackStreamInstanced is equivalent to calling DrawArraysInstanced with mode as specified, first set to zero, count set to the number of vertices captured from the vertex stream numbered stream the last time transform feedback was active on the transform feedback object named id, and instancecount as specified. Calling DrawTransformFeedbackInstanced is equivalent to calling DrawTransformFeedbackStreamInstanced with stream set to zero. OpenGL 4.6 (Core Profile) - May 5, 2022

13.4. PRIMITIVE QUERIES

451

Calling DrawTransformFeedbackStream is equivalent to calling DrawTransformFeedbackStreamInstanced with instancecount set to one. Finally, calling DrawTransformFeedback is equivalent to calling DrawTransformFeedbackStreamInstanced with stream set to zero and instancecount set to one. Note that the vertex count is from the number of vertices recorded to the selected vertex stream during the transform feedback operation. If no outputs belonging to the selected vertex stream are recorded, the corresponding vertex count will be zero even if complete primitives were emitted to the selected stream. No error is generated if the transform feedback object named by id is active; the vertex count used for the rendering operation is set by the previous EndTransformFeedback command. Errors An INVALID_VALUE error is generated if stream is greater than or equal to the value of MAX_VERTEX_STREAMS. An INVALID_VALUE error is generated if id is not the name of a transform feedback object. An INVALID_VALUE error is generated if instancecount is negative. An INVALID_OPERATION error is generated if EndTransformFeedback has never been called while the object named by id was bound.

13.4

Primitive Queries

Primitive queries use query objects to track the number of primitives in each vertex stream that are generated by the GL and the number of primitives in each vertex stream that are written to buffer objects in transform feedback mode. When BeginQueryIndexed is called with a target of PRIMITIVES_GENERATED, the primitives generated count maintained by the GL for the vertex stream index is set to zero. There is a separate query and counter for each vertex stream. The number of vertex streams is given by the value of the implementationdependent constant MAX_VERTEX_STREAMS. When a generated primitive query for a vertex stream is active, the primitives-generated count is incremented every time a primitive emitted to that stream reaches the transform feedback stage (see section 13.3), 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 BeginQueryIndexed is called with a target of TRANSFORM_-

OpenGL 4.6 (Core Profile) - May 5, 2022

13.5. TRANSFORM FEEDBACK OVERFLOW QUERIES

452

FEEDBACK_PRIMITIVES_WRITTEN, the transform feedback primitives written count maintained by the GL for vertex stream index is set to zero. There is a separate query and counter for each vertex stream. When a transform feedback primitives written query for a vertex stream is active, the counter for that vertex stream is incremented every time the vertices of a primitive written to that stream are recorded into one or more buffer objects. If transform feedback is not active or if a primitive to be recorded does not fit in a buffer object, the counter is not incremented. 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.

13.5

Transform Feedback Overflow Queries

Transform feedback overflow queries use query objects to track whether or not all stream vertices were written to buffer objects in transform feedback mode. When BeginQueryIndexed is called with a target of TRANSFORM_FEEDBACK_OVERFLOW, the transform feedback overflow state maintained by the GL is set to false. If transform feedback is active and a primitive to be recorded on any stream does not fit in one of the buffer objects used for capture, the overflow state is set to true. When BeginQueryIndexed is called with a target of TRANSFORM_FEEDBACK_STREAM_OVERFLOW, the transform feedback stream overflow state maintained by the GL for vertex stream index is set to false. There is a separate overflow state for each vertex stream. If transform feedback is active and a primitive to be recorded on stream index does not fit in one of the buffer objects used for capture, the overflow state for vertex stream index is set to true.

13.6

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. The provoking vertex is controlled with the command void ProvokingVertex( enum provokeMode );

OpenGL 4.6 (Core Profile) - May 5, 2022

13.7. PRIMITIVE CLIPPING

453

Primitive type of polygon i point independent line line loop

First vertex convention i 2i − 1 i

line strip independent triangle triangle strip triangle fan line adjacency line strip adjacency triangle adjacency triangle strip adjacency

i 3i − 2 i i+1 4i − 2 i+1 6i − 5 2i − 1

Last vertex convention 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 13.2: 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. provokeMode must be either FIRST_VERTEX_CONVENTION or LAST_VERTEX_CONVENTION, and controls selection of the vertex whose values are assigned to flatshaded colors and outputs, as shown in table 13.2. When a vertex processing stage is active, user-defined output variables may be flatshaded by using the flat qualifier when declaring the output, as described in section 4.5(“Interpolation Qualifiers”) of the OpenGL Shading Language Specification. The state required for flatshading is one bit for the provoking vertex mode. The initial value of the provoking vertex mode is LAST_VERTEX_CONVENTION.

13.7

Primitive Clipping

The command void ClipControl( enum origin, enum depth ); controls the clipping volume behavior. origin and depth specify the clip control origin and depth mode respectively. The interpretation of these parameters is discussed later in this section and in section 13.8. OpenGL 4.6 (Core Profile) - May 5, 2022

13.7. PRIMITIVE CLIPPING

454

Errors An INVALID_ENUM error is generated if origin is not LOWER_LEFT or UPPER_LEFT. An INVALID_ENUM error is generated if depth is not NEGATIVE_ONE_TO_ONE or ZERO_TO_ONE. Primitives are culled against the cull volume and then clipped to the clip volume. In clip coordinates, the view volume is defined by −wc ≤ xc ≤ wc −wc ≤ yc ≤ wc zmin ≤ zc ≤ wc . where zmin is −wc when the clip control depth mode is NEGATIVE_ONE_TO_ONE, and 0 when the mode is ZERO_TO_ONE. This view volume may be further restricted by as many as n client-defined halfspaces. n is an implementation-dependent maximum that must be at least 8, and may be determined by calling GetIntegerv with pname MAX_COMBINED_CLIP_AND_CULL_DISTANCES. The cull volume is the intersection of up to the value of MAX_CULL_DISTANCES client-defined half-spaces (if no client-defined cull half-spaces are enabled, culling against the cull volume is skipped). The number of enabled cull half-spaces is determined by the explicit or implicit size of the built-in array gl_CullDistance in the last shader stage before rasterization which has an active program. A shader may write a single cull distance for each enabled cull half-space to elements of the gl_CullDistance[] array. If the cull distance for any enabled cull half-space is negative for all of the vertices of the primitive under consideration, the primitive is discarded. Otherwise the primitive is clipped against the clip volume as defined below. The clip volume is the intersection of up to the value of MAX_CLIP_DISTANCES client-defined half-spaces with the view volume (if no client-defined clip half-spaces are enabled, the clip volume is the view volume). The last vertex processing stage may write a single clip distance for each enabled clip half-space to elements of the gl_ClipDistance[] array. Clip halfspace i is then given by the set of points satisfying the inequality ci (P ) ≥ 0, where ci (P ) is the value of clip distance i at point P . For point primitives, ci (P ) is simply the clip distance for the vertex in question. For line and triangle OpenGL 4.6 (Core Profile) - May 5, 2022

13.7. PRIMITIVE CLIPPING

455

primitives, per-vertex clip distances are interpolated using a weighted mean, with weights derived according to the algorithms described in sections 14.5 and 14.6. Client-defined clip half-spaces are enabled or disabled by calling Enable or Disable with target CLIP_DISTANCEi, where i is an integer between 0 and n − 1; specifying a value of i enables or disables the client-defined clip half-space with index i. The constants obey CLIP_DISTANCEi = CLIP_DISTANCE0 + i. Depth clamping is enabled or disabled by calling Enable or Disable with target DEPTH_CLAMP. If depth clamping is enabled, the zmin ≤ zc ≤ wc plane equation (see the clip volume definition above) is ignored by view volume clipping (effectively, there is no near or far plane clipping). If the primitive under consideration is a point, then clipping passes it unchanged if it lies within the near and far clip planes, and the client-defined halfspaces; otherwise, it is discarded. In addition, implementations may discard points that lie outside the view volume. If the primitive is a line segment, then clipping does nothing to it if it lies entirely within the clip volume, and discards it if it lies entirely outside the volume. If part of the line segment lies in the volume 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 boundary of the clip volume. 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 13.7.1. 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. Primitives rendered with user-defined half-spaces must satisfy a complementarity criterion. Suppose a series of primitives is drawn where each vertex i has a OpenGL 4.6 (Core Profile) - May 5, 2022

13.7. PRIMITIVE CLIPPING

456

single specified clip distance di (or a number of similarly specified clip distances, if multiple half-spaces are enabled). Next, suppose that the same series of primitives are drawn again with each such clip distance replaced by −di (and the GL is otherwise in the same state). In this case, primitives must not be missing any pixels, nor may any pixels be drawn twice in regions where those primitives are cut by the clip planes. The state required for clipping is one bit for clip control origin, one bit for clip control depth mode, at least 8 bits indicating which of the client-defined halfspaces are enabled. In the initial state, the clip control origin is LOWER_LEFT, the depth mode is NEGATIVE_ONE_TO_ONE, and all half-spaces are disabled. Implementations are allowed to pass incoming primitives unchanged and to output multiple primitives for an incoming primitive due to implementation dependent reasons as long as the results of rendering otherwise remain unchanged.

13.7.1

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. 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 13.7) for a clipped point P is used to obtain the output value associated with P as 2 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. For outputs of the last vertex processing stage specified to be interpolated without perspective correction (using the noperspective qualifier), the value of t used to obtain the output value associated with P will be adjusted to produce results that vary linearly in screen space. 2

Since this computation is performed in clip space before division by wc , clipped output values are perspective-correct.

OpenGL 4.6 (Core Profile) - May 5, 2022

13.8. COORDINATE TRANSFORMATIONS

457

Implementations need not support interpolation of output values of integer or unsigned integer type, as all such attributes must be flat shaded.

13.7.2 This subsection is only defined in the compatibility profile.

13.7.3

Primitive Clipping Queries

Primitive clipping queries use query objects to track the number of primitives that are processed by the primitive clipping stage and the number of primitives that are output by the primitive clipping stage and are further processed by the rasterization stage. When BeginQuery is called with a target of CLIPPING_INPUT_PRIMITIVES, the clipping input primitives count maintained by the GL is set to zero. When a clipping input primitives query is active, the counter is incremented every time a primitive reaches the primitive clipping stage (see section 13.7). When BeginQuery is called with a target of CLIPPING_OUTPUT_PRIMITIVES, the clipping output primitives count maintained by the GL is set to zero. When a clipping output primitives query is active, the counter is incremented every time a primitive passes the primitive clipping stage. The actual number of primitives output by the primitive clipping stage for a particular input primitive is implementation dependent (see section 13.7) but must satisfy the following conditions: • If at least one vertex of the input primitive lies inside the clipping volume, the counter is incremented by one or more. • Otherwise, the counter is incremented by zero or more. If RASTERIZER_DISCARD is enabled, implementations are allowed to discard primitives right after the optional transform feedback state (see section 14.1). As a result, if RASTERIZER_DISCARD is enabled, the clipping input and output primitives count may not be incremented.

13.8

Coordinate Transformations

Clip coordinates for a vertex result from shader execution, which yields a vertex coordinate gl_Position.

OpenGL 4.6 (Core Profile) - May 5, 2022

13.8. COORDINATE TRANSFORMATIONS

458

Perspective division on clip coordinates yields normalized device coordinates, followed by a viewport transformation (see section 13.8.1) to convert these coordinates into window coordinates. If a vertex in clip coordinates is given by   xc  yc     zc  wc then the vertex’s normalized device coordinates are    xc  xd wc f ×y   yd  =   wc c  zc zd w c

where f is 1 when the clip control origin is LOWER_LEFT and −1 when the origin is UPPER_LEFT.

13.8.1

Controlling the Viewport

The viewport transformation is determined by the selected viewport’s width and height in pixels, px and py , respectively,   and its center (ox , oy ) (also in pixels). xw  The vertex’s window coordinates, yw  , are given by zw    px  xw 2 xd + ox  yw  =  py yd + oy  . 2 zw s × zd + b n+f where s = f −n 2 and b = 2 when the clip control depth mode is NEGATIVE_ONE_TO_ONE, or s = f − n and b = n when the mode is ZERO_TO_ONE. Multiple viewports are available and are numbered zero through the value of MAX_VIEWPORTS minus one. If a geometry shader is active and writes to gl_ViewportIndex, the viewport transformation uses the viewport corresponding to the value assigned to gl_ViewportIndex taken from an implementationdependent primitive vertex. If the value of the viewport index is outside the range zero to the value of MAX_VIEWPORTS minus one, the results of the viewport transformation are undefined. If no geometry shader is active, or if the active geometry shader does not write to gl_ViewportIndex, the viewport numbered zero is used by the viewport transformation.

OpenGL 4.6 (Core Profile) - May 5, 2022

13.8. COORDINATE TRANSFORMATIONS A single vertex may be used in more than one individual primitive, in primitives such as TRIANGLE_STRIP. In this case, the viewport transformation is applied separately for each primitive. The factor and offset applied to zd for each viewport encoded by n and f are set using void DepthRangeArrayv( uint first, sizei count, const double *v ); void DepthRangeIndexed( uint index, double n, double f ); void DepthRange( double n, double f ); void DepthRangef( float n, float f ); DepthRangeArrayv is used to specify the depth range for multiple viewports simultaneously. first specifies the index of the first viewport to modify and count specifies the number of viewports. Viewports whose indices lie outside the range [f irst, f irst + count) are not modified. The v parameter contains the address of an array of double types specifying near (n) and far (f) for each viewport in that order. Values in v are each clamped to the range [0, 1] when specified. Errors An INVALID_VALUE error is generated if (first + count) is greater than the value of MAX_VIEWPORTS. An INVALID_VALUE error is generated if count is negative. DepthRangeIndexed specifies the depth range for a single viewport and is equivalent (assuming no errors are generated) to: double v[] = { n, f }; DepthRangeArrayv(index, 1, v); DepthRange sets the depth range for all viewports to the same values and is equivalent (assuming no errors are generated) to: for (uint i = 0; i < MAX_VIEWPORTS; i++) DepthRangeIndexed(i, n, 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 OpenGL 4.6 (Core Profile) - May 5, 2022

459

13.8. COORDINATE TRANSFORMATIONS

460

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). Viewport transformation parameters are specified using void ViewportArrayv( uint first, sizei count, const float *v ); void ViewportIndexedf( uint index, float x, float y, float w, float h ); void ViewportIndexedfv( uint index, const float *v ); void Viewport( int x, int y, sizei w, sizei h ); ViewportArrayv specifies parameters for multiple viewports simultaneously. first specifies the index of the first viewport to modify and count specifies the number of viewports. Viewports whose indices lie outside the range [first, first + count) are not modified. v contains the address of an array of floating-point values specifying the left (x), bottom (y), width (w) and height (h) of each viewport, in that order. x and y give the location of the viewport’s lower left corner and w and h give the viewport’s width and height, respectively. Errors An INVALID_VALUE error is generated if first + count is greater than the value of MAX_VIEWPORTS. An INVALID_VALUE error is generated if count is negative. ViewportIndexedf and ViewportIndexedfv specify parameters for a single viewport and are equivalent (assuming no errors are generated) to: float v[4] = { x, y, w, h }; ViewportArrayv(index, 1, v); and ViewportArrayv(index, 1, v); respectively. Viewport sets the parameters for all viewports to the same values and is equivalent (assuming no errors are generated) to: for (uint i = 0; i < MAX_VIEWPORTS; i++) ViewportIndexedf(i, 1, (float)x, (float)y, (float)w, (float)h); OpenGL 4.6 (Core Profile) - May 5, 2022

13.9.

461

The viewport parameters shown in the above equations are found from these values as ox = x + w2 oy = y + h2 px = w py = h. The location of the viewport’s bottom-left corner, given by (x, y), are clamped to be within the implementation-dependent viewport bounds range. The viewport bounds range [min, max] tuple may be determined by calling GetFloatv with pname VIEWPORT_BOUNDS_RANGE (see section 22). Viewport width and height are clamped to implementation-dependent maximums when specified. The maximum width and height may be found by calling GetFloatv with pname 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 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 for each viewport. In the initial state, w and h for each viewport 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. The precision with which the GL interprets the floating-point viewport bounds is implementation-dependent and may be determined by querying the implementation-defined constant VIEWPORT_SUBPIXEL_BITS.

13.9 This section is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 14

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 14.1-14.6 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 14.9, which may modify or discard fragments. Surviving fragments are processed by fragment shaders (see chapter 15). Fragment shaders determine color values for fragments, and may also modify or replace their assigned depth values. Figure 14.1 diagrams the rasterization process. Grid squares need not actually be square in the GL. Rasterization rules are not affected by the actual aspect ratio of the grid squares. Display of non-square grids, however, will cause rasterized points and line segments to appear fatter in one

462

463

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 14.1. Rasterization, early per-fragment tests, and fragment shading. Optional early tests described in section 14.9 are included in the “Other Early Tests” box.

OpenGL 4.6 (Core Profile) - May 5, 2022

14.1. DISCARDING PRIMITIVES BEFORE RASTERIZATION

464

direction than the other. We assume that fragments are square, since it simplifies antialiasing and texturing. Several factors affect rasterization. Primitives may be discarded before rasterization. Points may be given differing diameters and line segments differing widths. A point, line segment, or polygon may be antialiased. Rasterization only produces fragments corresponding to pixels in the framebuffer. 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 14.9.

14.1

Discarding Primitives Before Rasterization

Primitives sent to vertex stream zero (see section 13.3) are processed further; primitives emitted to any other stream are discarded. When geometry shaders are disabled, all vertices are considered to be emitted to stream zero. Primitives can be optionally discarded before rasterization by calling Enable and Disable with target RASTERIZER_DISCARD. When enabled, primitives are discarded immediately before the rasterization stage, but after the optional transform feedback stage (see section 13.3). 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.

14.2

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 .

14.3

Antialiasing

The R, G, and B values of the rasterized fragment are left unaffected, but the A value is multiplied by a floating-point value in the range [0, 1] that describes a

OpenGL 4.6 (Core Profile) - May 5, 2022

14.3. ANTIALIASING

465

fragment’s screen pixel coverage. The per-fragment stage of the GL can be set up to use the A value to blend the incoming fragment with the corresponding pixel already present in the framebuffer. The details of how antialiased fragment coverage values are computed are difficult to specify in general. The reason is that high-quality antialiasing may take into account perceptual issues as well as characteristics of the monitor on which the contents of the framebuffer are displayed. Such details cannot be addressed within the scope of this document. Further, the coverage value computed for a fragment of some primitive may depend on the primitive’s relationship to a number of grid squares neighboring the one corresponding to the fragment, and not just on the fragment’s grid square. Another consideration is that accurate calculation of coverage values may be computationally expensive; consequently we allow a given GL implementation to approximate true coverage values by using a fast but not entirely accurate coverage computation. In light of these considerations, we chose to specify the behavior of exact antialiasing in the prototypical case that each displayed pixel is a perfect square of uniform intensity. The square is called a fragment square and has lower left corner (x, y) and upper right corner (x+1, y +1). We recognize that this simple box filter may not produce the most favorable antialiasing results, but it provides a simple, well-defined model. A GL implementation may use other methods to perform antialiasing, subject to the following conditions: 1. If f1 and f2 are two fragments, and the portion of f1 covered by some primitive is a subset of the corresponding portion of f2 covered by the primitive, then the coverage computed for f1 must be less than or equal to that computed for f2 . 2. The coverage computation for a fragment f must be local: it may depend only on f ’s relationship to the boundary of the primitive being rasterized. It may not depend on f ’s x and y coordinates. Another property that is desirable, but not required, is: 3. The sum of the coverage values for all fragments produced by rasterizing a particular primitive must be constant, independent of any rigid motions in window coordinates, as long as none of those fragments lies along window edges. In some implementations, varying degrees of antialiasing quality may be obtained by providing GL hints (section 21.5), allowing a user to make an image quality versus speed tradeoff. OpenGL 4.6 (Core Profile) - May 5, 2022

14.3. ANTIALIASING

14.3.1

466

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 each time a pixel is updated, so the antialiasing appears to be automatic at the application level. 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 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 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 only lines are being rendered, the “smooth” antialiasing mechanism provided by the base GL may result in a higher quality image. This mechanism is designed to allow multisample and smooth antialiasing techniques to be alternated during the rendering of a single scene. 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 the value of 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 sample location is returned as two floating-point values in val[0] and val[1], each between 0 and 1, corresponding to the x and y locations respectively in GL pixel space of that sample. (0.5, 0.5) thus corresponds to the pixel center. If the multisample mode does not have fixed sample locations, the returned values may only reflect the locations of samples within some pixels. OpenGL 4.6 (Core Profile) - May 5, 2022

14.3. ANTIALIASING

467

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 enabled or disabled by calling Enable or Disable with target MULTISAMPLE. If MULTISAMPLE is disabled, multisample rasterization of all primitives is equivalent to single-sample (fragment-center) rasterization, except that the fragment coverage value is set to full coverage. The color and depth values and the sets of texture coordinates may all be set to the values that would have been assigned by single-sample rasterization, or they may be assigned as described below for multisample rasterization. If MULTISAMPLE is enabled, multisample rasterization of all primitives differs substantially from single-sample 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 MULTISAMPLE is enabled 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. OpenGL 4.6 (Core Profile) - May 5, 2022

14.4. POINTS

468

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 14.2 is relaxed for all multisample rasterization, because the sample locations may be a function of pixel location. 14.3.1.1

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 MULTISAMPLE 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 value 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 colors and other associated data are evaluated for each sample, and each set of values is evaluated at the sample location.

14.4

Points

A point is drawn by generating a set of fragments in the shape of a square or circle centered around the vertex of the point. Each vertex has an associated point size that controls the size of that square or circle. If program point size mode is enabled, the derived point size is taken from the (potentially clipped) shader built-in gl_PointSize written by the last vertex processing stage and 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 was written to gl_PointSize, results are undefined. If program point size mode is disabled, the derived point size is specified with the command OpenGL 4.6 (Core Profile) - May 5, 2022

14.4. POINTS

469

void PointSize( float size ); size specifies the requested size of a point. The default value is 1.0. Errors An INVALID_VALUE error is generated if size is less than or equal to zero. Program point size mode is enabled and disabled by calling Enable or Disable with target PROGRAM_POINT_SIZE. If multisampling is enabled, an implementation may optionally fade the point alpha (see section 17.2) instead of allowing the point width to go below a given threshold. In this case, the width of the rasterized point is derived size derived size ≥ threshold width = (14.1) threshold otherwise and the fade factor is computed as follows: ( 1 derived size ≥ threshold f ade = derived size 2 otherwise threshold

(14.2)

The point fade threshold, is specified with void PointParameter{if}( enum pname, T param ); void PointParameter{if}v( enum pname, const T *params ); If pname is POINT_FADE_THRESHOLD_SIZE, then param specifies, or params points to the point fade threshold. Data conversions are performed as specified in section 2.2.1. The point sprite texture coordinate origin is set with the PointParameter* commands where pname is POINT_SPRITE_COORD_ORIGIN and param is LOWER_LEFT or UPPER_LEFT. The default value is UPPER_LEFT. Errors An INVALID_ENUM error is generated if pname is not POINT_FADE_THRESHOLD_SIZE or POINT_SPRITE_COORD_ORIGIN. An INVALID_VALUE error is generated if negative values are specified for POINT_FADE_THRESHOLD_SIZE.

OpenGL 4.6 (Core Profile) - May 5, 2022

14.4. POINTS

14.4.1

470

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 contains point sprite texture coordinates. The s point sprite texture coordinate varies from zero to one across the point horizontally left-to-right. If POINT_SPRITE_COORD_ORIGIN is LOWER_LEFT, the t coordinate varies from zero to one vertically bottom-to-top. Otherwise if the point sprite texture coordinate origin is UPPER_LEFT, the t coordinate varies from zero to one vertically top-to-bottom. The following formula is used to evaluate the s and t point sprite texture coordinates: xf + 21 − xw 1 s= + (14.3) 2 size

  1 

(yf + 12 −yw ) , POINT_SPRITE_COORD_ORIGIN = LOWER_LEFT 2 + size 1 −y y + ) ( w f 1 2 , POINT_SPRITE_COORD_ORIGIN = UPPER_LEFT 2 − size

(14.4) 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. Not all point widths need be supported, but the width 1.0 must be provided. The range of supported widths and the width of evenly-spaced gradations within that range are implementation-dependent. The range and gradations may be obtained using the query mechanism described in chapter 22. 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 point widths may also be supported. There is no requirement that these widths must be equally spaced. If an unsupported width is requested, the nearest supported width is used instead.

14.4.2

Point Rasterization State

The state required to control point rasterization consists of the floating-point point width, a bit indicating whether or not vertex program point size mode is enabled, a bit for the point sprite texture coordinate origin, and a floating-point value specifying the point fade threshold size. OpenGL 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

14.4.3

471

Point Multisample Rasterization

If MULTISAMPLE is enabled, and 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 side equal to the current point width. 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. The set of point sizes supported is equivalent to those for point sprites without multisample.

14.5

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. Antialiasing may be enabled or disabled by calling Enable or Disable with target LINE_SMOOTH. Errors An INVALID_VALUE error is generated if width is less than or equal to zero.

14.5.1

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

OpenGL 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

472

of four half planes: Rf = { (x, y) | |x − xf | + |y − yf | <

1 }. 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 14.2. 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). 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 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

473

Figure 14.2. 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.

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

(14.5)

(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

(14.6)

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 (14.7) where za and zb are the depth values of the starting and ending endpoints of the segment, respectively.

OpenGL 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

474

The noperspective and flat keywords used to declare shader outputs affect how they are interpolated. When neither keyword is specified, interpolation is performed as described in equation 14.6. When the noperspective keyword is specified, interpolation is performed in the same fashion as for depth values, as described in equation 14.7. 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 13.6).

14.5.2

Other Line Segment Features

We have just described the rasterization of non-antialiased line segments of width one. We now describe the rasterization of line segments for general values of the line segment rasterization parameters. 14.5.2.1 This subsection is only defined in the compatibility profile. 14.5.2.2

Wide Lines

The actual width of non-antialiased lines is determined by rounding the supplied width to the nearest integer, then clamping it to the implementation-dependent maximum non-antialiased line width. This implementation-dependent value must be no less than the implementation-dependent maximum antialiased line width, rounded to the nearest integer value, and in any event no less than 1. If rounding the specified width results in the value 0, then it is as if the value was 1. Non-antialiased 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 14.3). 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

OpenGL 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

width = 2

475

width = 3

Figure 14.3. Rasterization of non-antialiased 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.

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. 14.5.2.3

Antialiasing

Rasterized antialiased line segments produce fragments which intersect a rectangle centered on the line segment. Two of the edges are parallel to the specified line segment; each is at a distance of one-half the current 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 values are computed for each fragment by computing the area of the intersection of the rectangle with the fragment square (see figure 14.4; see also section 14.3). Equation 14.6 is used to compute associated data values just as with non-antialiased lines; equation 14.5 is used to find the value of t for each fragment whose square is intersected by the line segment’s rectangle. Not all widths need be supported for line segment antialiasing, but width 1.0 antialiased segments must be provided. As with the point width, a GL implementation may be queried for the OpenGL 4.6 (Core Profile) - May 5, 2022

14.5. LINE SEGMENTS

476

Figure 14.4. The region used in rasterizing and finding corresponding coverage values for an antialiased line segment (an x-major line segment is shown).

range and number of gradations of available antialiased line widths.

14.5.3

Line Rasterization State

The state required for line rasterization consists of the floating-point line width and a bit indicating whether line antialiasing is on or off. The initial value of the line width is 1.0. The initial state of line segment antialiasing is disabled.

14.5.4

Line Multisample Rasterization

If MULTISAMPLE is enabled, and the value of SAMPLE_BUFFERS is one, then lines are rasterized using the following algorithm, regardless of whether line antialiasing (LINE_SMOOTH) is enabled or disabled. Line rasterization produces a fragment for each framebuffer pixel with one or more sample points that intersect the rectangular region that is described in the Antialiasing portion of section 14.5.2 (Other Line Segment Features). Coverage bits that correspond to sample points that intersect a retained 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 14.5, then using the result to evaluate equation 14.7. An implementation may choose to assign the associated data to more than one sample by evaluating equation 14.5 at any location within the pixel including the fragment center or any one of the samOpenGL 4.6 (Core Profile) - May 5, 2022

14.6. POLYGONS

477

ple locations, then substituting into equation 14.6. The different associated data values need not be evaluated at the same location. Line width range and number of gradations are equivalent to those supported for antialiased lines.

14.6

Polygons

A polygon results from a triangle arising from a triangle strip, triangle fan, or series of separate triangles. Like points and line segments, polygon rasterization is controlled by several variables. Polygon antialiasing is enabled or disabled by calling Enable or Disable with target POLYGON_SMOOTH.

14.6.1

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

1 X i i⊕1 i a= f xw yw − xi⊕1 w yw 2

(14.8)

i=0

where f = 1 when the clip control origin is LOWER_LEFT and f = −1 when the i are the x and y window coordinates of the ith origin is UPPER_LEFT, xiw and yw vertex of the 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. Errors An INVALID_ENUM error is generated if dir is not CW or CCW.

OpenGL 4.6 (Core Profile) - May 5, 2022

14.6. POLYGONS

478

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 must be FRONT, BACK or FRONT_AND_BACK. Culling is enabled or disabled by calling Enable or Disable with target CULL_FACE. Front-facing polygons are rasterized if either culling is disabled or the CullFace mode is BACK while backfacing 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 )

A(ppa pc ) , A(pa pb pc )

A(ppa pb ) , A(pa pb pc )

where A(lmn) denotes the area in window coordinates of the triangle with vertices l, m, and n. OpenGL 4.6 (Core Profile) - May 5, 2022

14.6. POLYGONS

479

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

(14.9)

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

(14.10)

where za , zb , and zc are the depth values of pa , pb , and pc , respectively. The noperspective and flat keywords used to declare shader outputs affect how they are interpolated. When neither keyword is specified, interpolation is performed as described in equation 14.9. When the noperspective keyword is specified, interpolation is performed in the same fashion as for depth values, as described in equation 14.10. 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 13.6). 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 14.9 should be iterated independently and a division performed for each fragment).

OpenGL 4.6 (Core Profile) - May 5, 2022

14.6. POLYGONS

480

14.6.2 This subsection is only defined in the compatibility profile.

14.6.3

Antialiasing

Polygon antialiasing rasterizes a polygon by producing a fragment wherever the interior of the polygon intersects that fragment’s square. A coverage value is computed at each such fragment, and this value is saved to be applied as described in section 17.1. An associated datum is assigned to a fragment by integrating the datum’s value over the region of the intersection of the fragment square with the polygon’s interior and dividing this integrated value by the area of the intersection. For a fragment square lying entirely within the polygon, the value of a datum at the fragment’s center may be used instead of integrating the value across the fragment.

14.6.4

Options Controlling Polygon Rasterization

The interpretation of polygons for rasterization is controlled using void PolygonMode( enum face, enum mode ); face must be FRONT_AND_BACK, indicating that the rasterizing method described by mode replaces the rasterizing method for both front- and back-facing polygons. mode must be one of POINT, LINE, or FILL. Calling PolygonMode with POINT causes the vertices of a polygon to be treated, for rasterization purposes, as if they had been drawn with mode POINTS. LINE causes edges to be rasterized as line segments. FILL is the default mode of polygon rasterization, corresponding to the description in sections 14.6.1, and 14.6.3. Note that these modes affect only the final rasterization of polygons: in particular, and the polygon is clipped and possibly culled before these modes are applied. Polygon antialiasing applies only to the FILL state of PolygonMode. For POINT or LINE, point antialiasing or line segment antialiasing, respectively, apply.

14.6.5

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 with the commands void PolygonOffsetClamp( float factor, float units, float clamp ); OpenGL 4.6 (Core Profile) - May 5, 2022

14.6. POLYGONS

481

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, which may then be clamped to a minimum or maximum value specified by clamp. The values factor, units, and clamp may each be positive, negative, or zero. Calling the command PolygonOffset is equivalent to calling the command PolygonOffsetClamp with clamp equal to zero. The maximum depth slope m of a triangle is s ∂zw 2 ∂zw 2 m= + (14.11) ∂xw ∂yw where (xw , yw , zw ) is a point on the triangle. m may be approximated as ∂zw ∂zw m = max , . (14.12) ∂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   clamp = 0 or N aN m × f actor + r × units, o = min(m × f actor + r × units, clamp), clamp > 0   max(m × f actor + r × units, clamp), clamp < 0 OpenGL 4.6 (Core Profile) - May 5, 2022

(14.13)

14.6. POLYGONS

482

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. The state required for polygon offset consists of three boolean state values, POLYGON_OFFSET_POINT, POLYGON_OFFSET_LINE, and POLYGON_OFFSET_FILL, which determine whether polygon offset o is applied during the rasterization of polygons in POINT, LINE, and FILL modes, respectively, and a floating-point state value, POLYGON_OFFSET_CLAMP, specifying the offset clamp. The initial value of each of the offset enables is FALSE, and the initial value of the offset clamp is 0.0. If polygon offset point, line, or fill is enabled, o is added to the depth value of each fragment produced by the rasterization of a polygon in POINT, LINE, or FILL mode, respectively. 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.

14.6.6

Polygon Multisample Rasterization

If MULTISAMPLE is enabled, and the value of SAMPLE_BUFFERS is one, then polygons are rasterized using the following algorithm, regardless of whether polygon antialiasing (POLYGON_SMOOTH) is enabled or disabled. Polygon rasterization produces a fragment for each framebuffer pixel with one or more sample points that satisfy the point sampling criteria described in section 14.6.1. 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 14.6.1, 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 noperspective and flat qualifiers affect how shader outputs are interpolated in the same fashion as described for basic polygon rasterization in section 14.6.1. The rasterization described above applies only to the FILL state of PolygonMode. For POINT and LINE, the rasterizations described in sections 14.4.3 (Point Multisample Rasterization) and 14.5.4 (Line Multisample Rasterization) apply.

OpenGL 4.6 (Core Profile) - May 5, 2022

14.7.

14.6.7

483

Polygon Rasterization State

The state required for polygon rasterization consists of the current state of polygon antialiasing (enabled or disabled), the current value of the PolygonMode setting, whether point, line, and fill mode polygon offsets are enabled or disabled, and the factor and bias values of the polygon offset equation. The initial setting of polygon antialiasing is disabled. The initial state for PolygonMode is FILL . The initial polygon offset factor and bias values are both 0; initially polygon offset is disabled for all modes.

14.7 This section is only defined in the compatibility profile.

14.8 This section is only defined in the compatibility profile.

14.9

Early Per-Fragment Tests

Once fragments are produced by rasterization, a number of per-fragment operations are performed prior to fragment shader execution (see section 15). If a fragment is discarded during any of these operations, it will not be processed by any subsequent stage, including fragment shader execution. 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 14.9.1); • the scissor test (see section 14.9.2); • multisample fragment operations (see section 14.9.3); If early per-fragment operations are enabled, these tests are also performed: • the stencil test (see section 17.3.3); • the depth buffer test (see section 17.3.4); and • occlusion query sample counting (see section 17.3.5).

OpenGL 4.6 (Core Profile) - May 5, 2022

14.9. EARLY PER-FRAGMENT TESTS

14.9.1

484

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 17.4.1), 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.

14.9.2

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 ScissorArrayv( uint first, sizei count, const int *v ); void ScissorIndexed( uint index, int left, int bottom, sizei width, sizei height ); void ScissorIndexedv( uint index, int *v ); void Scissor( int left, int bottom, sizei width, sizei height ); ScissorArrayv defines a set of scissor rectangles that are each applied to the corresponding viewport (see section 13.8.1). first specifies the index of the first scissor rectangle to modify, and count specifies the number of scissor rectangles. v contains the address of an array of integers containing the left, bottom, width and height of the scissor rectangles, in that order. If left ≤ xw < left + width and bottom ≤ yw < bottom + height for the selected scissor rectangle, then the scissor test passes. Otherwise, the test fails and the fragment is discarded. For points, lines, and polygons, the scissor rectangle for a primitive is selected in the same manner as the viewport (see section 13.8.1). For buffer clears (see section 17.4.3), the scissor rectangle numbered zero is used for the scissor test. Errors An INVALID_VALUE error is generated by ScissorArrayv if first +count

OpenGL 4.6 (Core Profile) - May 5, 2022

14.9. EARLY PER-FRAGMENT TESTS

485

is greater than the value of MAX_VIEWPORTS. An INVALID_VALUE error is generated by ScissorArrayv if count is negative. An INVALID_VALUE error is generated by ScissorIndexed and Scissor if width or height is negative. The scissor test is enabled or disabled for all viewports using Enable or Disable with target SCISSOR_TEST. The test is enabled or disabled for a specific viewport using Enablei or Disablei with the constant SCISSOR_TEST and the index of the selected viewport. When disabled, it is as if the scissor test always passes. The value of the scissor test enable for viewport i can be queried by calling IsEnabledi with target SCISSOR_TEST and index i. The value of the scissor test enable for viewport zero may also be queried by calling IsEnabled with the same target, but no index parameter. Errors An INVALID_VALUE error is generated by Enablei, Disablei and IsEnabledi if target is SCISSOR_TEST and index is greater than or equal to the value of MAX_VIEWPORTS. The state required consists of four integer values per viewport, and a bit indicating whether the test is enabled or disabled for each viewport. In the initial state, left = bottom = 0, and width and height are determined by the size of the window into which the GL is to do its rendering for all viewports. 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 for all viewports. ScissorIndexed and ScissorIndexedv specify the scissor rectangle for a single viewport and are equivalent (assuming no errors are generated) to: int v[] = { lef t, bottom, width, height }; ScissorArrayv(index, 1, v); and ScissorArrayv(index, 1, v); respectively. Scissor sets the scissor rectangle for all viewports to the same values and is equivalent (assuming no errors are generated) to: OpenGL 4.6 (Core Profile) - May 5, 2022

14.9. EARLY PER-FRAGMENT TESTS

486

for (uint i = 0; i < MAX_VIEWPORTS; i++) { ScissorIndexed(i, lef t, bottom, width, height); } Calling Enable or Disable with target SCISSOR_TEST is equivalent, assuming no errors, to: for (uint i = 0; i < MAX_VIEWPORTS; i++) { Enablei(SCISSOR_TEST, i); /* or */ Disablei(SCISSOR_TEST, i); }

14.9.3

Multisample Fragment Operations

This step modifies fragment coverage values based on the values of SAMPLE_COVERAGE, SAMPLE_COVERAGE_VALUE, SAMPLE_COVERAGE_INVERT, SAMPLE_MASK, SAMPLE_MASK_VALUE, and an output sample mask optionally written by the fragment shader. If MULTISAMPLE is disabled, or 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, index zero (see section 15.2.3). 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. Next, if SAMPLE_COVERAGE is enabled, the fragment coverage 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. OpenGL 4.6 (Core Profile) - May 5, 2022

14.9. EARLY PER-FRAGMENT TESTS

487

The values of SAMPLE_COVERAGE_VALUE and SAMPLE_COVERAGE_INVERT are specified by calling 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 23.11. 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 32 × M + B as described in section 14.3.1. The sample mask value is queried by calling GetIntegeri v with target set to SAMPLE_MASK_VALUE and the 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.

14.9.4

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 15.2.4). 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 17.3. If early fragment tests are enabled, any depth value computed by the fragment shader has no effect. Additionally, 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 4.6 (Core Profile) - May 5, 2022

Chapter 15

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 14). 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 17.

15.1

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 7.11. Fragment shaders can read input variables or inputs that correspond to the attributes of the fragments produced by rasterization. The OpenGL 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 Shading Language Specification). The values of these user-defined outputs are, if not flat shaded, interpolated across 488

15.2. SHADER EXECUTION

489

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 MULTISAMPLE is enabled, 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 the value of MIN_FRAGMENT_INTERPOLATION_OFFSET or greater than the value of 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 value of 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 15.2.3 describes how to direct these values to buffers.

15.2

Shader Execution

If there is an active program object present for the fragment stage, the executable code for that program is used to process incoming fragments that are the result of rasterization. Implementations are allowed to skip the execution of certain fragment shader invocations, and to execute additional fragment shader invocations during programmable fragment processing due to implementation dependent reasons, including the execution of fragment shader invocations when there is not an active program object present for the fragment shader stage, as long as the results of render-

OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

490

ing otherwise remain unchanged. Following shader execution, the fixed-function operations described in chapter 17 are performed. Special considerations for fragment shader execution are described in the following sections.

15.2.1

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 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 15.1, 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, respectively. The resulting four-component vector (Rs , Gs , Bs , As ) is returned to the fragdu ment shader. For the purposes of level-of-detail calculations, the derivatives du dx , dy , dv dv dw dw dx , dy , dx and dy may be approximated by a differencing algorithm as described in section 8.13.1(“Derivative Functions”) of the OpenGL Shading Language Specification. OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION Texture Base Internal Format RED RG RGB RGBA

491 Texture base color Cb Ab (Rt , 0, 0) 1 (Rt , Gt , 0) 1 (Rt , Gt , Bt ) 1 (Rt , Gt , Bt ) At

Table 15.1: Correspondence of filtered texture components to texture base components.

Texture lookups involving textures with depth and/or stencil component data are performed as described in section 11.1.3.5.

15.2.2

Shader Inputs

The OpenGL Shading Language Specification describes the values that are available as inputs to the fragment shader. The built-in variable gl_FragCoord holds the fragment coordinate xf yf zf wf for the fragment. Computing the fragment coordinate depends on the fragment processing pixel-center and origin conventions (discussed below) as follows: ( xw − 21 , pixel-center convention is integer xf = xw , otherwise ( H − yw , origin convention is upper-left yf0 = yw , otherwise ( yf0 − 12 , pixel-center convention is integer yf = yf0 otherwise

(15.1)

zf = zw 1 wf = wc where xw yw zw is the fragment’s window-space position, wc is the w component of the fragment’s clip-space position (see section 13.8), and H is the window’s height in pixels. Note that zw already has a polygon offset added in, if enabled (see section 14.6.5). zf must be precisely zero or one in the case where zw is either zero or one, respectively. OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

492

Unless otherwise specified by layout qualifiers in the fragment shader (see section 4.4.1.3(“Fragment Shader Inputs”) of the OpenGL Shading Language Specification), the fragment processing pixel-center convention is half-integer and the fragment processing origin convention is lower left. 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 (including ones resulting from primitives rendered as points or lines), the determination is made by examining the sign of the area computed by equation 14.8 of section 14.6.1 (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. For polygons drawn in point or line mode, the primitive ID counter is incremented only once, even though multiple points or lines may be 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_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 this variable in a fragment shader causes the entire shader to be evaluated per-sample. When rendering to a non-multisample buffer, or if multisample rasterization is disabled, gl_SampleID will always be zero. gl_NumSamples is the sample count of the framebuffer regardless of whether multisample rasterization is enabled 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 coordinates of the center of the pixel are always (0.5, 0.5). Using this variable in a fragment shader causes the entire shader to be evaluated per-sample. When rendering to a non-multisample buffer, or if multisample rasterization is disabled, gl_SamplePosition will alOpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

493

ways 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, or if multisample rasterization is disabled, 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 14.9.3). 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 11.3.4.5), 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. 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).

15.2.3

Shader Outputs

The OpenGL 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_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 13.8.1). For floating-point depth buffers, conversion OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

494

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 sample fragment operations, as described in section 17.3.10, 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 evaluated 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. 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_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. The binding of a user-defined output variable to components of a fragment color number can be specified explicitly in the shader text or SPIR-V shader, or using the command void BindFragDataLocationIndexed( uint program, uint colorNumber, uint index, const char * name ); OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

495

specifies that the output variable name in program should be bound to fragment color colorNumber when the program is next linked. index may be zero or one to specify that the color will be used as either the first or second color input to the blend equation, respectively, as described in section 17.3.6. If name was bound previously, its assigned binding is replaced with colorNumber. name must be a null-terminated string. 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 one, if colorNumber is greater than or equal to the value of MAX_DRAW_BUFFERS and index is zero, or if colorNumber is greater than or equal to the value of MAX_DUAL_SOURCE_DRAW_BUFFERS and index is equal to one. The command void BindFragDataLocation( uint program, uint colorNumber, const char * name ); is equivalent to BindFragDataLocationIndexed(program, colorN umber, 0, name); BindFragDataLocation has no effect until the program is linked. In particular, it doesn’t modify the bindings of outputs in a program that has already been linked. Errors An INVALID_OPERATION error is generated if name starts with the reserved gl_ prefix. When a program is linked, each active user-defined fragment shader output variable will have a binding consisting of a fragment color number, a fragment color index, and a component index. Output variables declared with location, component, or index layout qualifiers will use the values specified in the shader text. For SPIR-V shaders, these are specified by the Location, Component, and Index decorations. Output variables without such layout or decoration qualifiers OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

496

will use the bindings specified by BindFragDataLocationIndexed or BindFragDataLocation, if any. BindFragDataLocation* has no effect on SPIR-V shaders, since the locations must always be fully specified as described in section 15.2.3.1. Otherwise, the linker will automatically assign a fragment color number, using any color number not already assigned to another active fragment shader output variable. The fragment color index and component index of an output variable binding will default to zero unless values are explicitly specified by a layout qualifer or BindFragDataLocationIndexed. The properties of an active fragment shader output variable binding can be queried using the command GetProgramResourceiv with a programInterface of PROGRAM_OUTPUT and props values of LOCATION, LOCATION_INDEX, and LOCATION_COMPONENT. When a fragment shader terminates, the value of each active user-defined output variable is written to components of the fragment color output to which it is bound. The set of fragment color components written is determined according to the variable’s data type and component index binding, using the mappings in table 11.1. For an output variable declared as an array bound to fragment color number i, individual active array elements are written to consecutive fragment color numbers beginning with i, with the components written determined from the array element’s data type and the array variable’s component index binding. Output binding assignments will cause LinkProgram to fail: • if the number of active outputs is greater than the value of MAX_DRAW_BUFFERS; • if the program has an active output assigned to a location greater than or equal to the value of MAX_DUAL_SOURCE_DRAW_BUFFERS and has an active output assigned an index greater than or equal to one; • if two output variables are bound to the same output number and index with overlapping components selected; • if two output variables with different component types (signed integer, unsigned integer, or floating-point) are bound to the same output number, even if selected components do not overlap; or • if the explicit binding assigments do not leave enough space for the linker to automatically assign a location for an output array, which requires multiple contiguous locations. BindFragDataLocationIndexed may be issued before any shader objects are attached to a program object. Hence it is allowed to bind any name (except a name starting with gl_) to a color number and index, including a name that is never used OpenGL 4.6 (Core Profile) - May 5, 2022

15.2. SHADER EXECUTION

497

as an output in any fragment shader object. Assigned bindings for variables that do not exist are ignored. 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 commands int GetFragDataLocation( uint program, const char *name ); int GetFragDataIndex( uint program, const char *name ); are provided to query the location and fragment color index assigned to a fragment shader output variable. Errors If program has been linked successfully but contains no fragment shader, 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 commands are equivalent to GetProgramResourceLocation(program, PROGRAM_OUTPUT, name); and GetProgramResourceLocationIndex(program, PROGRAM_OUTPUT, name); respectively. 15.2.3.1

SPIR-V Fragment Output Interface

When a SPIR-V fragment stage is present, the variables listed by OpEntryPoint with the Output storage class form the fragment output interface. These variables must be decorated with a Location They can also be decorated with a Component and/or an Index. User-defined fragment shader output variables are matched only by their Location, Component, and Index decorations. If two outputs are placed within the same location, they must have the same underlying type (floating-point or integer). No component aliasing of output variables is allowed. That is, there must OpenGL 4.6 (Core Profile) - May 5, 2022

15.3. FRAGMENT SHADER QUERIES

498

not be two output variables which have the same location, component, and index, either explicitly declared or implied. Output values written by a fragment shader must be declared with either OpTypeFloat or OpTypeInt, and a Width of 32. Composites of these types are also permitted.

15.2.4

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 14.9 will be performed prior to fragment shader execution. Otherwise, they will be performed after fragment shader execution.

15.3

Fragment Shader Queries

Fragment shader queries use query objects to track the number of fragment shader invocations. When BeginQuery is called with a target of FRAGMENT_SHADER_INVOCATIONS, the fragment shader invocations count maintained by the GL is set to zero. When a fragment shader invocations query is active, the counter is incremented every time the fragment shader is invoked (see section 15.2). The result of fragment shader queries may be implementation dependent due to reasons described in section 15.2.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 16

This chapter is only defined in the compatibility profile.

499

Chapter 17

Writing Fragments and Samples to the Framebuffer After programmable fragment processing, the following fixed-function operations are applied to the resulting fragments: • Antialiasing application (see section 17.1). • Multisample point fade (see section 17.2). • Per-fragment operations and writing to the framebuffer (see section 17.3). Writing to the framebuffer 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 17.4.

17.1

Antialiasing Application

If antialiasing is enabled for the primitive from which a rasterized fragment was produced, then the computed coverage value is applied to the fragment. The value is multiplied by the fragment’s alpha value to yield a final alpha value. The coverage value is applied separately to each fragment color, and only applied if the corresponding color buffer in the framebuffer has a fixed- or floating-point format.

17.2

Multisample Point Fade

If multisampling is enabled and the rasterized fragment results from a point primitive, then the computed fade factor from equation 14.2 is applied to the fragment. 500

17.3. PER-FRAGMENT OPERATIONS

501

The fade factor is multiplied by the fragment’s alpha value to yield a final alpha value. The fade factor is applied separately to each fragment color, and only applied if the corresponding color buffer in the framebuffer has a fixed- or floatingpoint format.

17.3

Per-Fragment Operations

A fragment is produced by rasterization with window coordinates of (xw , yw ) and depth z, as described in chapter 14. The fragment is then modified by programmable fragment processing, which adds associated data as described in chapter 15. The fragment is then further modified, and possibly discarded by the perfragment operations described in this chapter. These operations are diagrammed in figure 17.1, 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 17.3.3, 17.3.4, and 17.3.5 may instead be performed prior to fragment processing, as described in section 14.9, if requested by the fragment program.

17.3.1

Alpha To Coverage

This step modifies fragment alpha and coverage values based on the values of SAMPLE_ALPHA_TO_COVERAGE and SAMPLE_ALPHA_TO_ONE. If MULTISAMPLE is disabled, if the value of SAMPLE_BUFFERS is not one, or if draw buffer zero is not NONE and the buffer it references has an integer format, these operations are skipped. Alpha to coverage and alpha to one operations are enabled or disabled by calling Enable and Disable with targets SAMPLE_ALPHA_TO_COVERAGE or SAMPLE_ALPHA_TO_ONE, respectively. All alpha values in this section refer only to the alpha component of the fragment shader output linked to color number zero, index zero (see section 15.2.3). 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 14.3.1). 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 14.9.3), 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

Fragment (or sample) from Rasterization

Fragment Shader

Blending

Fragment (or sample) and Associated Data

Occlusion Query (*)

Alpha To Coverage Operations

Depth Buffer Test (*)

Framebuffer

SRGB Conversion

502

Dithering

Stencil Test (*)

Framebuffer

Logicop

Additional Multisample Fragment Operations

Framebuffer Write To Framebuffer

Figure 17.1. Per-fragment operations. The boxes labelled with “(*)” may instead be performed during early per-fragment operations, as described in section 14.9.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

503

Next, if SAMPLE_ALPHA_TO_ONE is enabled, each alpha value is replaced by the maximum representable alpha value for fixed-point color buffers, or by 1.0 for floating-point buffers. Otherwise, the alpha values are not changed.

17.3.2 This subsection is only defined in the compatibility profile.

17.3.3

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 target 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. For the purposes of stencil testing, a primitive is still considered a polygon even if the polygon is to be rasterized as points or lines due to the current polygon mode. Whether a polygon is front- or back-facing is determined in the same manner used for face culling (see section 14.6.1). 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 OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

504

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 17.3.4) 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 values are both zero, the front and back stencil comparison functions are both ALWAYS, and the front and back stencil masks 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. 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.

17.3.4

Depth Buffer Test

The depth buffer test discards the incoming fragment if a depth comparison fails. The comparison is enabled or disabled with the generic Enable and Disable commands using target DEPTH_TEST. When disabled, the depth comparison and subOpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

505

sequent 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 depth clamping (see section 13.7) is enabled, before the incoming fragment’s zw is compared zw is clamped to the range [min(n, f ), max(n, f )], where n and f are the current near and far depth range values (see section 13.8.1) 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.

17.3.5

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 of SAMPLES_PASSED, ANY_SAMPLES_PASSED, or ANY_SAMPLES_PASSED_CONSERVATIVE. When an occlusion query is started with target SAMPLES_PASSED, the samples-passed count maintained by the GL is set to zero. When an occlusion query is active, the samples-passed count is incremented for each fragment still being processed after the depth test. If the value of SAMPLE_BUFFERS is zero, then the samples-passed count is incremented by one for each fragment. If the value of SAMPLE_BUFFERS is one, then the samples-passed count is incremented OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

506

by the number of samples whose coverage bit is set. However, implementations, at their discretion, may instead increase the samples-passed count by the value of SAMPLES if any sample in the fragment is covered. When an occlusion query finishes and all fragments generated by commands issued prior to EndQuery have been generated, the samples-passed count is written to the corresponding query object as the query result value, and the query result for that object is marked as available. 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.

17.3.6

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] or [−1, 1] respectively for an unsigned normalized or signed normalized color buffer prior to evaluating the blend equation. If the color buffer is floating-point, 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 ); OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

507

target is the symbolic constant BLEND and index is an integer i specifying the draw buffer associated with the symbolic constant DRAW_BUFFERi. If the color buffer associated with DRAW_BUFFERi is one of FRONT, BACK, LEFT, RIGHT, or FRONT_AND_BACK (specifying multiple color buffers), then the state enabled or disabled is applicable for all of the buffers. 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, or if logical operation on color values is enabled (section 17.3.9), proceed to the next operation. If multiple fragment colors are being written to multiple buffers (see section 17.4.1), blending is computed and applied separately for each fragment color and the corresponding buffer. 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. 17.3.6.1

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 equation while modeAlpha determines the alpha blend equation. BlendEquation and BlendEquationi argument mode determines both the RGB and alpha blend equations. mode, modeRGB, and modeAlpha must be one of the blend equation modes in table 17.1. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

508

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 mode, modeRGB, or modeAlpha are not one of the blend equation modes in table 17.1. Signed or unsigned normalized fixed-point destination (framebuffer) components are represented as described in section 2.3.5. Constant color components, floating-point 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, signed and unsigned normalized fixed-point color components undergo an implied conversion to floating-point using equations 2.2 and 2.1, respectively. This conversion must leave the values zero and one invariant. Blending computations are treated as if carried out in floating-point, and will be performed with a precision and dynamic range no lower than that used to represent destination components. If FRAMEBUFFER_SRGB is enabled and 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 fixed-point 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 8.24. If FRAMEBUFFER_SRGB is disabled or 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 17.1 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 comOpenGL 4.6 (Core Profile) - May 5, 2022

17.3. 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 )

509 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 17.1: RGB and alpha blend equations.

ponents 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 functions are described below. 17.3.6.2

Blend Functions

The weighting factors used by the blend equation are determined by the blend functions. There are four possible sources for weighting factors. These are the constant color (Rc , Gc , Bc , Ac ) set with BlendColor (see below), the first source color (Rs0 , Gs0 , Bs0 , As0 ), the second source color (Rs1 , Gs1 , Bs1 , As1 ), 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 OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

510

void BlendFunci( uint buf, enum src, enum dst ); 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 17.2. 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 17.2. 17.3.6.3

Dual Source Blending and Multiple Draw Buffers

Blend functions that require the second color input, (Rs1 , Gs1 , Bs1 , As1 ) (SRC1_COLOR, SRC1_ALPHA, ONE_MINUS_SRC1_COLOR, or ONE_MINUS_SRC1_ALPHA) may consume hardware resources that could otherwise be used for rendering to multiple draw buffers. Therefore, the number of draw buffers that can be attached to a framebuffer may be lower when using dual-source blending. The maximum number of draw buffers that may be attached to a single framebuffer when using dual-source blending functions is implementation-dependent and may be queried by calling GetIntegerv with pname MAX_DUAL_SOURCE_DRAW_BUFFERS. When using dual-source blending, MAX_DUAL_SOURCE_DRAW_BUFFERS should be used in place of MAX_DRAW_BUFFERS to determine the maximum number of draw buffers that may be attached to a single framebuffer. The value of MAX_DUAL_SOURCE_DRAW_BUFFERS must be at least 1. If the value of MAX_DUAL_SOURCE_DRAW_BUFFERS is 1, then dual-source blending and multiple draw buffers cannot be used simultaneously.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. 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 SRC1_COLOR ONE_MINUS_SRC1_COLOR SRC1_ALPHA ONE_MINUS_SRC1_ALPHA

RGB Blend Factors (Sr , Sg , Sb ) or (Dr , Dg , Db ) (0, 0, 0) (1, 1, 1) (Rs0 , Gs0 , Bs0 ) (1, 1, 1) − (Rs0 , Gs0 , Bs0 ) (Rd , Gd , Bd ) (1, 1, 1) − (Rd , Gd , Bd ) (As0 , As0 , As0 ) (1, 1, 1) − (As0 , As0 , As0 ) (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 (Rs1 , Gs1 , Bs1 ) (1, 1, 1) − (Rs1 , Gs1 , Bs1 ) (As1 , As1 , As1 ) (1, 1, 1) − (As1 , As1 , As1 )

511

Alpha Blend Factor Sa or Da 0 1 As0 1 − As0 Ad 1 − Ad As0 1 − As0 Ad 1 − Ad Ac 1 − Ac Ac 1 − Ac 1 As1 1 − As1 As1 1 − As1

Table 17.2: 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 ). s0 d

OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

512

Errors An INVALID_OPERATION error is generated by any command that transfers vertices to the GL if either blend function requires the second color input for any draw buffer, and any draw buffers greater than or equal to the value of MAX_DUAL_SOURCE_DRAW_BUFFERS have values other than NONE. 17.3.6.4

Generation of Second Color Source for Blending

When using a fragment shader with dual-source blending functions, the color outputs are bound to the first and second inputs of the blender using BindFragDataLocationIndexed as described in section 15.2.3. Data written to the first of these outputs becomes the first source color input to the blender (corresponding to SRC_COLOR and SRC_ALPHA). Data written to the second of these outputs generates the second source color input to the blender (corresponding to SRC1_COLOR and SRC1_ALPHA). If the second color input to the blender is not written in the shader, or if no output is bound to the second input of a blender, the result of the blending operation is not defined. 17.3.6.5

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. 17.3.6.6

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 OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

513

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 23.21 and index i. 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 (section 17.4.1) using each buffer’s color for Cd . If a color buffer has no A value, then Ad is taken to be 1.

17.3.7

sRGB Conversion

If FRAMEBUFFER_SRGB is enabled and 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 = (17.1) 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_SRGB is disabled or the value of 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. 1

Note that only unsigned normalized fixed-point color buffers may be SRGB-encoded. Signed normalized fixed-point + SRGB encoding is not defined.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS

17.3.8

514

Dithering

Dithering selects between two representable color values. 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, 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.

17.3.9

Logical Operation

Finally, a logical operation is applied between the incoming fragment’s color values and the color values stored at the corresponding location in the framebuffer. The result replaces the values in the framebuffer at the fragment’s (xw , yw ) coordinates. The logical operation on color values is enabled or disabled by calling Enable or Disable with target COLOR_LOGIC_OP. If the logical operation is enabled for color values, it is as if blending were disabled, regardless of the value of BLEND. If multiple fragment colors are being written to multiple buffers (see section 17.4.1), the logical operation is computed and applied separately for each fragment color and the corresponding buffer. Logical operation has no effect on a floating-point destination color buffer, or when FRAMEBUFFER_SRGB is enabled and the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the framebuffer attachment corresponding to the destination buffer is SRGB (see section 9.2.3). However, if logical operation is enabled, blending is still disabled. The logical operation is selected by void LogicOp( enum op ); op must be one of the logicop modes in table 17.3, which also describes the resulting operation when that mode is selected. s is the value of the incoming fragment OpenGL 4.6 (Core Profile) - May 5, 2022

17.3. PER-FRAGMENT OPERATIONS Logicop Mode CLEAR AND AND_REVERSE COPY AND_INVERTED NOOP XOR OR NOR EQUIV INVERT OR_REVERSE COPY_INVERTED OR_INVERTED NAND SET

515 Operation 0 s∧d s ∧ ¬d s ¬s ∧ d d s xor d s∨d ¬(s ∨ d) ¬(s xor d) ¬d s ∨ ¬d ¬s ¬s ∨ d ¬(s ∧ d) all 1’s

Table 17.3: Logical operation op arguments to LogicOp and their corresponding operations.

and d is the value stored in the framebuffer. Logical operations are performed independently for each red, green, blue, and alpha value of each color buffer that is selected for writing. The required state is an integer indicating the logical operation, and a bit indicating whether the logical operation is enabled or disabled. The initial state is for the logic operation to be given by COPY, and to be disabled. Errors An INVALID_VALUE error is generated if op is not one of the logicop modes in table 17.3.

17.3.10

Additional Multisample Fragment Operations

If the DrawBuffer mode (see section 17.4.1) is NONE, no change is made to any multisample or color buffer. Otherwise, fragment processing is as described below. If MULTISAMPLE is enabled, and the value of SAMPLE_BUFFERS is one, the

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

516

stencil test, depth test, blending, dithering, and logical operations 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 MULTISAMPLE is disabled, and the value of SAMPLE_BUFFERS is one, the fragment may be treated exactly as described above, with optimization possible because the fragment coverage must be set to full coverage. Further optimization is allowed, however. An implementation may choose to identify a centermost sample, and to perform stencil and depth tests on only that sample. Regardless of the outcome of the stencil test, all multisample buffer stencil sample values are set to the appropriate new stencil value. If the depth test passes, all multisample buffer depth sample values are set to the depth of the fragment’s centermost sample’s depth value, and all multisample buffer color sample values are set to the color value of the incoming fragment. Otherwise, no change is made to any multisample buffer color or depth value. 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 buffers selected by DrawBuffer or 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 17.3.6). Otherwise, a simple average computed independently for each color component is recommended.

17.4

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.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

17.4.1

517

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 either *DrawBuffer or *DrawBuffers commands described below. The set of buffers of a framebuffer object to which fragment color zero is written is controlled with the commands void DrawBuffer( enum buf ); void NamedFramebufferDrawBuffer( uint framebuffer, enum buf ); For DrawBuffer, the framebuffer object is that bound to the DRAW_For NamedFramebufferDrawBuffer, framebuffer is zero or the name of a framebuffer object. If framebuffer is zero, then the default draw framebuffer is affected. If the default framebuffer is affected (see section 9), buf must be one of the values listed in table 17.4. In this case, buf is a symbolic constant specifying zero, one, two, or four buffers for writing. These constants refer to the four potentially visible buffers (front left, front right, back left, and back right). Arguments that omit reference to LEFT or RIGHT refer to both left and right buffers. Arguments that omit reference to FRONT or BACK refer to both front and back buffers. If a framebuffer object is affected, buf must be one of the values listed in table 17.5, which summarizes the constants and the buffers they indicate. In this case, buf specifies a single color buffer for writing. Specifying COLOR_ATTACHMENTi enables drawing only to the image attached to the framebuffer at that attachment point. FRAMEBUFFER binding.

Errors An INVALID_OPERATION error is generated by NamedFramebufferDrawBuffer if framebuffer is not zero or the name of an existing framebuffer object. An INVALID_ENUM error is generated if buf is not one of the values in tables 17.5 or 17.4. An INVALID_OPERATION error is generated if the default framebuffer is affected and buf is a value (other than NONE) that does not indicate one of the color buffers allocated to the default framebuffer. An INVALID_OPERATION error is generated if a framebuffer object is affected and buf is one of the constants from table 17.4 (other than NONE), or

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS Symbolic Constant

Front Left

NONE FRONT_LEFT FRONT_RIGHT BACK_LEFT BACK_RIGHT FRONT BACK LEFT RIGHT FRONT_AND_BACK

Front Right

518 Back Left

Back Right

• • • • •

• • •

• •

•

• • •

Table 17.4: Arguments to DrawBuffer when the context is bound to a default framebuffer, and the buffers they indicate. The same arguments are valid for ReadBuffer, but only a single buffer is selected as discussed in section 18.2.1.

COLOR_ATTACHMENTm and m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS.

DrawBuffer will set the draw buffer for fragment colors other than zero to NONE.

The set of buffers of a framebuffer object to which all fragment colors are written is controlled with the commands The commands void DrawBuffers( sizei n, const enum *bufs ); Symbolic Constant NONE COLOR_ATTACHMENTi (see caption)

Meaning No buffer Output fragment color to image attached at color attachment point i

Table 17.5: Arguments to DrawBuffer(s) 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS Symbolic Constant NONE FRONT_LEFT FRONT_RIGHT BACK_LEFT BACK_RIGHT

Front Left

Front Right

519 Back Left

Back Right

• • • •

Table 17.6: Arguments to DrawBuffers when the default framebuffer is affected, and the buffers they indicate.

void NamedFramebufferDrawBuffers( uint framebuffer, sizei n, const enum *bufs ); For DrawBuffers, the framebuffer object is that bound to the DRAW_FRAMEBUFFER binding. For NamedFramebufferDrawBuffers, framebuffer is the name of the framebuffer object. If framebuffer is zero, then the default framebuffer is affected. n specifies the number of buffers in bufs. bufs is a pointer to an array of values specifying the buffer to which each fragment color is written. Each buffer listed in bufs must be one of the values from tables 17.5 or 17.6. Further, acceptable values for the constants in bufs depend on whether the default framebuffer or a framebuffer object is affected. For more information about framebuffer objects, see section 9. If the default framebuffer is affected, then each of the constants must be one of the values listed in table 17.6 or the special value BACK. When BACK is used, n must be 1 and color values are written into the left buffer for single-buffered contexts, or into the back left buffer for double-buffered contexts. If a framebuffer object is affected, then each of the constants must be one of the values listed in table 17.5. 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 may be queried by calling GetIntegerv with pname MAX_DRAW_BUFFERS. Except for NONE, a buffer may not appear more than once in the array pointed to by bufs.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

520

If a fragment shader writes to a user-defined output variable, DrawBuffers specifies a set of draw buffers into which each of the multiple output colors defined by these variables are separately written. If a fragment shader writes to no user-defined output variables, the values of the fragment colors following shader execution are undefined, and may differ for each fragment color. If some, but not all user-defined output variables are written, the values of fragment colors corresponding to unwritten variables are similarly 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_OPERATION error is generated by NamedFramebufferDrawBuffers if framebuffer is not zero or the name of an existing framebuffer object. 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 17.5 or 17.6. An INVALID_OPERATION error is generated if a buffer other than NONE is specified more than once in the array pointed to by bufs. An INVALID_ENUM error is generated if any value in bufs is FRONT, LEFT, RIGHT, or FRONT_AND_BACK. This restriction applies to both the default framebuffer and framebuffer objects, and exists because these constants may themselves refer to multiple buffers, as shown in table 17.4. An INVALID_OPERATION error is generated for the default framebuffer if any value in bufs is BACK, and n is not one. An INVALID_OPERATION error is generated if the default framebuffer is affected and any value in bufs is a constant (other than NONE or BACK) that does not indicate one of the color buffers allocated to the default framebuffer. 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 DrawBuffer or DrawBuffers causes subsequent pixel color value writes to affect the indicated buffers. If a framebuffer

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

521

object is affected 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. Monoscopic contexts include only left buffers, while stereoscopic contexts include both left and right buffers. Likewise, single buffered contexts include only front buffers, while double buffered contexts include both front and back buffers. The type of context is selected at GL initialization. The state required to handle color buffer selection for each framebuffer is an 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 back buffer; FRONT if there is no back buffer; and NONE if no default framebuffer is associated with the context. 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 then zero is NONE. The draw buffer of the currently bound draw framebuffer selected for fragment color i may be queried by calling GetIntegerv with pname set to DRAW_BUFFERi. DRAW_BUFFER is equivalent to DRAW_BUFFER0.

17.4.2

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. For any i where the value of DRAW_BUFFERi is one of FRONT, BACK, LEFT, RIGHT, or FRONT_AND_BACK, specifying multiple color buffers, the mask is applied to all of the buffers. OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

522

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 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 17.3.3). 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS 17.4.2.1

523

Fine Control of Multisample Buffer Updates

When 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.

17.4.3

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 ); 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. The command void ClearDepth( double d ); 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 13.8.1. No conversion is applied when clearing a floating-point depth buffer. The command OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

524

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 17.3.7), and dithering. The masking operations described in section 17.4.2 are also applied. If a buffer is not present, then a Clear directed at that buffer has no effect. Unsigned normalized fixed-point and signed 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] or [−1, 1] respectively, 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 is undefined. 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. 17.4.3.1

Clearing Individual Buffers

Individual buffers of a framebuffer object may be cleared with the commands void ClearBuffer{if ui}v( enum buffer, int drawbuffer, const T *value ); void ClearNamedFramebuffer{if ui}v( uint framebuffer, enum buffer, int drawbuffer, const T *value ); For ClearBuffer*, the framebuffer object is the bound draw framebuffer object. For ClearNamedFramebuffer*, framebuffer is the name of the framebuffer object. If framebuffer is zero, the default draw framebuffer is affected. buffer and drawbuffer identify a buffer to clear, and value specifies the value or values to clear it to. The *fv, *iv, and *uiv forms of these commands 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 one of the possible values in tables 17.4 and 17.5 identifying one or more color buffers, each of which is cleared to the same value. Clamping OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

525

and conversion for fixed-point color buffers are performed in the same fashion as for 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 for ClearDepth. Only the *fv forms of these commands should be used to clear depth buffers; other forms do not 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 for ClearStencil. Only the *iv forms of these commands should be used to clear stencil buffers; other forms do not accept a buffer of STENCIL. Both depth and stencil buffers of a framebuffer object may be cleared with the commands void ClearBufferfi( enum buffer, int drawbuffer, float depth, int stencil ); void ClearNamedFramebufferfi( uint framebuffer, enum buffer, int drawbuffer, float depth, int stencil ); For ClearBufferfi, the framebuffer object is the bound draw framebuffer object. For ClearNamedFramebufferfi, framebuffer is the name of the framebuffer object. If framebuffer is zero, the default draw framebuffer is affected. 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 for ClearDepth. Masking of stencil for stencil buffers is performed in the same fashion as for ClearStencil. These commands are equivalent to clearing the depth and stencil buffers separately, but may be faster when a buffer of internal format DEPTH_STENCIL is being cleared. The same per-fragment and masking operations defined for Clear are applied. For all forms of ClearBuffer* and ClearNamedFramebuffer*, 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. Errors An INVALID_OPERATION error is generated by ClearNamedFramebuffer* if framebuffer is not zero or the name of an existing framebuffer ob-

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS

526

ject. An INVALID_ENUM error is generated by ClearBufferiv and ClearNamedFramebufferiv if buffer is not COLOR or STENCIL. An INVALID_ENUM error is generated by ClearBufferuiv and ClearNamedFramebufferuiv if buffer is not COLOR. An INVALID_ENUM error is generated by ClearBufferfv and ClearNamedFramebufferfv if buffer is not COLOR or DEPTH. An INVALID_ENUM error is generated by ClearBufferfi and ClearNamedFramebufferfi 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. 17.4.3.2

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 Clear*Buffer* 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.

17.4.4

Invalidating Framebuffer Contents

To signal that the GL need not preserve all contents of a framebuffer object (invalidating portions of every pixel or a subregion of pixels), use the commands void InvalidateSubFramebuffer( enum target, sizei numAttachments, const enum *attachments, int x, int y, sizei width, sizei height ); void InvalidateNamedFramebufferSubData( uint framebuffer, sizei numAttachments, const enum *attachments, int x, int y, sizei width, sizei height ); For InvalidateSubFramebuffer, the framebuffer object is that bound to target, which must be DRAW_FRAMEBUFFER, READ_FRAMEBUFFER or FRAMEBUFFER.

OpenGL 4.6 (Core Profile) - May 5, 2022

17.4. WHOLE FRAMEBUFFER OPERATIONS FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER. For InvalidateNamed-

FramebufferSubData, framebuffer is the name of the framebuffer object. If framebuffer is zero, the default draw framebuffer is affected. numAttachments indicates how many attachments are supplied in the attachments list. If an attachment is specified that does not exist in the framebuffer object, 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 (for the default framebuffer), or outside of the attachments of the framebuffer object, are ignored. If the framebuffer object is not complete, these commands may be ignored. Errors An INVALID_ENUM error is generated by InvalidateSubFramebuffer if target is not FRAMEBUFFER, DRAW_FRAMEBUFFER, or READ_FRAMEBUFFER. An INVALID_OPERATION error is generated by InvalidateNamedFramebufferData if framebuffer is not zero or the name of an existing framebuffer object. An INVALID_VALUE error is generated if numAttachments, width, or height is negative. An INVALID_ENUM error is generated if a framebuffer object is affected, and any element of attachments is not one of the values in table 9.2. An INVALID_OPERATION error is generated if attachments contains COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS. An INVALID_ENUM error is generated if the default framebuffer is affected, and any elements of attachments are not one of: • FRONT_LEFT, FRONT_RIGHT, BACK_LEFT, and BACK_RIGHT, identifying that specific buffer • COLOR, which is treated as BACK_LEFT for a double-buffered context and FRONT_LEFT for a single-buffered context • DEPTH, identifying the depth buffer • STENCIL, identifying the stencil buffer. The commands void InvalidateFramebuffer( enum target, OpenGL 4.6 (Core Profile) - May 5, 2022

527

17.4. WHOLE FRAMEBUFFER OPERATIONS

528

sizei numAttachments, const enum *attachments ); void InvalidateNamedFramebufferData( uint framebuffer, sizei numAttachments, const enum *attachments ); are equivalent to InvalidateSubFramebuffer(target, numAttachments, attachments, 0, 0, vw, vh); and InvalidateNamedFramebufferSubData(f ramebuf f er, numAttachments, attachments, 0, 0, vw, vh); respectively, where vw and vh are equal to the maximum viewport width and height, respectively, obtained by querying MAX_VIEWPORT_DIMS.

17.4.5 This subsection is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 18

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.

18.1 This section is only defined in the compatibility profile.

18.2

Reading Pixels

The method for reading pixels from the framebuffer and placing them in pixel pack buffer or client memory is diagrammed in figure 18.1. We describe the stages of the pixel reading process in the order in which they occur.

18.2.1

Selecting Buffers for Reading

When reading pixels from a color buffer of a framebuffer object, the buffer selected for reading is termed the read buffer, and is controlled with the commands void ReadBuffer( enum src ); void NamedFramebufferReadBuffer( uint framebuffer, enum src ); For ReadBuffer, the target framebuffer object is that bound to READ_FRAMEBUFFER. For NamedFramebufferReadBuffer, framebuffer is zero or the name of the target framebuffer object. If framebuffer is zero, then the default read framebuffer is affected. 529

18.2. READING PIXELS

530

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 18.1. Operation of ReadPixels. Operations in dashed boxes are not performed for all data formats. Depth and stencil pixel paths are not shown.

If the default framebuffer is affected (see section 9), src must be one of the values listed in table 17.4, including NONE. FRONT_AND_BACK, FRONT, and LEFT refer to the front left buffer, BACK refers to the back left buffer, and RIGHT refers to the front right buffer. Other constants correspond directly to the buffers that they name. The initial value of the read buffer for the default framebuffer is FRONT if there is no back buffer; BACK if there is a back buffer; and NONE if no default framebuffer is associated with the context. If a framebuffer object is affected, src must be one of the values listed in table 17.5, including NONE. Specifying COLOR_ATTACHMENTi enables reading from the image attached to the framebuffer at that attachment point. The initial value of the read buffer for framebuffer objects is COLOR_ATTACHMENT0. The read buffer of the currently bound read framebuffer may be queried by calling GetIntegerv with pname set to READ_BUFFER. Errors An INVALID_OPERATION error is generated by NamedFramebufferReadBuffer if framebuffer is not zero or the name of an existing framebuffer

OpenGL 4.6 (Core Profile) - May 5, 2022

18.2. READING PIXELS

531

object. An INVALID_ENUM error is generated if src is not one of the values in tables 17.4 or 17.5. An INVALID_OPERATION error is generated if the default framebuffer is affected and src is a value (other than NONE) that does not indicate any of the color buffers allocated to the default framebuffer. An INVALID_OPERATION error is generated if a framebuffer object is affected, and src is one of the constants from table 17.4 (other than NONE, or COLOR_ATTACHMENTm where m is greater than or equal to the value of MAX_COLOR_ATTACHMENTS).

18.2.2

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. Pixels are read with the commands 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 8.4.4. The pixel storage modes that apply to ReadPixels and other commands that query images (see section 8.11) are summarized in table 18.1. Errors 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.2. READING PIXELS Parameter Name PACK_SWAP_BYTES PACK_LSB_FIRST PACK_ROW_LENGTH PACK_SKIP_ROWS PACK_SKIP_PIXELS PACK_ALIGNMENT PACK_IMAGE_HEIGHT PACK_SKIP_IMAGES PACK_COMPRESSED_BLOCK_WIDTH PACK_COMPRESSED_BLOCK_HEIGHT PACK_COMPRESSED_BLOCK_DEPTH PACK_COMPRESSED_BLOCK_SIZE

532 Type boolean boolean integer integer integer integer integer integer integer integer integer integer

Initial Value FALSE FALSE

0 0 0 4 0 0 0 0 0 0

Valid Range TRUE/FALSE TRUE/FALSE [0, ∞) [0, ∞) [0, ∞) 1,2,4,8 [0, ∞) [0, ∞) [0, ∞) [0, ∞) [0, ∞) [0, ∞)

Table 18.1: PixelStore parameters pertaining to ReadPixels, GetCompressedTexImage and GetTexImage.

Preferred values for format and type may be determined by calling GetIntegerv with pnames IMPLEMENTATION_COLOR_READ_FORMAT and IMPLEMENTATION_COLOR_READ_TYPE, respectively. The preferred format may vary depending on the format of the selected read buffer of the currently bound read framebuffer. Errors 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 18.2.1) has no image attached. • the selected read buffer is NONE.

18.2.3

Obtaining Pixels from the Framebuffer

If the format is DEPTH_COMPONENT, then values are obtained from the depth buffer.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.2. READING PIXELS

533

If there is a multisample buffer (the value of SAMPLE_BUFFERS is one), then values are obtained from the depth samples in this buffer. It is recommended that the depth value of the centermost sample be used, though implementations may choose any function of the depth sample values at each pixel. If the format is DEPTH_STENCIL, then values are taken from both the depth buffer and the stencil buffer. type must be UNSIGNED_INT_24_8 or FLOAT_32_UNSIGNED_INT_24_8_REV. If there is a multisample buffer, then values are obtained from the depth and stencil samples in this buffer. It is recommended that the depth and stencil values of the centermost sample be used, though implementations may choose any function of the depth and stencil sample values at each pixel. If the format is STENCIL_INDEX, then values are taken from the stencil buffer. If there is a multisample buffer, then values are obtained from the stencil samples in this buffer. It is recommended that the stencil value of the centermost sample be used, though implementations may choose any function of the stencil sample values at each pixel. For all other formats, values are obtained from the color buffer selected by the read buffer. Errors An INVALID_ENUM error is generated if format is DEPTH_STENCIL and type is not UNSIGNED_INT_24_8 or FLOAT_32_UNSIGNED_INT_24_8_REV. An INVALID_OPERATION error is generated if format is DEPTH_COMPONENT and there is no depth buffer; if format is STENCIL_INDEX and there is no stencil buffer; or if format is DEPTH_STENCIL and either there is no depth buffer, or there is no stencil buffer. An INVALID_FRAMEBUFFER_OPERATION error is generated if the object bound to READ_FRAMEBUFFER_BINDING is not framebuffer complete (as defined in section 9.4.2). An INVALID_OPERATION error is generated if format selects a color buffer while the read buffer is NONE, or if the GL is using a framebuffer object (the value of READ_FRAMEBUFFER_BINDING is non-zero) and the read buffer selects an attachment that has no image attached. 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

OpenGL 4.6 (Core Profile) - May 5, 2022

18.2. READING PIXELS

534

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, GREEN, BLUE, RG, RGB, RGBA, BGR, or BGRA, then red, green, blue, and alpha values are obtained from the selected buffer at each pixel location. An INVALID_OPERATION error is generated if format is an integer format and the color buffer is not an integer format, or if the color buffer is an integer format and format is not an integer format. 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 8.11). If G, B, or A values are not present in the internal format, they are taken to be zero, zero, and one respectively.

18.2.4

Conversion of RGBA values

This step applies only if format is not STENCIL_INDEX, DEPTH_COMPONENT, or DEPTH_STENCIL. The R, G, B, and A values form a group of elements. For a signed or unsigned normalized fixed-point color buffer, each element is converted to floating-point using equations 2.2 or 2.1, respectively. For an integer or floating-point color buffer, the elements are unmodified.

18.2.5

Conversion of Depth values

This step applies only if format is DEPTH_COMPONENT or DEPTH_STENCIL and the depth buffer uses a fixed-point representation. An element is taken to be a fixed-point value in [0, 1] with m bits, where m is the number of bits in the depth buffer (see section 13.8.1). No conversion is necessary if the depth buffer uses a floating-point representation.

18.2.6 This subsection is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.2. READING PIXELS

535

18.2.7 This subsection is only defined in the compatibility profile.

18.2.8

Final Conversion

Read color clamping is controlled by calling void ClampColor( enum target, enum clamp ); with target set to CLAMP_READ_COLOR. If clamp is TRUE, read color clamping is enabled; if clamp is FALSE, read color clamping is disabled. If clamp is FIXED_ONLY, read color clamping is enabled if the selected read color buffer has fixedpoint components. For an integer RGBA color, each component is clamped to the representable range of type. For a floating-point RGBA color, if type is FLOAT or HALF_FLOAT, each component is clamped to [0, 1] if read color clamping is enabled. Then the appropriate conversion formula from table 18.2 is applied to the component. If type is UNSIGNED_INT_10F_11F_11F_REV and format is RGB, each component is clamped to [0, 1] if read color clamping is enabled. The returned data are then 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, and 3rd components of the UNSIGNED_INT_10F_11F_11F_REV format as shown in table 8.8. If type is UNSIGNED_INT_5_9_9_9_REV and format is RGB, each component is clamped to [0, 1] if read color clamping is enabled. The returned data are then packed into a series of uint values. The red, green, and blue components are converted to reds , greens , blues , and exps integers as described in section 8.5.2 when internalformat is RGB9_E5. reds , greens , blues , and exps are then packed as the 1st, 2nd, 3rd, and 4th components of the UNSIGNED_INT_5_9_9_9_REV format as shown in table 8.8. For other types, and for a floating-point or unsigned normalized fixed-point color buffer, each component is clamped to [0, 1] whether or not read color clamping is enabled. For a signed normalized fixed-point color buffer, each component is clamped to [0, 1] if read color clamping is enabled, or if type represents unsigned integer components; otherwise type represents signed integer components,

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

536

and each component is clamped to [−1, 1]. Following clamping, the appropriate conversion formula from table 18.2 is applied to the component1 . For an index, if the type is not FLOAT or HALF_FLOAT, final conversion consists of masking the index with the value given in table 18.3. If the type is FLOAT or HALF_FLOAT, then the integer index is converted to a GL float or half data value.

18.2.9

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 of client memory and the pixels are packed into the client memory relative to the pointer. 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 8.2 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 8.4.4.1. 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, GREEN, or BLUE, only the corresponding single element is written. Likewise if the format is RG, RGB, or BGR, only the corresponding two or three elements are written. Otherwise all the elements of each group are written.

18.3

Copying Pixels

Several commands copy pixel data between regions of the framebuffer (see section 18.3.1), or between regions of textures and renderbuffers (see section 18.3.2). 1

OpenGL 4.2 changes the behavior of ReadPixels to allow readbacks from a signed normalized color buffer to a signed integer type without loss of information.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

type Parameter

537

GL Data Type

UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT HALF_FLOAT FLOAT UNSIGNED_BYTE_3_3_2 UNSIGNED_BYTE_2_3_3_REV UNSIGNED_SHORT_5_6_5 UNSIGNED_SHORT_5_6_5_REV UNSIGNED_SHORT_4_4_4_4 UNSIGNED_SHORT_4_4_4_4_REV UNSIGNED_SHORT_5_5_5_1 UNSIGNED_SHORT_1_5_5_5_REV UNSIGNED_INT_8_8_8_8 UNSIGNED_INT_8_8_8_8_REV UNSIGNED_INT_10_10_10_2 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

ubyte byte ushort short uint int half float ubyte ubyte ushort ushort ushort ushort ushort ushort uint uint uint uint uint uint uint float

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 Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width Equation 2.3, b = bitfield width 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 Special c = f (depth only)

Table 18.2: Reversed component conversions, used when component data are being returned to client memory. Color and depth components are converted from the internal floating-point representation (f ) to a datum of the specified GL data type (c). 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).

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS type Parameter UNSIGNED_BYTE BYTE UNSIGNED_SHORT SHORT UNSIGNED_INT INT UNSIGNED_INT_24_8 FLOAT_32_UNSIGNED_INT_24_8_REV

538 Index Mask 28 − 1 27 − 1 216 − 1 215 − 1 232 − 1 231 − 1 28 − 1 28 − 1

Table 18.3: Index masks used by ReadPixels. Floating-point data are not masked.

For all such commands, if the source and destination are identical or are different views of the same underlying texture image, and if the source and destination regions overlap in that framebuffer, renderbuffer, or texture image, pixel values resulting from the copy operation are undefined.

18.3.1

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 commands void BlitFramebuffer( int srcX0, int srcY0, int srcX1, int srcY1, int dstX0, int dstY0, int dstX1, int dstY1, bitfield mask, enum filter ); void BlitNamedFramebuffer( uint readFramebuffer, uint drawFramebuffer, int srcX0, int srcY0, int srcX1, int srcY1, int dstX0, int dstY0, int dstX1, int dstY1, bitfield mask, enum filter ); For BlitFramebuffer, the source and destination framebuffers are those bound to READ_FRAMEBUFFER and DRAW_FRAMEBUFFER respectively. For BlitNamedFramebuffer, readFramebuffer and drawFramebuffer are the names of the source and destination framebuffer objects respectively. If no framebuffer is bound to READ_FRAMEBUFFER or DRAW_FRAMEBUFFER (for BlitFramebuffer), or if readFramebuffer or drawFramebuffer is zero (for BlitNamedFramebuffer), then the default read or draw framebuffer is used as the corresponding source or destination framebuffer, respectively.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

539

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 17.4.3. 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. If the source and destination buffers are identical, and the source and destination rectangles overlap, the result of the blit operation is undefined as described in the introduction to section 18.3. When values are taken from the read buffer, if FRAMEBUFFER_SRGB is enabled and the value of FRAMEBUFFER_ATTACHMENT_COLOR_ENCODING for the frameOpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

540

buffer 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 8.17. When values are written to the draw buffers, blit operations bypass most of the fragment pipeline. The only fragment operations which affect a blit are the pixel ownership test, the scissor test, and sRGB conversion (see section 17.3.7). Color, depth, and stencil masks (see section 17.4.2) 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. filter 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 implementation-dependent 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. If the read framebuffer is not multisampled and the draw framebuffer is multisampled, the value of the source sample is replicated in each of the destination samples. If both the read and draw framebuffers are multisampled, and their effective values of SAMPLES are identical, the samples are copied without modification (other than possible format conversion) from the read framebuffer to the draw framebuffer. Note that the samples in the draw buffer are not guaranteed to be at the same sample location as the read buffer, so rendering using this newly created buffer can potentially have geometry cracks or incorrect antialiasing. This may occur if the sizes of the framebuffers do not match or if the source and destination rectangles are not defined with the same (X0, Y 0) and (X1, Y 1) bounds.

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

541

Errors An INVALID_OPERATION error is generated by BlitNamedFramebuffer if readFramebuffer or drawFramebuffer is not zero or the name of an existing framebuffer object. 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_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_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 either the read or draw framebuffer is multisampled, and the dimensions of the source and destination rectangles provided to BlitFramebuffer are not identical. An INVALID_OPERATION error is generated if both the read and draw framebuffers are multisampled, and their effective values of SAMPLES are not identical. 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.

18.3.2

Copying Between Images

The command

OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

542

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. CopyImageSubData also allows copying between certain types of compressed and uncompressed internal formats as described in table 18.4. 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 non-proxy texture targets are accepted, with the exception of TEXTURE_BUFFER and the cubemap face selectors described in table 8.19. 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. OpenGL 4.6 (Core Profile) - May 5, 2022

18.3. COPYING PIXELS

543

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 one-dimensional array, 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 8.19. For the purposes of CopyImageSubData, two internal formats are considered compatible if any of the following conditions are met: • the formats are the same If the formats are the same but are a base internal format, the the implementation’s effective internal format (see the end of section 8.5) for each image must be the same. • the formats are considered compatible according to the compatibility rules used for texture views as defined in section 8.18. In particular, if both internal formats are listed in the same entry of table 8.22, they are considered compatible • one format is compressed and the other is uncompressed and table 18.4 lists the two formats in the same row. Undefined pixel values result from overlapping copies, as described in the introduction to section 18.3. 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 non-proxy texture target; is TEXTURE_BUFFER or one of the cubemap face selectors described in table 8.19; 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

OpenGL 4.6 (Core Profile) - May 5, 2022

18.4. PIXEL DRAW AND READ STATE Texel / Block Size 128-bit

Uncompressed internal format RGBA32UI, RGBA32I, RGBA32F

64-bit

RGBA16F, RG32F, RGBA16UI, RG32UI, RGBA16I, RG32I, RGBA16, RGBA16_SNORM

544

Compressed internal format COMPRESSED_RG_RGTC2, COMPRESSED_SIGNED_RG_RGTC2, COMPRESSED_RGBA_BPTC_UNORM, COMPRESSED_SRGB_ALPHA_BPTC_UNORM, COMPRESSED_RGB_BPTC_SIGNED_FLOAT, COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT COMPRESSED_RED_RGTC1, COMPRESSED_SIGNED_RED_RGTC1

Table 18.4: Compatible internal formats for copying between compressed and uncompressed internal formats with CopyImageSubData. Formats in the same row can be copied between each other.

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 subregion 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.

18.4

Pixel Draw and Read State

The state required for pixel operations consists of the parameters that are set with PixelStore. This state has been summarized in tables 8.1 and 18.1. Additional state includes a three-valued integer controlling clamping during final conversion. The initial value of read color clamping is FIXED_ONLY. State set with PixelStore OpenGL 4.6 (Core Profile) - May 5, 2022

18.4. PIXEL DRAW AND READ STATE is GL client state.

OpenGL 4.6 (Core Profile) - May 5, 2022

545

Chapter 19

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 4.3.8(“Shared Variables”) of the OpenGL 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 546

547 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 one or more of the compute shaders attached to the program (see section 4.4.1.4(“Compute Shader Inputs”) of the OpenGL 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 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 4.6 (Core Profile) - May 5, 2022

19.1. COMPUTE SHADER VARIABLES

548

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.

19.1

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.

19.2

Compute Shader Queries

Compute shader queries use query objects to track the number of compute shader invocations. When BeginQuery is called with a target of COMPUTE_SHADER_INVOCATIONS, the compute shader invocations count maintained by the GL is set to zero. When a compute shader invocations query is active, the counter is incremented every time the compute shader is invoked (see chapter 19). Implementations are allowed to skip the execution of certain compute shader invocations, and to execute additional compute shader invocations due to implementation dependent reasons as long as the results of rendering otherwise remain unchanged.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 20

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. Debug output 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 22.2), 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 1 Debug contexts are specified at context creation time, using window system binding APIs such as those specified in the GLX_ARB_create_context and WGL_ARB_create_context extensions for GLX and WGL, respectively.

549

20.1. 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

550 Messages Generated by The GL The GLSL shader compiler or compilers for other extension-provided languages The window system, such as WGL or GLX External debuggers or third-party middleware libraries The application Sources that do not fit to any of the ones listed above

Table 20.1: Sources of debug output messages. Each message must originate from a source listed in this table.

output. Full debug output support is guaranteed only in a debug context.

20.1

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 20.1. The message type is one of the symbolic constants listed in table 20.2. 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 20.3. Because messages can be disabled by their severity, the global volume of debug output can be limited. 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 4.6 (Core Profile) - May 5, 2022

20.1. DEBUG MESSAGES

551

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 20.2: 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 20.3: Severity levels of messages. Each debug output message is associated with one of these severity levels.

OpenGL 4.6 (Core Profile) - May 5, 2022

20.2. DEBUG MESSAGE CALLBACK

552

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, is guaranteed to be less than 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.

20.2

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 4.6 (Core Profile) - May 5, 2022

20.3. DEBUG MESSAGE LOG

553

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 20.8 describes this in further detail. If DEBUG_OUTPUT is disabled, then the GL will not call the callback function.

20.3

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 20.2. If DEBUG_OUTPUT is disabled, then no messages are added to the message log.

20.4

Controlling Debug Messages

Applications can control the volume of debug output in the active debug group (see section 20.6) by disabling specific groups of messages with the command

OpenGL 4.6 (Core Profile) - May 5, 2022

20.4. CONTROLLING DEBUG MESSAGES

554

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 20.1, 20.2, and 20.3. An INVALID_VALUE error is generated if count is negative, An INVALID_OPERATION error is generated if count is greater than zero and either source or type is DONT_CARE, or severity is not DONT_CARE.

OpenGL 4.6 (Core Profile) - May 5, 2022

20.5. EXTERNALLY GENERATED MESSAGES

20.5

555

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 20.2, 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 20.3. 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.

20.6

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 void PushDebugGroup( enum source, uint id, sizei length, const char *message ); OpenGL 4.6 (Core Profile) - May 5, 2022

20.6. DEBUG GROUPS

556

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 only the default debug group is on the stack.

OpenGL 4.6 (Core Profile) - May 5, 2022

20.7. DEBUG LABELS

557 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 20.4: Object namespace identifiers and the corresponding object types.

20.7

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 chapter 20) 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 20.4, 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 types listed in table 20.4. An INVALID_VALUE error is generated if name is not the name of a valid object of the type specified by identifier.

OpenGL 4.6 (Core Profile) - May 5, 2022

20.8. ASYNCHRONOUS AND SYNCHRONOUS DEBUG OUTPUT

558

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.

20.8

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 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 OpenGL 4.6 (Core Profile) - May 5, 2022

20.9. DEBUG OUTPUT QUERIES

559

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. The 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.

20.9

Debug Output Queries

Pointers set with debug output commands are queried with the generic GetPointerv command (see section 22.2). 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 20.3. 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. 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 OpenGL 4.6 (Core Profile) - May 5, 2022

20.9. DEBUG OUTPUT QUERIES

560

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 20.7). 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 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

20.9. DEBUG OUTPUT QUERIES

561

Errors An INVALID_ENUM error is generated is identifier is not one of the object types listed in table 20.4. 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 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 4.6 (Core Profile) - May 5, 2022

Chapter 21

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 21.5).

21.1 This section is only defined in the compatibility profile.

21.2 This section is only defined in the compatibility profile.

21.3 This section is only defined in the compatibility profile.

21.4 This section is only defined in the compatibility profile.

21.5

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 ); 562

21.6.

563

Target LINE_SMOOTH_HINT POLYGON_SMOOTH_HINT TEXTURE_COMPRESSION_HINT FRAGMENT_SHADER_DERIVATIVE_HINT

Hint description Line sampling quality Polygon sampling quality Quality and performance of texture image compression Derivative accuracy for fragment processing built-in functions dFdx, dFdy and fwidth

Table 21.1: Hint targets and descriptions.

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 21.1. 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. For the texture compression hint, a hint of FASTEST indicates that texture images should be compressed as quickly as possible, while NICEST indicates that the texture images should be compressed with as little image degradation as possible. FASTEST should be used for one-time texture compression, and NICEST should be used if the compression results are to be retrieved by GetCompressedTexImage (section 8.11) for reuse. 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 21.1. An INVALID_ENUM error is generated if hint is not FASTEST, NICEST, or DONT_CARE.

21.6 This section is only defined in the compatibility profile.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 22

Context State Queries The state required to describe the GL machine is enumerated in chapter 23, 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.

22.1

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 23. There are five 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 ); void GetDoublev( enum pname, double *data );

564

22.1. SIMPLE QUERIES

565

The commands obtain boolean, integer, 64-bit integer, floating-point, or doubleprecision 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 GetFloati v( enum target, uint index, float *data ); void GetDoublei v( enum target, uint index, double *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. Finally, boolean IsEnabled( enum cap ); can be used to determine if cap is currently enabled (as with Enable) or disabled. Errors An INVALID_ENUM error is generated if cap is not enable state queriable with IsEnabled.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.2. POINTER, STRING, AND RELATED CONTEXT QUERIES

566

boolean IsEnabledi( enum target, uint index ); 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.

22.2

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 20.9. Errors An INVALID_ENUM error is generated if pname is not one of the names described above. String queries return pointers to UTF-8 encoded, null-terminated static strings describing properties of the current GL context 1 . The command ubyte *GetString( enum name ); accepts name values of RENDERER, VENDOR, VERSION, and SHADING_LANGUAGE_VERSION. The format of the RENDERER and VENDOR strings is implementation-dependent. The VERSION and SHADING_LANGUAGE_VERSION strings are laid out as follows: <version number><space><vendor-specific information> 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.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.2. POINTER, STRING, AND RELATED CONTEXT QUERIES Value CONTEXT_CORE_PROFILE_BIT CONTEXT_COMPATIBILITY_PROFILE_BIT

567

OpenGL Profile Core Compatibility

Table 22.1: Context profile bits returned by the CONTEXT_PROFILE_MASK query. 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 Shading Language Specification release number. For example, this query might return the string "4.20" while the corresponding VERSION query returns "4.2". 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) that can be supported by the current GL context. Thus, if the client and server support different versions a compatible version is returned. Errors An INVALID_ENUM error is generated if name is not RENDERER, VENDOR, 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.

The profile implemented by the context may be queried by calling GetIntegerv with value CONTEXT_PROFILE_MASK, which returns a mask containing one of the bits in table 22.1, corresponding to the API profile implemented by the context (see appendix E.1). Flags defining additional properties of the context may be queried by calling GetIntegerv with pname CONTEXT_FLAGS. If CONTEXT_FLAG_FORWARD_COMPATIBLE_BIT is set in CONTEXT_FLAGS, then the context is a forward-compatible context as defined in appendix E, and the deprecated features described in that appendix are not supported; otherwise the context is a full context, and all features described in the specification are supported. 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 20. OpenGL 4.6 (Core Profile) - May 5, 2022

22.2. POINTER, STRING, AND RELATED CONTEXT QUERIES

568

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 10.3.7. If CONTEXT_FLAG_NO_ERROR_BIT is set in CONTEXT_FLAGS, then no error behavior is enabled, as described in section 2.3.1.1. The behavior of the context when it is made no longer current (released) by the platform binding may be queried by calling GetIntegerv with pname CONTEXT_RELEASE_BEHAVIOR. If the behavior is CONTEXT_RELEASE_BEHAVIOR_FLUSH, any pending commands on the context will be flushed. If the behavior is NONE, pending commands are not flushed. The default value is CONTEXT_RELEASE_BEHAVIOR_FLUSH, and may in some cases be changed using platform-specific context creation extensions. 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. 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. 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. If name is SHADING_LANGUAGE_VERSION, a version string for one of the supported versions of the OpenGL Shading Language and OpenGL ES Shading Language is returned. index may range from zero to the value of NUM_SHADING_LANGUAGE_VERSIONS minus one. The format of the returned string is identical to the text that may follow #version in shader program source and is formatted as the version number followed, for versions in which language profiles are defined, by a space and a profile name. For example, a returned string containing "420 core" indicates support for OpenGL Shading Language 4.20, core profile. An empty string indicates support for OpenGL Shading Language 1.10, which did not include the #version compiler directive. The profile string will always be present in the returned string when it is accepted by that version of the OpenGL Shading Language, even though there is a default profile string in versions 1.50 and greater. Version strings 100, 300 es, and 310 es correspond to OpenGL ES Shading Language versions 1.00, 3.00 and 3.10, respectively. An index of zero will always return the string for the version of the most recent shading language supported by the GL and the profile of the shading language corresponding to the profile of the API (e.g. the first entry returned in an OpenGL OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

569

4.30 core profile context will be "430 core" and the first entry returned in an OpenGL 4.30 compatibility profile context will be "430 compatibility"). There is no defined ordering of the returned strings for other values of index. If name is SPIR_V_EXTENSIONS, the SPIR-V extension name corresponding to the indexth supported SPIR-V extension (specified as the Name String in the corresponding SPIR-V extension document) will be returned. index may range from zero to the value of NUM_SPIR_V_EXTENSIONS minus one. The value of NUM_SPIR_V_EXTENSIONS can be queried using the GetInteger command. There is no defined relationship between any particular extension name and the index values; a SPIR-V 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, SHADING_LANGUAGE_VERSION or SPIR_V_EXTENSIONS. An INVALID_VALUE error is generated if index is outside the valid range for the indexed state name.

22.3

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 ); void GetInternalformati64v( enum target, enum internalformat, enum pname, sizei count, int64 *params ); internalformat can be any value. The INTERNALFORMAT_SUPPORTED pname can be used to determine if the internal format is supported, and the other pnames are defined in terms of whether or not the format is supported. target indicates the usage of the internalformat, and must be one the targets listed in table 22.2. 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. pname indicates the information to query. The following subsections list the valid values for pname and define their meaning and the values that may be returned. In the following descriptions, the term resource is used to generically refer to an object of the appropriate type that has been created with internalformat OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES Target TEXTURE_1D TEXTURE_1D_ARRAY TEXTURE_2D TEXTURE_2D_ARRAY TEXTURE_2D_MULTISAMPLE TEXTURE_2D_MULTISAMPLE_ARRAY TEXTURE_3D TEXTURE_BUFFER TEXTURE_CUBE_MAP TEXTURE_CUBE_MAP_ARRAY TEXTURE_RECTANGLE RENDERBUFFER

570 Usage 1D texture 1D array texture 2D texture 2D array texture 2D multisample texture 2D multisample array texture 3D texture buffer texture cube map texture cube map array texture rectangle texture renderbuffer

Table 22.2: Possible targets that internalformat can be used with and the corresponding usage meaning.

and target. If the particular target and internalformat combination does not make sense, or if a particular type of target is not supported by the implementation the unsupported answer should be given. This is not an error. All properties may be queried via either GetInternalformat* command. Data conversions are done as defined in section 2.2.2.

22.3.1

Supported Operation Queries

Queries that return information about supported types of operations will return one of the following values in params: • NONE: the requested resource or operation is not supported at all by the implementation. • CAVEAT_SUPPORT: the requested operation is supported by the implementation, but there may be some implementation-specific caveats that make support less than optimal. For example using the feature may result in reduced performance (relative to other formats or features), such as software rendering or other mechanisms of emulating the desired feature. If a query reports that there is a caveat and the debug output functionality is enabled (see section 20), the GL will generate a debug output message

OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

571

describing the caveat. The message has the source DEBUG_SOURCE_API, the type DEBUG_TYPE_PERFORMANCE, and an implementation-dependent ID. • FULL_SUPPORT: the requested operation is fully supported by the implementation. Possible pnames for supported types of operations, and their meanings, include: • CLEAR_BUFFER: Support for using the resource with ClearBuffer*Data commands is returned in params. • CLEAR_TEXTURE: Support for using the resource with ClearTex*Image commands is returned in params. • COMPUTE_TEXTURE: Support for using the resource as a source for texture sampling in a compute shader is written to params. • FILTER: Support for filter types other than NEAREST or NEAREST_MIPMAP_NEAREST for the resource is written to params. This indicates if sampling from such resources supports setting the MIN/MAG filters to LINEAR values. • FRAGMENT_TEXTURE: Support for using the resource as a source for texture sampling in a fragment shader is written to params. • FRAMEBUFFER_BLEND: Support for rendering to the resource via framebuffer attachment when blending is enabled is returned in params. • FRAMEBUFFER_RENDERABLE: Support for rendering to the resource via framebuffer attachment is returned in params. • FRAMEBUFFER_RENDERABLE_LAYERED: Support for layered rendering to the resource via framebuffer attachment is returned in params. • GEOMETRY_TEXTURE: Support for using the resource as a source for texture sampling in a geometry shader is written to params. • MANUAL_GENERATE_MIPMAP: Support for manually generating mipmaps for the resource is returned in params. • READ_PIXELS: Support for reading pixels from the resource when it is attached to a framebuffer is returned in params.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

572

• SHADER_IMAGE_ATOMIC: Support for using the resource with atomic memory operations from shaders is written to params. • SHADER_IMAGE_LOAD: Support for using the resource with image load operations in shaders is written to params. In this case the internalformat is the value of the format parameter that would be passed to BindImageTexture. • SHADER_IMAGE_STORE: Support for using the resource with image store operations in shaders is written to params. In this case the internalformat is the value of the format parameter that is passed to BindImageTexture. • SIMULTANEOUS_TEXTURE_AND_DEPTH_TEST: Support for using the resource both as a source for texture sampling while it is bound as a buffer for depth test is written to params. For example, a depth (or stencil) texture could be bound simultaneously for texturing while it is bound as a depth (and/or stencil) buffer without causing a feedback loop, provided that depth writes are disabled. • SIMULTANEOUS_TEXTURE_AND_DEPTH_WRITE: Support for using the resource both as a source for texture sampling while performing depth writes to the resources is written to params. For example, a depth-stencil texture could be bound simultaneously for stencil texturing while it is bound as a depth buffer. Feedback loops cannot occur because sampling a stencil texture only returns the stencil portion, and thus writes to the depth buffer do not modify the stencil portions. • SIMULTANEOUS_TEXTURE_AND_STENCIL_TEST: Support for using the resource both as a source for texture sampling while it is bound as a buffer for stencil test is written to params. For example, a depth (or stencil) texture could be bound simultaneously for texturing while it is bound as a depth (and/or stencil) buffer without causing a feedback loop, provided that stencil writes are disabled. • SIMULTANEOUS_TEXTURE_AND_STENCIL_WRITE: Support for using the resource both as a source for texture sampling while performing stencil writes to the resources is written to params. For example, a depth-stencil texture could be bound simultaneously for depth-texturing while it is bound as a stencil buffer. Feedback loops cannot occur because sampling a depth texture only returns the depth portion, and thus writes to the stencil buffer could not modify the depth portions. • SRGB_READ: Support for converting from sRGB colorspace on read operations (see section 8.24) from the resource is returned in params. OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

573

• SRGB_WRITE: Support for converting to sRGB colorspace on write operations to the resource is returned in params. This indicates that writing to framebuffers with this internal format will encode to sRGB color spaces when FRAMEBUFFER_SRGB is enabled (see section 17.3.7). • TESS_CONTROL_TEXTURE: Support for using the resource as a source for texture sampling in a tessellation control shader is written to params. • TESS_EVALUATION_TEXTURE: Support for using the resource as a source for texture sampling in a tessellation evaluation shader is written to params. • TEXTURE_GATHER: Support for using the resource with texture gather operations is written to params. • TEXTURE_GATHER_SHADOW: Support for using resource with texture gather operations with shadow samplers is written to params. • TEXTURE_SHADOW: Support for using the resource with shadow samplers is written to params. • TEXTURE_VIEW: Support for using the resource with the TextureView command is returned in params. • VERTEX_TEXTURE: Support for using the resource as a source for texture sampling in a vertex shader is written to params.

22.3.2

Other Internal Format Queries

Other supported values for pname, their meanings, and their possible return values include: • COLOR_COMPONENTS: If the internal format contains any color components (R, G, B, or A), TRUE is returned in params. If the internal format is unsupported or contains no color components, FALSE is returned. • COLOR_ENCODING: The color encoding for the resource is returned in params. Possible values for color buffers are LINEAR or SRGB, for linear or sRGB-encoded color components, respectively. For non-color formats (such as depth or stencil), or for unsupported resources, the value NONE is returned. • COLOR_RENDERABLE: If internalformat is color-renderable (as defined in section 9.4), TRUE is returned in params. If the internal format is unsupported, or the internal format is not color-renderable, FALSE is returned. OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

574

• DEPTH_COMPONENTS: If the internal format contains a depth component (D), TRUE is returned in params. If the internal format is unsupported or contains no depth component, FALSE is returned. • DEPTH_RENDERABLE: If internalformat is depth-renderable (as defined in section 9.4), TRUE is returned in params. If the internal format is unsupported, or if the internal format is not depth-renderable, FALSE is returned. • GET_TEXTURE_IMAGE_FORMAT: The implementation-preferred format to pass to GetTexImage when querying texture image data from this resource. Possible values include any value that is legal to pass for the format parameter to GetTexImage, or NONE if the resource does not support this operation, or if GetTexImage is not supported. • GET_TEXTURE_IMAGE_TYPE: The implementation-preferred type to pass to GetTexImage when querying texture image data from this resource. Possible values include any value that is legal to pass for the type parameter to GetTexImage, or NONE if the resource does not support this operation, or if GetTexImage is not supported. • IMAGE_COMPATIBILITY_CLASS: The compatibility class of the resource when used as an image texture is returned in params. This corresponds to the value from the Class column in table 8.27. The possible values returned are IMAGE_CLASS_4_X_32, IMAGE_CLASS_2_X_32, IMAGE_CLASS_1_X_32, IMAGE_CLASS_4_X_16, IMAGE_CLASS_2_X_16, IMAGE_CLASS_1_X_16, IMAGE_CLASS_4_X_8, IMAGE_CLASS_2_X_8, IMAGE_CLASS_1_X_8, IMAGE_CLASS_11_11_10, and IMAGE_CLASS_10_10_10_2, which correspond to the 4x32, 2x32, 1x32, 4x16, 2x16, 1x16, 4x8, 2x8, 1x8, the class (a) 11/11/10 packed floating-point format, and the class (b) 10/10/10/2 packed formats, respectively. If the resource is not supported for image textures, or if image textures are not supported, NONE is returned. • IMAGE_FORMAT_COMPATIBILITY_TYPE: The matching criteria use for the resource when used as an image textures is returned in params. This is equivalent to calling GetTexParameter with pname set to IMAGE_FORMAT_COMPATIBILITY_TYPE. Possible values are IMAGE_FORMAT_COMPATIBILITY_BY_SIZE or IMAGE_FORMAT_COMPATIBILITY_BY_CLASS. If the resource is not supported for image textures, or if image textures are not supported, NONE is returned.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

575

• IMAGE_PIXEL_FORMAT: The pixel format of the resource when used as an image texture is returned in params. This is the value from the Pixel format column in table 8.27. If the resource is not supported for image textures, or if image textures are not supported, NONE is returned. • IMAGE_PIXEL_TYPE: The pixel type of the resource when used as an image texture is returned in params. This is the value from the Pixel type column in table 8.27. If the resource is not supported for image textures, or if image textures are not supported, NONE is returned. • IMAGE_TEXEL_SIZE: The size of a texel of the resource when used as an image texture is returned in params. This is the value from the Size column in table 8.27. If the resource is not supported for image textures, or if image textures are not supported, zero is returned. • INTERNALFORMAT_PREFERRED: The implementation-preferred internal format for representing resources of the specified internalformat is returned in params. The preferred internal format should have no less precision than the requested one. If the specified internalformat is already a preferred format, or if there is no better format that is compatible, the queried internalformat value is written to params. If the internalformat is not supported, NONE is returned. • INTERNALFORMAT_RED_SIZE, INTERNALFORMAT_GREEN_SIZE, INTERNALFORMAT_BLUE_SIZE, INTERNALFORMAT_ALPHA_SIZE, INTERNALFORMAT_DEPTH_SIZE, INTERNALFORMAT_STENCIL_SIZE, or INTERNALFORMAT_SHARED_SIZE: For uncompressed internal formats, queries of these values return the actual resolutions that would be used for storing image components for the resource. For compressed internal formats, the resolutions returned specify the component resolution of an uncompressed internal format that produces an image of roughly the same quality as the compressed algorithm. For textures this query will return the same information as querying GetTexLevelParameter* for the corresponding TEXTURE_*_SIZE (except in cases where GetTexLevelParameter* does not support such a query). If the internal format is unsupported, or if a particular component is not present in the format, 0 is written to params. • INTERNALFORMAT_RED_TYPE, INTERNALFORMAT_GREEN_TYPE, INTERNALFORMAT_BLUE_TYPE, INTERNALFORMAT_ALPHA_TYPE, INTERNALFORMAT_DEPTH_TYPE, or INTERNALFORMAT_STENCIL_TYPE: OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

576

For uncompressed internal formats, queries for these values return the data type used to store the component. For compressed internal formats the types returned specify how components are interpreted after decompression. For textures this query returns the same information as querying GetTexLevelParameter* for the corresponding TEXTURE_*_TYPE (except in cases where GetTexLevelParameter* does not support such a query). Possible values returned include NONE, SIGNED_NORMALIZED, UNSIGNED_NORMALIZED, FLOAT, INT, and UNSIGNED_INT, representing missing, signed normalized fixed-point, unsigned normalized fixed-point, floating-point, signed unnormalized integer, and unsigned unnormalized integer components respectively. NONE is returned for all component types if the format is unsupported. • INTERNALFORMAT_SUPPORTED: If internalformat is an internal format that is supported by the implementation in at least some subset of possible operations, TRUE is written to params. If internalformat if not a valid token for any internal format usage, FALSE is returned. internalformats that must be supported include: – sized internal formats from tables 8.12- 8.13 and 8.16, – any specific compressed internal format from table 8.14, – any image unit format from table 8.26, – any generic compressed internal format from table 8.14, if the implementation accepts it for any texture specification commands, and – any unsized or base internal format, if the implementation accepts it for texture or image specification. In other words, any internalformat accepted by any of the commands: ClearBufferData, ClearBufferSubData, CompressedTexImage1D, CompressedTexImage2D, CompressedTexImage3D, CopyTexImage1D, CopyTexImage2D, RenderbufferStorage, RenderbufferStorageMultisample, TexBuffer, TexImage1D, TexImage2D, TexImage3D, TexImage2DMultisample, TexImage3DMultisample, TexStorage1D, TexStorage2D, TexStorage3D, TexStorage2DMultisample, TexStorage3DMultisample, and TextureView, and any valid format accepted by BindImageTexture, must be supported. • MAX_COMBINED_DIMENSIONS: The maximum combined dimensions for the resource is returned in params. The combined dimensions is the product of the individual dimensions of the resource. For multisampled surfaces OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

577

the number of samples is considered an additional dimension. Note that the value returned can be ≥ 232 and should be queried with GetInternalformati64v. This value should be considered a recommendation for applications. There may be system-dependant reasons why allocations larger than this size may fail, even if there might appear to be sufficient memory available when queried via some other means. This also does not provide a guarantee that allocations smaller than this will succeed because this value is not affected by existing resource allocations. For one-dimensional targets this is the maximum single dimension. For one-dimensional array targets this is the maximum combined width and layers. For two-dimensional targets this is the maximum combined width and height. For two-dimensional multisample targets this is the combined width, height and samples. For two-dimensional array targets this is the max combined width, height and layers. For two-dimensional multisample array targets, this is the max combined width, height, layers and samples. For three-dimensional targets this is the maximum combined width, height and depth. For cube map targets this is the maximum combined width, height and faces. For cube map array targets this is the maximum width, height and layer-faces. If the resource is unsupported, zero is returned. • MAX_DEPTH: The maximum supported depth for the resource is returned in params. For resources with three or more dimensions, the third dimension is considered the depth. If the resource does not have at least three dimensions, or if the resource is unsupported, zero is returned. • MAX_HEIGHT: The maximum supported height for the resource is returned in params. For resources with two or more dimensions, the second dimension is considered the height. If the resource does not have at least two dimensions, or if the resource is unsupported, zero is returned. • MAX_LAYERS: The maximum supported number of layers for the resource is returned in params. For 1D array targets, the value returned is the same as the MAX_HEIGHT. For 2D and cube array targets, the value returned is the same as the MAX_DEPTH. If the resource does not support layers, or if the resource is unsupported, zero is returned. • MAX_WIDTH: The maximum supported width for the resource is returned in params. For resources with only one dimension, that dimension is considered the width. If the resource is unsupported, zero is returned.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

578

• MIPMAP: If the resource supports mipmaps, TRUE is returned in params. If the resource is not supported, or if mipmaps are not supported for this type of resource, FALSE is returned. • NUM_SAMPLE_COUNTS: The number of sample counts that would be returned by querying SAMPLES is returned in params. – If internalformat is not color-renderable, depth-renderable, or stencilrenderable (as defined in section 9.4), or if target does not support multiple samples (is not TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY, or RENDERBUFFER), zero is returned. • READ_PIXELS_FORMAT: The format to pass to ReadPixels to obtain the best performance and image quality when reading from framebuffers with internalformat is returned in params. Possible values include any value that is legal to pass for the format parameter to ReadPixels, or NONE if internalformat is not supported or can never be a valid source for ReadPixels. • READ_PIXELS_TYPE: The type to pass to ReadPixels to obtain the best performance and image quality when reading from framebuffers with internalformat is returned in params. Possible values include any value that is legal to pass for the type parameter to ReadPixels, or NONE if the internal format is not supported or can never be a source for ReadPixels. • 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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

22.3. INTERNAL FORMAT QUERIES

579

* The value of MAX_SAMPLES. – If internalformat is not color-renderable, depth-renderable, or stencilrenderable (as defined in section 9.4), or if target does not support multiple samples (i.e. other than TEXTURE_2D_MULTISAMPLE, TEXTURE_2D_MULTISAMPLE_ARRAY, or RENDERBUFFER), params is not modified. • STENCIL_COMPONENTS: If the internal format contains a stencil component (S), TRUE is returned in params. If the internal format is unsupported or contains no stencil component, FALSE is returned. • STENCIL_RENDERABLE: If internalformat is stencil-renderable (as defined in section 9.4), TRUE is returned in params. If the internal format is unsupported, or if the internal format is not stencil-renderable, FALSE is returned. • TEXTURE_COMPRESSED: If internalformat is a compressed format that is supported for this type of resource, TRUE is returned in params. If the internal format is not compressed, or the type of resource is not supported, FALSE is returned. • TEXTURE_COMPRESSED_BLOCK_HEIGHT: If the resource contains a compressed format, the height of a compressed block (in texels) is returned in params. If the internal format is not compressed, or the resource is not supported, 0 is returned. • TEXTURE_COMPRESSED_BLOCK_SIZE: If the resource contains a compressed format, the number of bytes per block is returned in params. If the internal format is not compressed, or the resource is not supported, 0 is returned. Together with the block width and height queries this allows the bitrate to be computed, and may be useful in conjunction with ARB compressed texture pixel storage). • TEXTURE_COMPRESSED_BLOCK_WIDTH: If the resource contains a compressed format, the width of a compressed block (in texels) is returned in params. If the internal format is not compressed, or the resource is not supported, 0 is returned. • TEXTURE_IMAGE_FORMAT: The implementation-preferred format to pass to TexImage*D or TexSubImage*D when specifying texture image data for this resource is returned in params. Possible values include any value that is legal to pass for the format parameter to the Tex*Image*D commands, or NONE if the resource is not supported for this operation. OpenGL 4.6 (Core Profile) - May 5, 2022

22.4. TRANSFORM FEEDBACK STATE QUERIES

580

• TEXTURE_IMAGE_TYPE: The implementation-preferred type to pass to TexImage*D or TexSubImage*D when specifying texture image data for this resource is returned in params. Possible values include any value that is legal to pass for the type parameter to the Tex*Image*D commands, or NONE if the resource is not supported for this operation. • VIEW_COMPATIBILITY_CLASS: The compatibility class of the resource when used as a texture view is returned in params. The compatibility class is one of the values from the Class column of table 8.22. If the resource has no other formats that are compatible, the resource does not support views, or if texture views are not supported, NONE is returned.

Errors An INVALID_ENUM error is generated if target is not one of the targets in table 22.2, or if pname is not one of the parameters described above. An INVALID_VALUE error is generated if count is negative.

22.4

Transform Feedback State Queries

State of the currently bound transform feedback object may be queried by calling GetIntegerv, GetIntegeri v, GetInteger64i v, GetBooleanv, or other query functions with the query target set to one of the tokens listed in table 23.48. Alternatively, the state of a transform feedback object may be queried with the commands void GetTransformFeedbackiv( uint xfb, enum pname, int *param ); void GetTransformFeedbacki v( uint xfb, enum pname, uint index, int *param ); void GetTransformFeedbacki64 v( uint xfb, enum pname, uint index, int64 *param ); xfb must be zero, indicating the default transform feedback object, or the name of an existing transform feedback object. pname must be one of the tokens listed in table 23.48, depending on the command name as shown in the errors section below. For indexed state, index is the index of the transform feedback stream. param is the address of a variable to receive the result of the query.

OpenGL 4.6 (Core Profile) - May 5, 2022

22.5. INDEXED BINDING STATE QUERIES

581

Errors An INVALID_OPERATION error is generated by GetTransformFeedback* if xfb is not zero or the name of an existing transform feedback object. An INVALID_ENUM error is generated by GetTransformFeedbackiv if pname is not TRANSFORM_FEEDBACK_PAUSED or TRANSFORM_FEEDBACK_ACTIVE. An INVALID_ENUM error is generated by GetTransformFeedbacki v if pname is not TRANSFORM_FEEDBACK_BUFFER_BINDING. An INVALID_ENUM error is generated by GetTransformFeedbacki64 v if pname is not TRANSFORM_FEEDBACK_BUFFER_START or TRANSFORM_FEEDBACK_BUFFER_SIZE. An INVALID_VALUE error is generated by GetTransformFeedbacki v and GetTransformFeedbacki64 v if index is greater than or equal to the number of binding points for transform feedback, as described in section 6.7.1.

22.5

Indexed Binding State Queries

The name of the texture object bound to the active texture unit may be queried by calling GetIntegerv with pname TEXTURE_BINDING_1D, TEXTURE_BINDING_1D_ARRAY, TEXTURE_BINDING_2D, TEXTURE_BINDING_2D_ARRAY, TEXTURE_BINDING_2D_MULTISAMPLE, TEXTURE_BINDING_2D_MULTISAMPLE_ARRAY, TEXTURE_BINDING_3D, TEXTURE_BINDING_BUFFER, TEXTURE_BINDING_CUBE_MAP, TEXTURE_BINDING_CUBE_MAP_ARRAY, or TEXTURE_BINDING_RECTANGLE. Likewise, the current sampler bound to the active texture unit may be queried by calling GetIntegerv with pname SAMPLER_BINDING.

To query the bound texture or sampler object bound to a specific texture unit without changing the active texture selector, call GetIntegeri v with one of the valid pnames listed above, and with index set to the zero-based texture unit index to be queried.

OpenGL 4.6 (Core Profile) - May 5, 2022

Chapter 23

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 the simple queries in section 22.1 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 any of the simple queries. 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 floating-point 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 23.1 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.

582

583 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 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 23.1: State Variable Types

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.2: Current Values and Associated Data

OpenGL 4.6 (Core Profile) - May 5, 2022

PATCH VERTICES

Type Z+

Get Command GetIntegerv

Initial Value 3

PATCH DEFAULT OUTER LEVEL

4×R

GetFloatv

(1.0, 1.0, 1.0, 1.0)

PATCH DEFAULT INNER LEVEL

2×R

GetFloatv

(1.0, 1.0)

Get value

Description No. of vertices in input patch Default outer tess. level w/o control shader Default inner tess. level w/o control shader

Sec. 10.1 11.2.2 11.2.2

584

Table 23.3: Vertex Array Object State (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

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 LONG

16 ∗ ×B

GetVertexAttribiv

FALSE

VERTEX ATTRIB ARRAY DIVISOR

16 ∗ ×Z +

GetVertexAttribiv

VERTEX ATTRIB ARRAY POINTER

16 ∗ ×Y

Get value VERTEX ATTRIB ARRAY ENABLED VERTEX ATTRIB ARRAY SIZE VERTEX ATTRIB ARRAY STRIDE

–

GetVertexAttribPointerv GetObjectLabel

NULL empty

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 has unconverted doubles Vertex attrib array instance divisor Vertex attrib array pointer Debug label

Sec. 10.3 10.3 10.3 10.3 10.3 10.3 10.3 10.4 10.3 20.9

585

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.4: Vertex Array Object State (cont.) † The ith attribute defaults to a value of i.

Get value ELEMENT ARRAY BUFFER BINDING

Type Z+

Get Command

Initial Value

GetIntegerv

VERTEX ATTRIB ARRAY BUFFER BINDING

16 ∗ ×Z +

GetVertexAttribiv

VERTEX ATTRIB BINDING

16 × Z16∗

GetVertexAttribiv

i†

VERTEX ATTRIB RELATIVE OFFSET

16 × Z +

GetVertexAttribiv

VERTEX BINDING OFFSET

16 × Z

GetInteger64i v

VERTEX BINDING STRIDE

16 × Z

GetIntegeri v

VERTEX BINDING DIVISOR

16 × Z

GetIntegeri v

VERTEX BINDING BUFFER

16 × Z

GetIntegeri v

Description 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 for vertex binding i Name of buffer bound to vertex binding i

Sec. 10.3.10 6 10.3 10.3

10.3

10.3 10.3 10.3

586

Table 23.5: Vertex Array Data (not in Vertex Array objects)

OpenGL 4.6 (Core Profile) - May 5, 2022

ARRAY BUFFER BINDING

Type Z+

Get Command GetIntegerv

Initial Value 0

DRAW INDIRECT BUFFER BINDING

Z+

GetIntegerv

VERTEX ARRAY BINDING

Z+

GetIntegerv

PARAMETER BUFFER BINDING

Z+

GetIntegerv

PRIMITIVE RESTART

IsEnabled

FALSE

PRIMITIVE RESTART FIXED INDEX

IsEnabled

FALSE

PRIMITIVE RESTART INDEX

Z+

GetIntegerv

Get value

Description Current buffer binding Indirect command buffer binding Current vertex array object binding Indirect parameter buffer binding Primitive restart enable Primitive restart fixed index enable Primitive restart index

Sec. 6 10.3.11 10.3.1 10.4 10.3 10.3.6 10.3.6

587

Table 23.6: Buffer Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value – BUFFER SIZE BUFFER USAGE BUFFER ACCESS BUFFER ACCESS FLAGS BUFFER IMMUTABLE STORAGE BUFFER STORAGE FLAGS BUFFER MAPPED BUFFER MAP POINTER BUFFER MAP OFFSET BUFFER MAP LENGTH –

Type n × BM U n × Z+ n×E n×E n × Z+

Get Command GetBufferSubData GetBufferParameteri64v GetBufferParameteriv GetBufferParameteriv GetBufferParameteriv

Initial Value 0 STATIC_DRAW READ_WRITE 0

GetBufferParameteriv

FALSE

Z+ n×B n×Y n × Z+ n × Z+ S

GetBufferParameteriv GetBufferParameteriv GetBufferPointerv GetBufferParameteri64v GetBufferParameteri64v GetObjectLabel

0 FALSE NULL 0 0 empty

Description Buffer data Buffer data size Buffer usage pattern Buffer access flag Extended buffer access flag TRUE if buffer’s data store is immutable, FALSE otherwise Buffer object storage flags Buffer map flag Mapped buffer pointer Start of mapped buffer range Size of mapped buffer range Debug label

Sec. 6 6 6 6.3 6.3 6 6 6.3 6.3 6.3 6.3 20.9

588

Table 23.7: Transformation state

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Get Command GetFloati v GetDoublei v

Initial Value See sec. 13.8.1 0,1

DEPTH RANGE

Type 16 ∗ ×4 × R 16 ∗ ×2×R[0,1]

CLIP DISTANCEi

8∗ × B

IsEnabled

FALSE

DEPTH CLAMP

Z2 Z2 B

GetIntegerv GetIntegerv IsEnabled

LOWER_LEFT NEGATIVE_ONE_TO_ONE FALSE

TRANSFORM FEEDBACK BINDING

Z+

GetIntegerv

Z+

GetIntegerv

VIEWPORT

CLIP ORIGIN CLIP DEPTH MODE

TRANSFORM FEEDBACK BUFFER BINDING

Description Viewport origin & extent Depth range near & far ith user clipping plane enabled Clip origin Clip depth mode Depth clamping enabled Object bound for transform feedback operations Buffer object bound to generic bind point for transform feedback

Sec. 13.8.1 13.8.1 13.7 13.7 13.7 13.7 13.3 6.7

589

Table 23.8: Coloring

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

CLAMP READ COLOR

GetIntegerv

PROVOKING VERTEX

GetIntegerv

Initial Value Description FIXED_Read color clamping ONLY LAST_Provoking vertex conVERTEX_vention CONVENTION

Sec. 18.2.8 13.6

590

Table 23.9: Rasterization

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value RASTERIZER DISCARD POINT SIZE POINT FADE THRESHOLD SIZE POINT SPRITE COORD ORIGIN LINE WIDTH LINE SMOOTH

Type

Get Command

Initial Value

IsEnabled

FALSE

R+ R+ E R+ B

GetFloatv GetFloatv GetIntegerv GetFloatv IsEnabled

1.0 1.0 UPPER_LEFT 1.0 FALSE

Description Discard primitives before rasterization Point size Threshold for alpha attenuation Origin orientation for point sprites Line width Line antialiasing on

Sec. 14.1 14.4 14.4 14.4 14.5 14.5

591

Table 23.10: Rasterization (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type B E

Get Command IsEnabled GetIntegerv

Initial Value FALSE BACK

FRONT FACE

GetIntegerv

CCW

POLYGON SMOOTH

IsEnabled

FALSE

POLYGON MODE

GetIntegerv

FILL

POLYGON OFFSET CLAMP

POLYGON OFFSET UNITS

R R R

GetFloatv GetFloatv GetFloatv

0 0 0

POLYGON OFFSET POINT

IsEnabled

FALSE

POLYGON OFFSET LINE

IsEnabled

FALSE

POLYGON OFFSET FILL

IsEnabled

FALSE

Get value CULL FACE CULL FACE MODE

POLYGON OFFSET FACTOR

Description Polygon culling enabled Cull front-/back-facing polygons Polygon frontface CW/CCW indicator Polygon antialiasing on Polygon rasterization mode (front & back) Polygon offset clamp Polygon offset factor Polygon offset units Polygon offset enable for POINT mode rasterization Polygon offset enable for LINE mode rasterization Polygon offset enable for FILL mode rasterization

Sec. 14.6.1 14.6.1 14.6.1 14.6 14.6.4 14.6.5 14.6.5 14.6.5 14.6.5 14.6.5 14.6.5

592

Table 23.11: Multisampling

OpenGL 4.6 (Core Profile) - May 5, 2022

SAMPLE SHADING

Type B B B B R+ B B

Get Command IsEnabled IsEnabled IsEnabled IsEnabled GetFloatv GetBooleanv IsEnabled

Initial Value TRUE FALSE FALSE FALSE 1 FALSE FALSE

MIN SAMPLE SHADING VALUE

R+

GetFloatv

B n × Z+

IsEnabled GetIntegeri v

FALSE All bits of all words set

Get value MULTISAMPLE SAMPLE ALPHA TO COVERAGE SAMPLE ALPHA TO ONE SAMPLE COVERAGE SAMPLE COVERAGE VALUE SAMPLE COVERAGE INVERT

SAMPLE MASK SAMPLE MASK VALUE

Description Multisample rasterization Modify coverage from alpha Set alpha to max 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. 14.3.1 17.3.1 17.3.1 14.9.3 14.9.3 14.9.3 14.9.3 14.3.1.1 14.9.3 14.9.3

593

Type

Get Command

Initial Value

80 ∗ ×3 × Z +

GetIntegerv

TEXTURE BINDING 1D ARRAY

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING 2D ARRAY

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING CUBE MAP ARRAY

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING RECTANGLE

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING BUFFER

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING CUBE MAP

80 ∗ ×Z +

GetIntegerv

TEXTURE BINDING 2D MULTISAMPLE

80 ∗ ×Z +

GetIntegerv

80 ∗ ×Z +

GetIntegerv

Get value TEXTURE BINDING xD

Table 23.12: Textures (state per texture unit)

OpenGL 4.6 (Core Profile) - May 5, 2022

TEXTURE BINDING 2D MULTISAMPLE ARRAY

Description Texture object bound to TEXTURE_xD Texture object bound to TEXTURE_1D_ARRAY Texture object bound to TEXTURE_2D_ARRAY Texture object bound to TEXTURE_CUBE_MAP_ARRAY Texture object bound to TEXTURE_RECTANGLE Texture object bound to TEXTURE_BUFFER Texture object bound to TEXTURE_CUBE_MAP Texture object bound to TEXTURE_2D_MULTISAMPLE Texture object bound to TEXTURE_2D_MULTISAMPLE_ARRAY

Sec. 8.1 8.1 8.1 8.1

8.1 8.1 8.1 8.22

8.22

594

Type

Get Command

Initial Value

SAMPLER BINDING

80 ∗ ×Z +

GetIntegerv

TEXTURE xD

0 ∗ ×3 × I

GetTexImage

See ch. 8

TEXTURE 1D ARRAY

0 ∗ ×I

GetTexImage

See ch. 8

TEXTURE 2D ARRAY

0 ∗ ×I

GetTexImage

See ch. 8

TEXTURE CUBE MAP ARRAY

0 ∗ ×I

GetTexImage

See ch. 8

TEXTURE RECTANGLE

0 ∗ ×I

GetTexImage

See ch. 8

TEXTURE CUBE MAP POSITIVE X

0 ∗ ×I

GetTexImage

See sec. 8.5

TEXTURE CUBE MAP NEGATIVE X

0 ∗ ×I

GetTexImage

See sec. 8.5

TEXTURE CUBE MAP POSITIVE Y

0 ∗ ×I

GetTexImage

See sec. 8.5

TEXTURE CUBE MAP NEGATIVE Y

0 ∗ ×I

GetTexImage

See sec. 8.5

TEXTURE CUBE MAP POSITIVE Z

0 ∗ ×I

GetTexImage

See sec. 8.5

TEXTURE CUBE MAP NEGATIVE Z

0 ∗ ×I

GetTexImage

See sec. 8.5

Get value

Table 23.13: Textures (state per texture unit) (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Description Sampler object bound to active texture unit xD texture image at l.o.d. i 1D texture image at row i 2D texture image at slice i Cube map array texture image at l.o.d. i Rectangle texture image at l.o.d. zero +x face cube map texture image at l.o.d. i −x face cube map texture image at l.o.d. i +y face cube map texture image at l.o.d. i −y face cube map texture image at l.o.d. i +z face cube map texture image at l.o.d. i −z face cube map texture image at l.o.d. i

Sec. 8.2 8 8 8 8 8 8.5 8.5 8.5 8.5 8.5 8.5

595

TEXTURE SWIZZLE R

Type E

Get Command GetTexParameteriv

Initial Value RED

TEXTURE SWIZZLE G

GetTexParameteriv

GREEN

TEXTURE SWIZZLE B

GetTexParameteriv

BLUE

TEXTURE SWIZZLE A

GetTexParameteriv

ALPHA

TEXTURE BORDER COLOR

TEXTURE WRAP S

C E E E

GetTexParameterfv GetTexParameteriv GetTexParameteriv GetTexParameteriv

0.0,0.0,0.0,0.0 See sec. 8.22 LINEAR See sec. 8.22

TEXTURE WRAP T

GetTexParameteriv

See sec. 8.22

TEXTURE WRAP R

GetTexParameteriv

See sec. 8.22

TEXTURE TARGET

TEXTURE MAX LEVEL

E R R Z+ Z+

GetTextureParameteriv GetTexParameterfv GetTexParameterfv GetTexParameterfv GetTexParameterfv

NONE -1000 1000 0 1000

TEXTURE LOD BIAS

GetTexParameterfv

0.0

Get value

Table 23.14: Textures (state per texture object)

OpenGL 4.6 (Core Profile) - May 5, 2022

TEXTURE MIN FILTER TEXTURE MAG FILTER

TEXTURE MIN LOD 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) Target of texture object Min level-of-detail Max. level-of-detail Base texture array Max. texture array level Texture level-of-detail bias (biastexobj )

Sec. 8.10 8.10 8.10 8.10 8 8.14 8.15 8.14.2 8.14.2 8.14.2 8.11 8 8 8 8 8.14

596

Table 23.15: Textures (state per texture object) (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Initial Value

DEPTH STENCIL TEXTURE MODE

GetTexParameteriv

DEPTH_COMPONENT

TEXTURE COMPARE MODE TEXTURE COMPARE FUNC

E E

GetTexParameteriv GetTexParameteriv

NONE LEQUAL

IMAGE FORMAT COMPATIBILITY TYPE

GetTexParameteriv

See sec. 8.26

TEXTURE IMMUTABLE FORMAT

GetTexParameteriv

FALSE

–

Z+ Z+ Z+ Z+ Z+ S

GetTexParameteriv GetTexParameteriv GetTexParameteriv GetTexParameteriv GetTexParameteriv GetObjectLabel

0 0 0 0 0 empty

TEXTURE MAX ANISOTROPY

GetTexParameterfv

1.0

Get value

TEXTURE IMMUTABLE LEVELS TEXTURE VIEW MIN LEVEL TEXTURE VIEW NUM LEVELS TEXTURE VIEW MIN LAYER TEXTURE VIEW NUM LAYERS

Description Depth stencil texture mode Comparison mode Comparison function Compatibility rules for texture use with image units Size and format immutable Storage no. of levels View base texture level View no. of texture levels View min array layer View no. of array layers Debug label Maximum degree of anisotropy

Sec. 8.16 8.23 8.23 8.26 8.19 8.18 8.18 8.18 8.18 8.18 20.9 8.14

597

Table 23.16: Textures (state per texture image)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type Z+ Z+ Z+ Z+

Get Command GetTexLevelParameteriv GetTexLevelParameteriv GetTexLevelParameteriv GetTexLevelParameteriv

Initial Value 0 0 0 0

TEXTURE FIXED SAMPLE LOCATIONS

GetTexLevelParameteriv

TRUE

TEXTURE INTERNAL FORMAT

GetTexLevelParameteriv

RGBA or R8

6 × Z+

GetTexLevelParameteriv

TEXTURE SHARED SIZE

Z+

GetTexLevelParameteriv

TEXTURE x TYPE

GetTexLevelParameteriv

NONE

Get value TEXTURE WIDTH TEXTURE HEIGHT TEXTURE DEPTH TEXTURE SAMPLES

TEXTURE x SIZE

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 8.22) 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)

Sec. 8 8 8 8.8 8.8 8

8 8.11

598

Table 23.17: Textures (state per texture image) (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Initial Value

TEXTURE COMPRESSED

GetTexLevelParameteriv

FALSE

TEXTURE COMPRESSED IMAGE SIZE

Z+

GetTexLevelParameteriv

Z+

GetTexLevelParameteriv

TEXTURE BUFFER OFFSET

n×Z

GetTexLevelParameteriv

TEXTURE BUFFER SIZE

n×Z

GetTexLevelParameteriv

Get value

TEXTURE BUFFER DATA STORE BINDING

Description True if image has a compressed internal format Size (in ubytes) of compressed image 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. 8.7 8.7

8.1

8.9

599

Table 23.18: Textures (state per sampler object)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type C E E

Get Command GetSamplerParameterfv GetSamplerParameteriv GetSamplerParameteriv

Initial Value 0.0,0.0,0.0,0.0 LEQUAL NONE

TEXTURE LOD BIAS

GetSamplerParameterfv

0.0

TEXTURE MAX LOD

TEXTURE MIN LOD

R E E R

GetSamplerParameterfv GetSamplerParameteriv GetSamplerParameteriv GetSamplerParameterfv

1000 LINEAR NEAREST_MIPMAP_LINEAR -1000

TEXTURE MAX ANISOTROPY

GetSamplerParameterfv

1.0

TEXTURE WRAP S

GetSamplerParameteriv

REPEAT

TEXTURE WRAP T

GetSamplerParameteriv

REPEAT

TEXTURE WRAP R

GetSamplerParameteriv

REPEAT

–

GetObjectLabel

empty

Get value TEXTURE BORDER COLOR TEXTURE COMPARE FUNC TEXTURE COMPARE MODE

TEXTURE MAG FILTER TEXTURE MIN FILTER

Description Border color Comparison function Comparison mode Texture level-of-detail bias (biastexobj ) Max. level-of-detail Magnification function Minification function Min level-of-detail Maximum degree of anisotropy Texcoord s wrap mode Texcoord t wrap mode (2D, 3D, cube map textures only) Texcoord r wrap mode (3D textures only) Debug label

Sec. 8 8.23 8.23 8.14 8 8.15 8.14 8 8.14 8.14.2 8.14.2 8.14.2 20.9

600

Table 23.19: Texture Environment and Generation

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value ACTIVE TEXTURE

Type E

Get Command GetIntegerv

Initial Value TEXTURE0

Description Active texture unit selector

Sec. 10.2

601

STENCIL PASS DEPTH FAIL

Type 16 ∗ ×B 16 ∗ ×4 × Z B E Z+ Z+ E E

Get Command IsEnabledi GetIntegeri v IsEnabled GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv

Initial Value FALSE See sec. 14.9.2 FALSE ALWAYS See sec. 17.3.3 0 KEEP KEEP

STENCIL PASS DEPTH PASS

GetIntegerv

KEEP

STENCIL BACK FUNC

E Z+ Z+ E E E B E

GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv IsEnabled GetIntegerv

ALWAYS See sec. 17.3.3 0 KEEP KEEP KEEP FALSE LESS

Get value SCISSOR BOX

Table 23.20: Pixel Operations

OpenGL 4.6 (Core Profile) - May 5, 2022

SCISSOR TEST

STENCIL TEST STENCIL FUNC STENCIL VALUE MASK STENCIL REF STENCIL FAIL

STENCIL BACK VALUE MASK STENCIL BACK REF STENCIL BACK FAIL STENCIL BACK PASS DEPTH FAIL STENCIL BACK PASS DEPTH PASS DEPTH TEST DEPTH FUNC

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 buffer enabled Depth buffer test function

Sec. 14.9.2 14.9.2 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.3 17.3.4 17.3.4

602

Type

Get Command

Initial Value

BLEND

8 ∗ ×B

IsEnabledi

FALSE

BLEND SRC RGB

8 ∗ ×E

GetIntegeri v

ONE

BLEND SRC ALPHA

8 ∗ ×E

GetIntegeri v

ONE

BLEND DST RGB

8 ∗ ×E

GetIntegeri v

ZERO

BLEND DST ALPHA

8 ∗ ×E

GetIntegeri v

ZERO

BLEND EQUATION RGB

8 ∗ ×E

GetIntegeri v

FUNC_ADD

BLEND EQUATION ALPHA

8 ∗ ×E

GetIntegeri v

FUNC_ADD

Get value

Table 23.21: Pixel Operations (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

BLEND COLOR

GetFloatv

0.0,0.0,0.0,0.0

FRAMEBUFFER SRGB

IsEnabled

FALSE

DITHER

B B E

IsEnabled IsEnabled GetIntegerv

TRUE FALSE COPY

COLOR LOGIC OP LOGIC OP MODE

Description 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 sRGB update and blending enable Dithering enabled Color logic op enabled Logic op function

Sec. 17.3.6 17.3.6 17.3.6 17.3.6 17.3.6 17.3.6 17.3.6 17.3.6 17.3.6 17.3.8 17.3.9 17.3.9

603

Table 23.22: Framebuffer Control

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Initial Value

COLOR WRITEMASK

8 ∗ ×4 × B

GetBooleani v

(TRUE,TRUE,TRUE,TRUE)

DEPTH WRITEMASK

GetBooleanv

TRUE

STENCIL WRITEMASK

Z+

GetIntegerv

1’s

STENCIL BACK WRITEMASK

Z+

GetIntegerv

1’s

COLOR CLEAR VALUE

GetFloatv

0.0,0.0,0.0,0.0

DEPTH CLEAR VALUE

R+

GetFloatv

STENCIL CLEAR VALUE

Z+

GetIntegerv

Get value

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

Sec. 17.4.2 17.4.2 17.4.2 17.4.2 17.4.3 17.4.3 17.4.3

604

Table 23.23: Framebuffer (state per target binding point)

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

Initial Value

DRAW FRAMEBUFFER BINDING

Z+

GetIntegerv

READ FRAMEBUFFER BINDING

Z+

GetIntegerv

Description Framebuffer object bound DRAW_FRAMEBUFFER Framebuffer object bound READ_FRAMEBUFFER

Sec. to to

9.2 9.2

605

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.24: Framebuffer (state per framebuffer object) † This state is queried from the currently bound read framebuffer.

Get value

Type

DRAW BUFFERi

8 ∗ ×E

READ BUFFER

E +

Get Command

Minimum Value

GetIntegerv

See sec. 17.4.1

GetIntegerv GetFramebufferParameteriv GetFramebufferParameteriv

See sec. 18.2

FRAMEBUFFER DEFAULT WIDTH

FRAMEBUFFER DEFAULT HEIGHT

Z+

FRAMEBUFFER DEFAULT LAYERS

Z+

GetFramebufferParameteriv

FRAMEBUFFER DEFAULT SAMPLES

Z+

GetFramebufferParameteriv

FALSE

GetObjectLabel

empty

FRAMEBUFFER DEFAULT FIXED SAMPLE LOCATIONS –

0 0

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 Debug label

Sec. 17.4.1 18.2 9.2.1 9.2.1 9.2.1

9.2.1

9.2.1 20.9

606

Get value

Type

Table 23.25: Framebuffer (state per attachment point)

OpenGL 4.6 (Core Profile) - May 5, 2022

FRAMEBUFFER ATTACHMENT OBJECT TYPE

FRAMEBUFFER ATTACHMENT OBJECT NAME

Z+

FRAMEBUFFER ATTACHMENT TEXTURE LEVEL

Z+

FRAMEBUFFER ATTACHMENT TEXTURE CUBE MAP FACE

FRAMEBUFFER ATTACHMENT TEXTURE LAYER

FRAMEBUFFER ATTACHMENT LAYERED

FRAMEBUFFER ATTACHMENT COLOR ENCODING

FRAMEBUFFER ATTACHMENT COMPONENT TYPE

Z+

GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv

Minimum Value NONE

NONE

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

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

Size in bits of attached image’s x component; x is RED, GREEN, BLUE, ALPHA, DEPTH, or STENCIL

9.2.3

607

FRAMEBUFFER ATTACHMENT x SIZE

Get Command GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv GetFramebufferAttachmentParameteriv

Table 23.26: Renderbuffer (state per target and binding point)

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value RENDERBUFFER BINDING

Type Z

Get Command GetIntegerv

Initial Value 0

Description Renderbuffer object bound RENDERBUFFER

Sec. to

9.2.4

608

Table 23.27: Renderbuffer (state per renderbuffer object)

OpenGL 4.6 (Core Profile) - May 5, 2022

RENDERBUFFER INTERNAL FORMAT

Type Z+ Z+ E

Get Command GetRenderbufferParameteriv GetRenderbufferParameteriv GetRenderbufferParameteriv

Initial Value 0 0 RGBA

RENDERBUFFER RED SIZE

Z+

GetRenderbufferParameteriv

RENDERBUFFER GREEN SIZE

Z+

GetRenderbufferParameteriv

RENDERBUFFER BLUE SIZE

Z+

GetRenderbufferParameteriv

RENDERBUFFER ALPHA SIZE

Z+

GetRenderbufferParameteriv

RENDERBUFFER DEPTH SIZE

Z+

GetRenderbufferParameteriv

RENDERBUFFER STENCIL SIZE

Z+

GetRenderbufferParameteriv

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 20.9

609

UNPACK LSB FIRST

Type B B

Get Command GetBooleanv GetBooleanv

Initial Value FALSE FALSE

UNPACK IMAGE HEIGHT

Z+

GetIntegerv

UNPACK ALIGNMENT

Z+ Z+ Z+ Z+ Z+

GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv

0 0 0 0 4

UNPACK COMPRESSED BLOCK WIDTH

Z+

GetIntegerv

UNPACK COMPRESSED BLOCK HEIGHT

Z+

GetIntegerv

UNPACK COMPRESSED BLOCK DEPTH

Z+

GetIntegerv

UNPACK COMPRESSED BLOCK SIZE

Z+

GetIntegerv

PIXEL UNPACK BUFFER BINDING

Z+

GetIntegerv

Get value

Table 23.28: Pixels

OpenGL 4.6 (Core Profile) - May 5, 2022

UNPACK SWAP BYTES

UNPACK SKIP IMAGES UNPACK ROW LENGTH UNPACK SKIP ROWS UNPACK SKIP PIXELS

Description Value of UNPACK_SWAP_BYTES Value of UNPACK_LSB_FIRST 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 UNPACK_COMPRESSED_BLOCK_WIDTH Value of UNPACK_COMPRESSED_BLOCK_HEIGHT Value of UNPACK_COMPRESSED_BLOCK_DEPTH Value of UNPACK_COMPRESSED_BLOCK_SIZE Pixel unpack buffer binding

Sec. 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 8.4.1 6.7

610

PACK ALIGNMENT

Type B B Z+ Z+ Z+ Z+ Z+ Z+

Get Command GetBooleanv GetBooleanv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv GetIntegerv

Initial Value FALSE FALSE 0 0 0 0 0 4

PACK COMPRESSED BLOCK WIDTH

Z+

GetIntegerv

PACK COMPRESSED BLOCK HEIGHT

Z+

GetIntegerv

PACK COMPRESSED BLOCK DEPTH

Z+

GetIntegerv

PACK COMPRESSED BLOCK SIZE

Z+

GetIntegerv

PIXEL PACK BUFFER BINDING

Z+

GetIntegerv

Get value PACK LSB FIRST

Table 23.29: Pixels (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

PACK SWAP BYTES

PACK IMAGE HEIGHT PACK SKIP IMAGES PACK ROW LENGTH PACK SKIP ROWS PACK SKIP PIXELS

Description Value of PACK_SWAP_BYTES Value of PACK_LSB_FIRST Value of PACK_IMAGE_HEIGHT Value of PACK_SKIP_IMAGES Value of PACK_ROW_LENGTH Value of PACK_SKIP_ROWS Value of PACK_SKIP_PIXELS Value of PACK_ALIGNMENT Value of PACK_COMPRESSED_BLOCK_WIDTH Value of PACK_COMPRESSED_BLOCK_HEIGHT Value of PACK_COMPRESSED_BLOCK_DEPTH Value of PACK_COMPRESSED_BLOCK_SIZE Pixel pack buffer binding

Sec. 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2 18.2

611

Table 23.30: Shader Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value SHADER TYPE DELETE STATUS COMPILE STATUS – INFO LOG LENGTH – SHADER SOURCE LENGTH – SPIR V BINARY

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

GetShaderiv

FALSE

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 Shader is associated with a SPIR-V module.

Sec. 7.1 7.1 7.1 7.14 7.14 7.1 7.14 20.9 7.2

612

Get value

Type

Get Command

Initial Value

Table 23.31: Program Pipeline Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

ACTIVE PROGRAM

Z+

GetProgramPipelineiv

VERTEX SHADER

Z+

GetProgramPipelineiv

GEOMETRY SHADER

Z+

GetProgramPipelineiv

FRAGMENT SHADER

Z+

GetProgramPipelineiv

COMPUTE SHADER

Z+

GetProgramPipelineiv

TESS CONTROL SHADER

Z+

GetProgramPipelineiv

TESS EVALUATION SHADER

Z+

GetProgramPipelineiv

VALIDATE STATUS

GetProgramPipelineiv

FALSE

–

GetProgramPiplineInfoLog

empty

Z+ S

GetProgramPipelineiv GetObjectLabel

0 empty

INFO LOG LENGTH –

Description Program object updated by Uniform* when PPO bound Name of current vertex shader program object Name of current geometry shader program object Name of current fragment shader program object Name of current compute shader program object Name of current TCS program object Name of current TES 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 7.14 7.4 20.9

613

Get value

Type

Get Command

Initial Value

Table 23.32: Program Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

CURRENT PROGRAM

Z+

GetIntegerv

PROGRAM PIPELINE BINDING

Z+

GetIntegerv

PROGRAM SEPARABLE

GetProgramiv

FALSE

DELETE STATUS

GetProgramiv

FALSE

LINK STATUS

GetProgramiv

FALSE

VALIDATE STATUS

GetProgramiv

FALSE

ATTACHED SHADERS

Z+

GetProgramiv

–

0 ∗ ×Z +

GetAttachedShaders

empty

–

GetProgramInfoLog

empty

INFO LOG LENGTH

Z+

GetProgramiv

PROGRAM BINARY LENGTH

Z+

GetProgramiv

PROGRAM BINARY RETRIEVABLE HINT

GetProgramiv

FALSE

–

COMPUTE WORK GROUP SIZE

GetProgramBinary

–

3 × Z+

GetProgramiv

{0, . . . }

GetObjectLabel

empty

Sec. 7.3 7.4 7.3 7.3 7.3 7.3 7.14 7.14 7.14 7.3 7.5 7.5 7.5 19 20.9

614

–

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

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.33: Program Object State (cont.)

Get value ACTIVE UNIFORMS – – – – ACTIVE UNIFORM MAX LENGTH – ACTIVE ATTRIBUTES – – – – ACTIVE ATTRIBUTE MAX LENGTH

Get Command GetProgramiv

Type Z+ 0 ∗ ×Z +

0 ∗ ×Z 0 ∗ ×Z + 0 ∗ ×char

Initial Value 0

GetUniformLocation

–

GetActiveUniform GetActiveUniform GetActiveUniform

– – empty

Z+

GetProgramiv

− Z+

GetUniform GetProgramiv

0 0

GetAttribLocation

–

0 ∗ ×Z 0 ∗ ×Z + 0 ∗ ×char

GetActiveAttrib GetActiveAttrib GetActiveAttrib

– – empty

Z+

GetProgramiv

0 ∗ ×Z +

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 Location of active generic attribute Size of active attribute Type of active attribute Name of active attribute Max. active attribute name length

Sec. 7.6 7.14 7.6 7.6 7.6 7.14 7.6 11.1.1 11.1.1 11.1.1 11.1.1 11.1.1 7.14

615

Type Z+ E E

Get Command GetProgramiv GetProgramiv GetProgramiv

0 TRIANGLES TRIANGLE_STRIP

Z+

GetProgramiv

INTERLEAVED_ATTRIBS

Z+

GetProgramiv

Z+

GetProgramiv

–

Z+

–

Z+

–

0+ × char

GetTransformFeedbackVarying GetTransformFeedbackVarying GetTransformFeedbackVarying

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.34: Program Object State (cont.)

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 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. 11.3.4 11.3.1 11.3.2 11.3.4.2 7.14 7.14 7.14 11.1.2.1 11.1.2.1 11.1.2.1

616

Table 23.35: Program Object State (cont.)

OpenGL 4.6 (Core Profile) - 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

Z+

GetProgramiv

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

617

Get Command

Initial Value

Table 23.36: Program Object State (cont.)

OpenGL 4.6 (Core Profile) - 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

UNIFORM BLOCK DATA SIZE

Z+

GetActiveUniformBlockiv

UNIFORM BLOCK NAME LENGTH

Z+

GetActiveUniformBlockiv

Z+

GetActiveUniformBlockiv

n × Z+

GetActiveUniformBlockiv

UNIFORM BLOCK ACTIVE UNIFORMS

UNIFORM BLOCK ACTIVE UNIFORM INDICES

UNIFORM BLOCK REFERENCED BY VERTEX SHADER

UNIFORM BLOCK REFERENCED BY TESS CONTROL SHADER

UNIFORM BLOCK REFERENCED BY TESS EVALUTION SHADER

Description Uniform buffer array stride Uniform buffer intramatrix stride Whether uniform is a row-major matrix Uniform buffer binding point 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 tess. control stage True if uniform block is actively referenced by tess. evaluation 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

618

Get value

Type

Get Command

GetActiveUniformBlockiv

FALSE

TESS CONTROL OUTPUT VERTICES

Z+

GetProgramiv

TESS GEN MODE

GetProgramiv

QUADS

TESS GEN SPACING

GetProgramiv

EQUAL

TESS GEN VERTEX ORDER

GetProgramiv

CCW

TESS GEN POINT MODE

GetProgramiv

FALSE

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.37: Program Object State (cont.)

UNIFORM BLOCK REFERENCED BY GEOMETRY SHADER

UNIFORM BLOCK REFERENCED BY FRAGMENT SHADER

UNIFORM BLOCK REFERENCED BY COMPUTE SHADER

Initial Value

Description True if uniform block is actively referenced by the geometry stage True if uniform block is actively referenced by the fragment stage True if uniform block is referenced by the compute stage 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?

Sec. 7.6.2

7.6.2

7.6.2 11.2.1 11.2.2 11.2.2

11.2.2 11.2.2

619

Get value

Type

Get Command

ACTIVE SUBROUTINE UNIFORM -

6 × Z+

GetProgramStageiv

ACTIVE SUBROUTINE UNIFORMS

6 × Z+

GetProgramStageiv

ACTIVE SUBROUTINES

6 × Z+

GetProgramStageiv

6 × Z+

GetProgramStageiv

6 × Z+

GetProgramStageiv

LOCATIONS

Table 23.38: Program Object State (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

ACTIVE SUBROUTINE UNIFORM MAX LENGTH ACTIVE SUBROUTINE MAX LENGTH

Initial Value

GetActiveSubroutineUniformiv

6 × 0 ∗ ×0 ∗ ×Z +

GetActiveSubroutineUniformiv

UNIFORM SIZE

6 × 0 ∗ ×Z +

UNIFORM NAME LENGTH

6 × 0 ∗ ×Z +

–

6 × 0 ∗ ×S

–

6 × 0 ∗ ×S

–

6 × 0 ∗ ×S

GetActiveSubroutineUniformiv GetActiveSubroutineUniformiv GetActiveSubroutineUniformName GetActiveSubroutineName GetActiveSubroutineName

COMPATIBLE SUBROUTINES

Sec. 7.10 7.10 7.10 7.10 7.10 7.10

7.10 7.10 7.10

Sub. uniform name string

7.10

Length name

7.10

Subroutine name string

subroutine

7.10

620

6 × 0 ∗ ×Z +

NUM COMPATIBLE SUBROUTINES

Description No. of subroutine unif. locations in the shader No. of subroutine unif. variables in the shader No. of subroutine functions in the shader Max. subroutine uniform name length Max. subroutine name length No. of subroutines compatible with a sub. uniform List of subroutines compatible with a sub. uniform No. of elements in sub. uniform array Length of sub. uniform name

Get value ACTIVE ATOMIC COUNTER BUFFERS ATOMIC COUNTER BUFFER BIND-

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.39: Program Object State (cont.)

ING ATOMIC COUNTER BUFFER DATA SIZE ATOMIC COUNTER BUFFER ACTIVE ATOMIC COUNTERS ATOMIC COUNTER BUFFER ACTIVE ATOMIC COUNTER INDICES ATOMIC COUNTER BUFFER REFERENCED BY VERTEX SHADER

Type

Get Command

Z+

GetProgramiv

n × Z+ n × Z+ n × Z+ m × n × Z+ n×B

ATOMIC COUNTER BUFFER REFERENCED BY TESS CONTROL -

ATOMIC COUNTER BUFFER REFERENCED BY FRAGMENT SHADER ATOMIC COUNTER BUFFER REFERENCED BY COMPUTE SHADER UNIFORM ATOMIC COUNTER -

FALSE

n×B

GetActiveAtomicCounterBufferiv

FALSE

n×B n×B B n × Z+

GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv

FALSE FALSE FALSE

GetActiveUniformsiv -

Description No. of active atomic counter buffers (AACBs) used by a program Binding point associated with an AACB Min size required by an AACB No. of active atomic counters in an AACB List of active atomic counters in an AACB AACB has a counter used by vertex shaders AACB has a counter used by tess. control shaders AACB has a counter used by tess. evaluation shaders AACB has a counter used by geometry shaders AACB has a counter used by fragment shaders AACB has a counter used by compute shaders AACB associated with an active uniform

Sec. 7.7 7.7 7.7 7.7 7.7 7.7 7.7

7.7 7.7 7.7 7.7 7.7

621

BUFFER INDEX

FALSE

SHADER

ENCED BY GEOMETRY SHADER

GetActiveAtomicCounterBufferiv

ATOMIC COUNTER BUFFER REF-

ATOMIC COUNTER BUFFER REFER-

n×B

SHADER

ERENCED BY TESS EVALUTION -

GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv GetActiveAtomicCounterBufferiv

Initial Value

Table 23.40: Program Interface State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type +

ACTIVE RESOURCES

n×Z

MAX NAME LENGTH

n × Z+

MAX NUM ACTIVE VARIABLES

n × Z+

MAX NUM COMPATIBLE SUBROUTINES

n × Z+

Get Command GetProgramInterfaceiv GetProgramInterfaceiv GetProgramInterfaceiv GetProgramInterfaceiv

Initial Value 0 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 Max. no. of compatible subroutines for subroutine uniforms

Sec. 7.3.1 7.3.1 7.3.1 7.3.1

622

Get value

Type

Table 23.41: Program Object Resource State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get Command GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv

Z+

ARRAY SIZE

Z+

ARRAY STRIDE

Z+

ATOMIC COUNTER BUFFER INDEX

Z+

BLOCK INDEX

Z+

BUFFER BINDING

Z+

BUFFER DATA SIZE

Z+

COMPATIBLE SUBROUTINES

Z+

GetProgramResourceiv

IS PER PATCH

Z+

IS ROW MAJOR

Z+

LOCATION

Z+

LOCATION COMPONENT

Z+

LOCATION INDEX

Z+

MATRIX STRIDE

Z+

GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv

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 List of compatible subroutines for active subroutine uniform Is active input/output a per-patch attribute? Active resource stored as a row major matrix? Location assigned to active resource Location component assigned to active resources Location index 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 7.3.1 7.3.1 7.3.1 7.3.1

623

ACTIVE VARIABLES

Initial Value

Get value

Type

Table 23.42: Program Object Resource State (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Get Command GetProgramResourceiv GetProgramResourceiv

NAME LENGTH

Z+

NUM ACTIVE VARIABLES

Z+

NUM COMPATIBLE SUBROUTINES

Z+

GetProgramResourceiv

OFFSET

Z+

REFERENCED BY VERTEX SHADER

Z+

GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv GetProgramResourceiv

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+

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 7.3.1

624

TOP LEVEL ARRAY SIZE

Description Length of active resource name No. of active variables owned by active resource No. of compatible subroutines for active subroutine uniform 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 23.43: Vertex and Geometry Shader State(not part of program objects)

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value CURRENT VERTEX ATTRIB PROGRAM POINT SIZE

Type 16 ∗ ×R B

Get Command

Initial Value

GetVertexAttribfv

0.0,0.0,0.0,1.0

IsEnabled

FALSE

Description Current generic vertex attribute values Point size mode

Sec. 10.2 14.4

625

Table 23.44: Query Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value QUERY RESULT QUERY RESULT AVAILABLE – QUERY TARGET

Type Z+ B S E

Get Command GetQueryObjectuiv GetQueryObjectiv GetObjectLabel GetQueryObjectiv

Initial Value 0 or FALSE TRUE empty NONE

Description Query object result Is the query object result available? Debug label Target of query object

Sec. 4.2.3 4.2.3 20.9 4.2

626

Table 23.45: Image State (state per image unit)

OpenGL 4.6 (Core Profile) - 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

IMAGE BINDING ACCESS

8 ∗ ×E

GetIntegeri v

READ_ONLY

IMAGE BINDING FORMAT

8 ∗ ×Z +

GetIntegeri v

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. 8.26 8.26 8.26 8.26 8.26 8.26

627

Table 23.46: Atomic Counter Buffer Binding State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

ATOMIC COUNTER BUFFER BINDING

Z+

GetIntegerv

ATOMIC COUNTER BUFFER BINDING

n × Z+

GetIntegeri v

ATOMIC COUNTER BUFFER START

n × Z+

GetInteger64i v

ATOMIC COUNTER BUFFER SIZE

n × Z+

GetInteger64i v

Initial Value

Description Current value of generic atomic counter buffer binding 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. 6.8

6.8

6.8 6.8

628

Table 23.47: Shader Storage Buffer Binding State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

SHADER STORAGE BUFFER BINDING

Z+

GetIntegerv

SHADER STORAGE BUFFER BINDING

n × Z+

GetIntegeri v

SHADER STORAGE BUFFER START

n × Z+

GetInteger64i v

SHADER STORAGE BUFFER SIZE

n × Z+

GetInteger64i v

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

629

Table 23.48: Transform Feedback Object State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

Initial Value

TRANSFORM FEEDBACK BUFFER BINDING

n × Z+

GetIntegeri v

TRANSFORM FEEDBACK BUFFER START

n × Z+

GetInteger64i v

TRANSFORM FEEDBACK BUFFER SIZE

n × Z+

GetInteger64i v

TRANSFORM FEEDBACK PAUSED

GetBooleanv

FALSE

TRANSFORM FEEDBACK ACTIVE

GetBooleanv

FALSE

–

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. 6.7

6.7

6.7 6.7 6.7 20.9

630

Table 23.49: Uniform Buffer Binding State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Type

Get Command

UNIFORM BUFFER BINDING

Z+

GetIntegerv

UNIFORM BUFFER BINDING

n × Z+

GetIntegeri v

UNIFORM BUFFER START

n × Z+

GetInteger64i v

UNIFORM BUFFER SIZE

n × Z+

GetInteger64i v

Initial Value

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 6.7 6.7

631

Table 23.50: Sync Object State

OpenGL 4.6 (Core Profile) - 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 20.9

632

Table 23.51: Hints

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value LINE SMOOTH HINT POLYGON SMOOTH HINT TEXTURE COMPRESSION HINT FRAGMENT SHADER DERIVATIVE HINT

Type E E E

Get Command GetIntegerv GetIntegerv GetIntegerv

Initial Value DONT_CARE DONT_CARE DONT_CARE

GetIntegerv

DONT_CARE

Description Line smooth hint Polygon smooth hint Texture compression quality hint Fragment shader derivative accuracy hint

Sec. 21.5 21.5 21.5 21.5

633

Table 23.52: Compute Dispatch State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value DISPATCH INDIRECT BUFFER BINDING

Type Z+

Get Command GetIntegerv

Initial Value 0

Description Indirect dispatch buffer binding

Sec. 19

634

Type

Get Command

Minimum Value

GetIntegerv

See sec. 22.2

MAX COMBINED CLIP AND CULL DISTANCES

Z+ Z+ Z+

GetIntegerv GetIntegerv GetIntegerv

8 8 8

SUBPIXEL BITS

Z+

GetIntegerv

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.53: Implementation Dependent Values

CONTEXT RELEASE BEHAVIOR MAX CLIP DISTANCES MAX CULL DISTANCES

MAX ELEMENT INDEX

GetInteger64v

PRIMITIVE RESTART FOR PATCHES SUPPORTED

GetBooleanv

–

MAX 3D TEXTURE SIZE

GetIntegerv

2048

MAX TEXTURE SIZE

Z+

GetIntegerv

16384

MAX ARRAY TEXTURE LAYERS

Z+

GetIntegerv

2048

MAX TEXTURE LOD BIAS

GetFloatv

2.0

MAX CUBE MAP TEXTURE SIZE

Z+

GetIntegerv

16384

MAX RENDERBUFFER SIZE

Z+

GetIntegerv

16384

MAX TEXTURE MAX ANISOTROPY

GetFloatv

16.0

−1

Description Flush behavior when context is released Max. no. of user clipping planes Max. no. of user culling planes Max. combined no. of user clipping No. of bits of subpixel precision in screen xw and yw Max. element index Primitive restart support for PATCHES Max. 3D texture image dimension Max. 2D/1D texture image dimension Max. no. of layers for texture arrays Max. absolute texture level-ofdetail bias Max. cube map texture image dimension Max. width and height of renderbuffers Limit of maximum degree of anisotropy

Sec. 22.2 13.7 13.7 13.7 14 10.4 10.3.6 8.5 8.5 8.5 8.14 8.5 9.2.4 8.14

635

Get Command GetFloatv

Minimum Value See sec. 13.8.1

MAX VIEWPORTS

Z+

GetIntegerv

VIEWPORT SUBPIXEL BITS

Z+

GetIntegerv

VIEWPORT BOUNDS RANGE

2×R

GetFloatv

†

LAYER PROVOKING VERTEX

GetIntegerv

See sec. 11.3.4

VIEWPORT INDEX PROVOKING VERTEX

GetIntegerv

See sec. 11.3.4

2 × R+

GetFloatv

1,1

R+

GetFloatv

–

ALIASED LINE WIDTH RANGE

2 × R+

GetFloatv

1,1

SMOOTH LINE WIDTH RANGE

2 × R+

GetFloatv

1,1

SMOOTH LINE WIDTH GRANULARITY

R+

GetFloatv

–

MAX ELEMENTS INDICES

Z+

GetIntegerv

–

MAX ELEMENTS VERTICES

Z+

GetIntegerv

–

MAX VIEWPORT DIMS

Table 23.54: Implementation Dependent Values (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

POINT SIZE RANGE

POINT SIZE GRANULARITY

Description Max. viewport dimensions Max. no. of active viewports No. of bits of subpixel precision for viewport bounds Viewport bounds range [min, max] † (at least [−32768, 32767]) Vertex convention followed by gl_Layer Vertex convention followed by gl_ViewportIndex Range (lo to hi) of point sprite sizes Point sprite size granularity Range (lo to hi) of aliased line widths Range (lo to hi) of antialiased line widths Antialiased line width granularity Recommended max no. of DrawRangeElements indices Recommended max no. of DrawRangeElements vertices

Sec. 13.8.1 13.8.1 13.8.1

13.8.1 11.3.4 11.3.4 14.4 14.4 14.5 14.5 14.5 10.3

636

Type 2 × Z+

Get value

10.3

Type

Get Command

Minimum Value

GetIntegerv

2047

MAX VERTEX ATTRIB BINDINGS

Z16∗

GetIntegerv

MAX VERTEX ATTRIB STRIDE

GetIntegerv

2048

NUM COMPRESSED TEXTURE FORMATS

Z+

GetIntegerv

18 ∗ ×Z +

GetIntegerv

MAX TEXTURE BUFFER SIZE

Z+

GetIntegerv

65536

MAX RECTANGLE TEXTURE SIZE

Z+

GetIntegerv

16384

NUM PROGRAM BINARY FORMATS

Z+

GetIntegerv

0 ∗ ×Z +

GetIntegerv

N/A

Z+

GetIntegerv

0 ∗ ×Z +

GetIntegerv

SHADER COMPILER

GetBooleanv

TRUE

MIN MAP BUFFER ALIGNMENT

Z+

GetIntegerv

TEXTURE BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

COMPRESSED TEXTURE FORMATS

PROGRAM BINARY FORMATS

NUM SHADER BINARY FORMATS

SHADER BINARY FORMATS

Description Max. offset added to vertex buffer binding offset Max. no. of vertex buffers Max. vertex attribute stride No. of compressed texture formats Enumerated compressed texture formats No. of addressable texels for buffer textures Max. width & height of rectangle textures No. of program binary formats Enumerated program binary formats No. of shader binary formats Enumerated shader binary formats Shader compiler supported Min byte alignment of pointers returned by Map*Buffer Min. required alignment for texture buffer offsets

Sec. 10.3 10.3 10.3 8.7 8.7 8.9 8.5 7.5 7.5 7.2 7.2 7 6.3 8.9

637

MAX VERTEX ATTRIB RELATIVE OFFSET

Table 23.55: Implementation Dependent Values (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

† The value of TEXTURE_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum.

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Minimum Value

MAJOR VERSION

Z+

GetIntegerv

–

MINOR VERSION

Z+

GetIntegerv

–

CONTEXT FLAGS

Z+

GetIntegerv

–

CONTEXT PROFILE MASK

Z+

GetIntegerv

–

n×S

GetStringi

–

NUM EXTENSIONS

Z+

GetIntegerv

RENDERER

GetString

–

SHADING LANGUAGE VERSION

GetString

–

SHADING LANGUAGE VERSION

n×S

GetStringi

–

Z+

GetIntegerv

n×S

GetStringi

NUM SPIR V EXTENSIONS

Z+

GetIntegerv

VENDOR

GetString

–

VERSION

GetString

–

EXTENSIONS

NUM SHADING LANGUAGE VERSIONS

SPIR V EXTENSIONS

Description Major version no. supported Minor version no. supported Context full/forwardcompatible flag Context profile mask Supported individual extension names No. of individual extension names Renderer string Latest Shading Language version supported Supported Shading Language versions No. of supported Shading Language versions Supported SPIR-V extension names No. of supported SPIR-V extension names Vendor string OpenGL version supported

Sec. 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2 22.2

638

Table 23.56: Implementation Dependent Version and Extension Support

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.57: Implementation Dependent Vertex Shader Limits

Get value

Type

Get Command

Minimum Value

MAX VERTEX ATTRIBS

Z+

GetIntegerv

MAX VERTEX UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX VERTEX UNIFORM VECTORS

Z+

GetIntegerv

256

MAX VERTEX UNIFORM BLOCKS

Z+

GetIntegerv

14∗

MAX VERTEX OUTPUT COMPONENTS

Z+

GetIntegerv

MAX VERTEX TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX VERTEX ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX VERTEX ATOMIC COUNTERS

Z+

GetIntegerv

MAX VERTEX SHADER STORAGE BLOCKS

Z+

GetIntegerv

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

11.1.3.6

7.8

639

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.58: Implementation Dependent Tessellation Shader Limits

Get value

Type

Get Command

Minimum Value

MAX TESS GEN LEVEL

GetIntegerv

MAX PATCH VERTICES

Z+

GetIntegerv

MAX TESS CONTROL UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX TESS CONTROL TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX TESS CONTROL OUTPUT COMPONENTS

Z+

GetIntegerv

128

MAX TESS PATCH COMPONENTS

Z+

GetIntegerv

120

Z+

GetIntegerv

4096

MAX TESS CONTROL INPUT COMPONENTS

Z+

GetIntegerv

128

MAX TESS CONTROL UNIFORM BLOCKS

Z+

GetIntegerv

14∗

MAX TESS CONTROL ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX TESS CONTROL ATOMIC COUNTERS

Z+

GetIntegerv

MAX TESS CONTROL SHADER STORAGE BLOCKS

Z+

GetIntegerv

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

640

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 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 4.6 (Core Profile) - May 5, 2022

Minimum Value

Type

MAX TESS EVALUATION UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX TESS EVALUATION TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX TESS EVALUATION OUTPUT COMPONENTS

Z+

GetIntegerv

128

MAX TESS EVALUATION INPUT COMPONENTS

Z+

GetIntegerv

128

MAX TESS EVALUATION UNIFORM BLOCKS

Z+

GetIntegerv

14∗

Z+

GetIntegerv

Z+

GetIntegerv

Z+

GetIntegerv

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

641

Table 23.59: Implementation Dependent Tessellation Shader Limits (cont.)

Get Command

Get value

Type

Get Command

Minimum Value

Table 23.60: Implementation Dependent Geometry Shader Limits

OpenGL 4.6 (Core Profile) - May 5, 2022

Z+

GetIntegerv

1024

MAX GEOMETRY UNIFORM BLOCKS

Z+

GetIntegerv

14∗

MAX GEOMETRY INPUT COMPONENTS

Z+

GetIntegerv

MAX GEOMETRY OUTPUT COMPONENTS

Z+

GetIntegerv

128

MAX GEOMETRY OUTPUT VERTICES

Z+

GetIntegerv

256

MAX GEOMETRY TOTAL OUTPUT COMPONENTS

Z+

GetIntegerv

1024

MAX GEOMETRY TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX GEOMETRY SHADER INVOCATIONS

Z+

GetIntegerv

MAX VERTEX STREAMS

Z+

GetIntegerv

MAX GEOMETRY ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX GEOMETRY ATOMIC COUNTERS

Z+

GetIntegerv

MAX GEOMETRY SHADER STORAGE BLOCKS

Z+

GetIntegerv

Sec. 11.3.3 7.6.2 11.3.4.4 11.3.4.5 11.3.4

11.3.4

11.3.4 11.3.4.2 11.3.4.2 7.7 11.1.3.6 7.8

642

MAX GEOMETRY UNIFORM COMPONENTS

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 Total no. of vertex streams 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

OpenGL 4.6 (Core Profile) - May 5, 2022

Table 23.61: Implementation Dependent Fragment Shader Limits

Get value

Type

Get Command

Minimum Value

MAX FRAGMENT UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX FRAGMENT UNIFORM VECTORS

Z+

GetIntegerv

256

MAX FRAGMENT UNIFORM BLOCKS

Z+

GetIntegerv

14∗

MAX FRAGMENT INPUT COMPONENTS

Z+

GetIntegerv

128

MAX TEXTURE IMAGE UNITS

Z+

GetIntegerv

MIN PROGRAM TEXTURE GATHER OFFSET

GetIntegerv

-8

MAX PROGRAM TEXTURE GATHER OFFSET

Z+

GetIntegerv

MAX FRAGMENT ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX FRAGMENT ATOMIC COUNTERS

Z+

GetIntegerv

MAX FRAGMENT SHADER STORAGE BLOCKS

Z+

GetIntegerv

Description No. of components for fragment shader (FS) uniform variables No. of vectors for FS uniform variables Max. no. of FS uniform buffers per program Max. no. of components of inputs read by a FS No. of texture image units accessible by a FS Min texel offset for textureGather Max. texel offset for textureGather No. of atomic counter buffers accessed by a FS No. of atomic counters accessed by a FS No. of shader storage blocks accessed by a FS

Sec. 15.1 15.1 7.6.2 15.2.2 11.1.3.5 8.14.1 8.14.1 7.7 11.1.3.6 7.8

643

Get Command

MAX COMPUTE WORK GROUP COUNT

3 × Z+

GetIntegeri v

65535

MAX COMPUTE WORK GROUP SIZE

3 × Z+

GetIntegeri v

1024 (x, y), 64 (z)

MAX COMPUTE WORK GROUP INVOCATIONS

Z+

GetIntegerv

1024

MAX COMPUTE UNIFORM BLOCKS

Z+

GetIntegerv

14∗

MAX COMPUTE TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX COMPUTE ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX COMPUTE ATOMIC COUNTERS

Z+

GetIntegerv

MAX COMPUTE SHARED MEMORY SIZE

Z+

GetIntegerv

32768

MAX COMPUTE UNIFORM COMPONENTS

Z+

GetIntegerv

1024

MAX COMPUTE IMAGE UNIFORMS

Z+

GetIntegerv

Z+

GetIntegerv

Z+

GetIntegerv

Table 23.62: Implementation Dependent Compute Shader Limits

OpenGL 4.6 (Core Profile) - May 5, 2022

MAX COMBINED COMPUTE UNIFORM COMPONENTS

MAX COMPUTE SHADER STORAGE BLOCKS

Minimum Value

Description Max. no. of workgroups (WG) that may be dispatched by a single dispatch command (per dimension) Max. local size of a compute WG (per dimension) Max. total compute shader (CS) invocations in a single local WG Max. no. of uniform blocks per compute program Max. no. of texture image units accessible by a CS No. of atomic counter buffers accessed by a CS No. of atomic counters accessed 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 image variables in compute shaders No. of words for compute shader uniform variables in all uniform blocks, including the default No. of shader storage blocks accessed by a compute shader

Sec.

19 19

7.6.2

11.1.3.5 7.7 11.1.3.6

19.1

19.1 11.1.3

19.1

7.8

644

Type

Get value

Table 23.63: Implementation Dependent Aggregate Shader Limits

Get Command GetIntegerv GetIntegerv

Minimum Value -8 7

MAX UNIFORM BUFFER BINDINGS

Z+

GetIntegerv

MAX UNIFORM BLOCK SIZE

Z+

GetIntegerv

16384

UNIFORM BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

MAX COMBINED UNIFORM BLOCKS

Z+

GetIntegerv

70∗

MAX VARYING COMPONENTS

Z+

GetIntegerv

MAX VARYING VECTORS

Z+

GetIntegerv

MAX COMBINED TEXTURE IMAGE UNITS

Z+

GetIntegerv

MAX SUBROUTINES

Z+

GetIntegerv

256

MAX SUBROUTINE UNIFORM LOCATIONS

Z+

GetIntegerv

1024

MAX UNIFORM LOCATIONS

Z+

GetIntegerv

1024

MIN PROGRAM TEXEL OFFSET

Description Min texel offset allowed in lookup Max. texel offset allowed in lookup 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 components for output variables No. of vectors for output variables Total no. of texture units accessible by the GL Max. no. of subroutines per shader stage Max. no. of subroutine uniform locations per stage Max. no. of user-assignable uniform locations

Sec. 11.1.3.5 11.1.3.5 7.6.2 7.6.2 7.6.2 7.6.2 11.1.2.1 11.1.2.1 11.1.3.5 7.10 7.10 7.6

645

† The value of UNIFORM_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum.

OpenGL 4.6 (Core Profile) - May 5, 2022

MAX PROGRAM TEXEL OFFSET

Type Z Z

Get value

Get Command

Minimum Value

MAX ATOMIC COUNTER BUFFER BINDINGS

Z+

GetIntegerv

MAX ATOMIC COUNTER BUFFER SIZE

Z+

GetIntegerv

MAX COMBINED ATOMIC COUNTER BUFFERS

Z+

GetIntegerv

MAX COMBINED ATOMIC COUNTERS

Z+

GetIntegerv

MAX SHADER STORAGE BUFFER BINDINGS

Z+

GetIntegerv

MAX SHADER STORAGE BLOCK SIZE

Z+

GetInteger64v

227

MAX COMBINED SHADER STORAGE BLOCKS

Z+

GetIntegerv

SHADER STORAGE BUFFER OFFSET ALIGNMENT

Z+

GetIntegerv

256†

Description 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 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. 6.8 7.7 7.7 11.1.3.6 7.8 7.8 7.8 7.8

646

Table 23.64: Implementation Dependent Aggregate Shader Limits (cont.)

† The value of SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT is the maximum allowed, not the minimum.

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Minimum Value

MAX IMAGE UNITS

Z+

GetIntegerv

MAX COMBINED SHADER OUTPUT RESOURCES

Z+

GetIntegerv

MAX IMAGE SAMPLES

Z+

GetIntegerv

MAX VERTEX IMAGE UNIFORMS

Z+

GetIntegerv

MAX TESS CONTROL IMAGE UNIFORMS

Z+

GetIntegerv

MAX TESS EVALUATION IMAGE UNIFORMS

Z+

GetIntegerv

MAX GEOMETRY IMAGE UNIFORMS

Z+

GetIntegerv

MAX FRAGMENT IMAGE UNIFORMS

Z+

GetIntegerv

MAX COMBINED IMAGE UNIFORMS

Z+

GetIntegerv

Description No. of units for image load/store/atom Limit on active image units + shader storage blocks + fragment outputs Max. allowed samples for a texture level bound to an image unit 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 all shaders

Sec. 8.26

8.26

8.26 11.1.3.7 11.1.3.7 11.1.3.7 11.1.3.7 11.1.3.7 11.1.3.7

647

Table 23.65: Implementation Dependent Aggregate Shader Limits (cont.)

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

MAX COMBINED VERTEX UNIFORM COMPONENTS

MAX COMBINED GEOMETRY UNIFORM COMPONENTS

MAX COMBINED TESS CONTROL UNIFORM COMPONENTS

MAX COMBINED TESS EVALUATION UNIFORM COMPONENTS

MAX COMBINED FRAGMENT UNIFORM COMPONENTS

Type

Get Command

Minimum Value

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Z+

GetIntegerv

†

Description No. of words for vertex shader uniform variables in all uniform blocks (including default) No. of words for geometry shader uniform variables in all uniform blocks (including default) No. of words for TCS uniform variables in all uniform blocks (including default) No. of words for TES uniform variables in all uniform blocks (including default) No. of words for fragment shader uniform variables in all uniform blocks (including default)

Sec. 7.6.2

7.6.2

11.2.1.1

11.2.3.1

7.6.2

648

Table 23.66: Implementation Dependent Aggregate Shader Limits (cont.) † The minimum value for each stage is MAX_stage_UNIFORM_BLOCKS × MAX_UNIFORM_BLOCK_SIZE / 4 + MAX_stage_UNIFORM_COMPONENTS

Get value

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Initial Value

DEBUG CALLBACK FUNCTION

GetPointerv

NULL

DEBUG CALLBACK USER PARAM

GetPointerv

NULL

DEBUG LOGGED MESSAGES

Z+

GetIntegerv

Z+

GetIntegerv

DEBUG OUTPUT SYNCHRONOUS

IsEnabled

FALSE

DEBUG GROUP STACK DEPTH

Z+

GetIntegerv

DEBUG OUTPUT

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. 20.2 20.2 20.3

20.3

20.8 20.6 20

649

Table 23.67: 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 23.68: Implementation Dependent Debug Output State

OpenGL 4.6 (Core Profile) - May 5, 2022

Get Command

Minimum Value

Get value

Type

MAX DEBUG MESSAGE LENGTH

Z+

GetIntegerv

MAX DEBUG LOGGED MESSAGES

Z+

GetIntegerv

MAX DEBUG GROUP STACK DEPTH

Z+

GetIntegerv

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. 20.1

20.3 20.6 20.7

650

Table 23.69: Implementation Dependent Values (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get Command

Minimum Value

MAX FRAMEBUFFER WIDTH

Z+

GetIntegerv

16384

MAX FRAMEBUFFER HEIGHT

Z+

GetIntegerv

16384

MAX FRAMEBUFFER LAYERS

Z+

GetIntegerv

2048

MAX FRAMEBUFFER SAMPLES

Z+

GetIntegerv

MAX SAMPLE MASK WORDS

Z+

GetIntegerv

MAX SAMPLES

Z+

GetIntegerv

MAX COLOR TEXTURE SAMPLES

Z+

GetIntegerv

MAX DEPTH TEXTURE SAMPLES

Z+

GetIntegerv

MAX INTEGER SAMPLES

Z+

GetIntegerv

QUERY COUNTER BITS

n × Z+

GetQueryiv

See sec. 4.2.3

Z+

GetInteger64v

MAX SERVER WAIT TIMEOUT

Description 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 sample mask words Max. no. of samples supported for multisampling† Max. no. of samples supported for all color formats in a multisample texture† Max. no. of samples supported for all depth/stencil formats in a multisample texture† Max. no. of samples supported for all integer format multisample buffers† Asynchronous query counter bits Max. WaitSync timeout interval

Sec. 9.2.1 9.2.1 9.2.1 9.2.1 14.9.3 9.2.4

22.3

22.3 4.2.3 4.1.1

651

† These queries return the maximum no. of samples for all internal formats required to support multisampled rendering.

Get value

Table 23.70: Implementation Dependent Values (cont.)

OpenGL 4.6 (Core Profile) - May 5, 2022

† “ULP” means “Units in the Last Place” or “Unit of Least Precision”.

Get value

Type

Get Command

Minimum Value

MIN FRAGMENT INTERPOLATION OFFSET

GetFloatv

-0.5

MAX FRAGMENT INTERPOLATION OFFSET

GetFloatv

+0.5 - 1 ULP†

FRAGMENT INTERPOLATION OFFSET BITS

Z+

GetIntegerv

MAX DRAW BUFFERS

Z+

GetIntegerv

MAX DUAL SOURCE DRAW BUFFERS

Z+

GetIntegerv

MAX COLOR ATTACHMENTS

Z+

GetIntegerv

Description Furthest negative offset for interpolateAtOffset Furthest positive offset for interpolateAtOffset Subpixel bits for interpolateAtOffset Max. no. of active draw buffers Max. no. of active draw buffers when using dualsource blending Max. no. of FBO attachment points for color buffers

Sec. 15.1

15.1

15.1 17.4.1 17.3.6

9.2.7

652

Table 23.71: Internal Format Dependent Values

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value SAMPLES

NUM SAMPLE COUNTS

Type

Get Command

Minimum Value

0 ∗ ×Z +

GetInternalformativ

†

Z+

GetInternalformativ

Description Supported sample counts † See section 22.3 No. of supported sample counts

Sec. 22.3 22.3

653

MAX TRANSFORM FEEDBACK INTERLEAVED COMPONENTS

MAX TRANSFORM FEEDBACK SEPARATE ATTRIBS

MAX TRANSFORM FEEDBACK SEPARATE COMPONENTS

MAX TRANSFORM FEEDBACK BUFFERS

Type

Get Command

Minimum Value

Z+

GetIntegerv

Z+

GetIntegerv

Z+

GetIntegerv

Z+

GetIntegerv

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 Max. no. of buffer objs to write with transform feedback

Sec. 13.3

13.3

654

Table 23.72: Implementation Dependent Transform Feedback Limits

OpenGL 4.6 (Core Profile) - May 5, 2022

Get value

Table 23.73: Framebuffer Dependent Values

Get Command

DOUBLEBUFFER

GetBooleanv

–

STEREO

GetBooleanv

–

SAMPLE BUFFERS

Minimum Value

GetIntegerv

GetIntegerv GetMultisamplefv

0 –

IMPLEMENTATION COLOR READ FORMAT

GetIntegerv

†

IMPLEMENTATION COLOR READ TYPE

GetIntegerv

†

SAMPLES SAMPLE POSITION

Z n × 2 × R[0,1]

Description True if front & back buffers exist True if left & right buffers exist No. of multisample buffers Coverage mask size Explicit sample positions Implementation preferred pixel format Implementation preferred pixel type

Sec. 17.4.1 22 14.3.1 14.3.1 14.3.1 18.2.2 18.2.2

655

† Unlike most framebuffer-dependent state, which is queried from the currently bound draw framebuffer, this state is queried from the currently bound read framebuffer. n is the value of SAMPLES.

OpenGL 4.6 (Core Profile) - May 5, 2022

Type

Get value

Table 23.74: Miscellaneous

OpenGL 4.6 (Core Profile) - May 5, 2022

Type n×E n×B n × Z+ Z+

Get Command GetError – GetQueryiv GetIntegeriv

Initial Value 0 FALSE 0 0

COPY READ BUFFER BINDING

Z+

GetIntegerv

COPY WRITE BUFFER BINDING

Z+

GetIntegerv

RESET NOTIFICATION STRATEGY

GetIntegerv

See sec. 2.3.2

TEXTURE BUFFER BINDING

GetIntegerv

TEXTURE CUBE MAP SEAMLESS

IsEnabled

FALSE

Get value – – CURRENT QUERY QUERY BUFFER BINDING

Description Current error code(s) True if there is a corresponding error Active query object names Query result buffer binding 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 Seamless cube map filtering enable

Sec. 2.3.1 2.3.1 4.2.3 4.2.3 6.6 6.6 2.3.2 8.1 8.13

656

Appendix A

Invariance The OpenGL 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 and buffer variable stores and atomic operations, and atomic counter 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.

657

A.2. MULTI-PASS ALGORITHMS

A.2

658

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, either in a different color or using the XOR logical operation. • Using stencil operations to compute capping planes. 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.

A.3

Invariance Rules

For a given instantiation of an OpenGL 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 • Scissor parameters (other than enable) OpenGL 4.6 (Core Profile) - May 5, 2022

A.3. INVARIANCE RULES

659

• 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) • Logical operation 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 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 and buffer variable stores and atomic operations, and atomic counter operations), invariance rules 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 rules 1 and 4 apply to GL commands involving shader side effects: OpenGL 4.6 (Core Profile) - May 5, 2022

A.4. TESSELLATION INVARIANCE

660

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 or buffer variable stores or 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, atomic operations, or atomic counter operations, and subsequent GL commands read the memory written by those shader invocations, these operations must be explicitly synchronized. For more details, see section 7.13.

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, 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. OpenGL 4.6 (Core Profile) - May 5, 2022

A.4. TESSELLATION INVARIANCE

661

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 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. OpenGL 4.6 (Core Profile) - May 5, 2022

A.5. ATOMIC COUNTER INVARIANCE

662

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 4.6 (Core Profile) - May 5, 2022

A.6. WHAT ALL THIS MEANS

A.6

663

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 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 state values may change subtly when renderers are swapped. This is the type of state value change that invariance rule 1 in section A.3 seeks to avoid.

OpenGL 4.6 (Core Profile) - 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 revisions may change, and features deprecated in a previous revision may be removed. 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 point size and 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 rendering commands, even at the fragment level. 6. Because rasterization of non-antialiased polygons is point sampled, polygons that have no area generate no fragments when they are rasterized in FILL mode, and the fragments generated by the rasterization of “narrow” polygons may not form a continuous array. 664

665 7. OpenGL 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 (or if active, geometry) 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 4.6 (Core Profile) - May 5, 2022

Appendix C

The OpenGL SPIR-V Execution Environment C.1

Required Versions and Formats

Implementations must support the 1.0 version of SPIR-V and the 1.0 version of the SPIR-V Extended Instructions for the OpenGL Shading Language (see section 1.3.4). A SPIR-V module passed into ShaderBinary is interpreted as a series of 32bit words in host endianness, with literal strings packed as described in section 2.2 of the SPIR-V Specification. The first few words of the SPIR-V module must be a magic number and a SPIR-V version number, as described in section 2.3 of the SPIR-V Specification.

C.2

Valid SPIR-V Built-In Variable Decorations

Implementations must support the built-in variable decorations described in table C.1:

C.3

Valid SPIR-V Capabilities

Implementations must support the capability operands declared by OpCapability, described in table C.2.

666

C.4. VALIDATION RULES

C.4

667

Validation Rules

In addition to what is allowed and disallowed above: • Every entry point must have no return value and accept no arguments. • The Logical addressing model must be selected. • Scope for execution must be limited to – Workgroup – Subgroup • Scope for memory must be limited to – Device – Workgroup – Invocation • Storage Class must be limited to – UniformConstant – Input – Uniform – Output – Workgroup – Private – Function – AtomicCounter – Image • Images – OpTypeImage must declare a scalar 32-bit float or 32-bit integer type for the Sampled Type. – OpSampledImage, OpImageQuerySizeLod, and OpImageQueryLevels must only consume an Image operand whose type has its Sampled operand set to 1. – The Depth operand of OpTypeImage is ignored. OpenGL 4.6 (Core Profile) - May 5, 2022

C.4. VALIDATION RULES

668

– The Image format of OpImageWrite must not be Unknown, unless the StorageImageWriteWithoutFormat OpCapability was declared. • AtomicCounter storage class – Unless otherwise specified, variables declared in the AtomicCounter storage class must be unsigned 32-bit integers. – When using the Atomic Instructions, the Memory Semantics operand must be AtomicCounterMemory, without inclusion of memory-ordering bits (implying Relaxed semantics). • Decorations – The Flat, NoPerspective, Sample, and Centroid decorations must not be used on variables with storage class other than Input or on variables used in the interface of non-fragment shader entry points. – The Patch decoration must not be used on variables in the interface of a vertex, geometry, or fragment shader stage’s entry point. – OpTypeRuntimeArray must only be used for the last member of an OpTypeStruct in the Uniform storage class, and be decorated as BufferBlock

– If the Xfb Execution Mode is set, any output variable that is at least partially captured: * must be decorated with an XfbBuffer, declaring the capturing buffer * must have at least one captured output variable in the capturing buffer decorated with an XfbStride (and all such XfbStride values for the capturing buffer must be equal) – If the Xfb Execution Mode is set, any captured output: * must be a non-structure decorated with Offset or a member of a structure whose type member is decorated with Offset (not all members of such a struct need be captured) * must be a numerical type scalar, vector, matrix, or array of these * must have an Offset that is a multiple of its first component * must have an Offset that is a multiple of 8 if any captured output in the capturing buffer is a 64-bit type, in which case the corresponding XfbStride must also be a multiple of 8

OpenGL 4.6 (Core Profile) - May 5, 2022

C.5. PRECISION AND OPERATION OF SPIR-V INSTRUCTIONS

669

* must not overlap (i.e., alias in any capturing buffer) any other captured output * must not have an Offset that causes overflow of the XfbStride • Compute Shaders – For each compute shader entry point, either a LocalSize execution mode or an object decorated with the WorkgroupSize decoration must be specified. • Recursion – For all entry points, the static function-call graph rooted at that entry point must not contain cycles. • User declared input and output variables – must not use OpTypeBool, either directly or as part of an aggregate. • Atomic Access Rules – Atomic instructions must declare a scalar 32-bit integer type, or a scalar 64-bit integer type if the Int64Atomics capability is enabled, for the value pointed to by Pointer.

C.5

Precision and Operation of SPIR-V Instructions

The following rules apply to both single and double-precision floating point instructions: • Positive and negative infinities and positive and negative zeros are generated as dictated by IEEE-754, but subject to the precisions allowed in the following table. • Dividing a non-zero by a zero results in the appropriately signed IEEE-754 Inf . • Any denormalized value input into a shader or potentially generated by any instruction in a shader may be flushed to 0. • The rounding mode cannot be set and is undefined. • NaN s are not required to be generated. Instructions that operate on a NaN are not required to return a NaN as the result. OpenGL 4.6 (Core Profile) - May 5, 2022

C.6. PRECISION OF GLSL.STD.450 INSTRUCTIONS

670

• Support for signaling NaN s is optional and exceptions are never raised. The precision of double-precision instructions is at least that of single precision. For single precision (32 bit) instructions, precisions are required to be at least as described in table C.3.

C.6

Precision of GLSL.std.450 Instructions

GLSL.std.450 extended instructions specifically defined in terms of the instructions in table C.4 inherit the precisions described in that table. GLSL.std.450 extended instructions not listed in, or defined in terms of those instructions have undefined precision. These include, for example, the trigonometric functions and determinant. For the OpSRem and OpSMod instructions, if either operand is negative the result is undefined.

OpenGL 4.6 (Core Profile) - May 5, 2022

C.6. PRECISION OF GLSL.STD.450 INSTRUCTIONS

Built-in Variable Decoration BaseInstance BaseVertex ClipDistance CullDistance DrawIndex FragCoord FragDepth FrontFacing GlobalInvocationId HelperInvocation InstanceId InvocationId Layer LocalInvocationId LocalInvocationIndex NumWorkgroups PatchVertices PointCoord PointSize Position PrimitiveId SampleId SampleMask SamplePosition TessCoord TessLevelInner TessLevelOuter VertexId ViewportIndex WorkgroupId WorkgroupSize

Table C.1: Built-in Variable Decorations

OpenGL 4.6 (Core Profile) - May 5, 2022

671

C.6. PRECISION OF GLSL.STD.450 INSTRUCTIONS OpCapability Operand AtomicStorageOps AtomicStorage ClipDistance CullDistance DerivativeControl DrawParameters Float64 GeometryPointSize GeometryStreams Geometry Image1D ImageBuffer ImageCubeArray ImageGatherExtended ImageMSArray ImageQuery ImageRect InterpolationFunction Matrix MultiViewport SampleRateShading Sampled1D SampledBuffer SampledCubeArray SampledImageArrayDynamicIndexing SampledRect Shader StorageBufferArrayDynamicIndexing StorageImageArrayDynamicIndexing StorageImageExtendedFormats StorageImageMultisample StorageImageWriteWithoutFormat SubgroupVoteKHR TessellationPointSize Tessellation TransformFeedback UniformBufferArrayDynamicIndexing

Table C.2: Valid SPIR-V Capabilities OpenGL 4.6 (Core Profile) - May 5, 2022

672

C.6. PRECISION OF GLSL.STD.450 INSTRUCTIONS

Instruction OpFAdd OpFSub OpFMul OpFOrdEqual, OpFUnordEqual OpFOrdLessThan, OpFUnordLessThan OpFOrdGreaterThan, OpFUnordGreaterThan OpFOrdLessThanEqual, OpFUnordLessThanEqual OpFOrdGreaterThanEqual, OpFUnordGreaterThanEqual OpFDiv

conversions between types

673

Precision Correctly rounded Correctly rounded Correctly rounded Correct result Correct result Correct result Correct result Correct result 2.5 ULP for b in the range [2−126 , 2126 ] Correctly rounded

Table C.3: 32-Bit Precision

Instruction fma() exp(x), exp2(x) log(), log2() pow(x,y) sqrt() inversesqrt()

Precision Inherited from OpFMul followed by OpFAdd. (3 + 2|x|) ULP 3 ULP outside the range [0.5, 2.0]; absolute error < 2−21 inside the range [0.5, 2.0] Inherited from exp2(y * log2(x)) Inherited from 1.0 / inversesqrt() 2 ULP Table C.4: GLSL.std.450 Precision

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix D

Compressed Texture Image Formats The compressed texture formats used by OpenGL 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.

D.1

RGTC Compressed Texture Image Formats

RGTC formats are described in the “RGTC Compressed Texture Image Formats” chapter of the Khronos Data Format Specification. The mapping between OpenGL RGTC formats and that specification is shown in table D.1.

674

D.2. BPTC COMPRESSED TEXTURE IMAGE FORMATS OpenGL format COMPRESSED_RED_RGTC1 COMPRESSED_SIGNED_RED_RGTC1 COMPRESSED_RG_RGTC2 COMPRESSED_SIGNED_RG_RGTC2

675

Data Format Specification description BC4 unsigned BC4 signed BC5 unsigned BC5 signed

Table D.1: Mapping of OpenGL RGTC formats to descriptions.

OpenGL format COMPRESSED_RGBA_BPTC_UNORM COMPRESSED_SRGB_ALPHA_BPTC_UNORM COMPRESSED_RGB_BPTC_SIGNED_FLOAT COMPRESSED_RGB_BPTC_UNSIGNED_FLOAT

Data Format Specification description BC7, linear encoding BC7, sRGB encoding BC6H, signed format BC6H, unsigned format

Table D.2: Mapping of OpenGL BPTC formats to descriptions.

D.2

BPTC Compressed Texture Image Formats

BPTC formats are described in the “BPTC2 Compressed Texture Image Formats” chapter of the Khronos Data Format Specification. The mapping between OpenGL BPTC formats and that specification is shown in table D.2.

D.3

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 ETC formats and that specification is shown in table D.3.

OpenGL 4.6 (Core Profile) - May 5, 2022

D.3. ETC COMPRESSED TEXTURE IMAGE FORMATS

OpenGL 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

676

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 D.3: Mapping of OpenGL ETC formats to descriptions.

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix E

Profiles and the Deprecation Model OpenGL 3.0 introduced a deprecation model in which certain features are marked as deprecated. Deprecated features are expected to be completely removed from a future version of OpenGL. Deprecated features are summarized in section E.2. To aid developers in writing applications which will run on such future versions, it is possible to create an OpenGL context which does not support deprecated features. Such a context is called a forward compatible context, while a context supporting all OpenGL features is called a full context. Forward compatible contexts cannot restore deprecated functionality through extensions, but they may support additional, non-deprecated functionality through extensions. Profiles define subsets of OpenGL functionality targeted to specific application domains. Starting with OpenGL 3.2, two profiles are defined (see below). Future versions may define additional profiles addressing embedded systems or other domains. OpenGL implementations are not required to support all defined profiles, but must support the core profile described below. To enable application control of deprecation and profiles, new context creation APIs have been defined as extensions to GLX, WGL and EGL. These APIs allow specifying a particular version, profile, and full or forward compatible status, and will either create a context compatible with the request, or fail (if, for example, requesting an OpenGL version or profile not supported by the implementation). Only the ARB may define OpenGL profiles and deprecated features.

677

E.1. CORE AND COMPATIBILITY PROFILES

E.1

678

Core and Compatibility Profiles

The core profile of OpenGL defines essential functionality for the modern programmable shading model introduced in OpenGL 2.0, but does not include features marked as removed for that version of the Specification (see section E.2). The compatibility profile does not remove any functionality. It is not possible to implement both core and compatibility profiles in a single GL context, since the core profile mandates functional restrictions not present in the compatibility profile. Refer to the WGL_ARB_create_context_profile and GLX_ARB_create_context_profile extensions (see appendix K.3.3.66) for information on creating a context implementing a specific profile.

E.2

Deprecated and Removed Features

OpenGL 3.0 defined a set of deprecated features. OpenGL 3.1 removed most of the deprecated features and moved them into the optional GL_ARB_compatibility extension. The OpenGL 3.2 core profile removes the same features as OpenGL 3.1, while the optional compatibility profile supports all those features. Deprecated and removed features are summarized below in two groups: features which are marked deprecated by the core profile, but have not yet been removed, and features actually removed from the core profile of the current version of OpenGL (no features have been removed from or deprecated in the compatibility profile). Functions which have been removed will generate an INVALID_OPERATION error if called in the core profile or in a forward-compatible context. Functions which are partially removed (e.g. no longer accept some parameter values) will generate the errors appropriate for any other unrecognized value of that parameter when a removed parameter value is passed in the core profile or a forwardcompatible context. Functions which are deprecated but have not yet been removed from the core profile continue to operate normally except in a forward-compatible context, where they are also removed.

E.2.1

Deprecated But Still Supported Features

The following features are deprecated, but still present in the core profile. They may be removed from a future version of OpenGL, and are removed in a forwardcompatible context implementing the core profile. • Wide lines - LineWidth values greater than 1.0 will generate an INVALID_VALUE error. OpenGL 4.6 (Core Profile) - May 5, 2022

E.2. DEPRECATED AND REMOVED FEATURES

679

• Global component limit query - the implementation-dependent values MAX_VARYING_COMPONENTS and MAX_VARYING_FLOATS. • The

query targets NUM_COMPRESSED_TEXTURE_FORMATS COMPRESSED_TEXTURE_FORMATS (see section 8.5).

and

• Bitmap pack/unpack state for bitmaps - the pixel pack parameters UNPACK_LSB_FIRST and PACK_LSB_FIRST.

E.2.2

Removed Features

• Application-generated object names - the names of all object types, such as buffer, query, and texture objects, must be generated using the corresponding Gen* commands. Trying to bind an object name not returned by a Gen* command will result in an INVALID_OPERATION error. This behavior is already the case for framebuffer, renderbuffer, and vertex array objects. Object types which have default objects (objects named zero), such as vertex array, framebuffer, and texture objects, may also bind the default object, even though it is not returned by Gen*. • Color index mode - no color index visuals are supplied by the window system-binding APIs such as GLX and WGL, so the default framebuffer is always in RGBA mode. All language and state related to color index mode vertex, rasterization, and fragment processing behavior is removed. COLOR_INDEX formats are also deprecated. • OpenGL Shading Language versions 1.10 and 1.20. These versions of the shading language depend on many API features that have also been deprecated. • Begin / End primitive specification - Begin, End, and EdgeFlag*; Color*, FogCoord*, Index*, Normal3*, SecondaryColor3*, TexCoord*, Vertex*; and all associated state. Vertex arrays and array drawing commands must be used to draw primitives. However, VertexAttrib* and the current vertex attribute state are retained in order to provide default attribute values for disabled attribute arrays. • Edge flags and fixed-function vertex processing - ColorPointer, EdgeFlagPointer, FogCoordPointer, IndexPointer, NormalPointer, SecondaryColorPointer, TexCoordPointer, VertexPointer, EnableClientState, DisableClientState, and InterleavedArrays, ClientActiveTexture; Frustum, LoadIdentity, LoadMatrix, LoadTransposeMatrix, MatrixMode, OpenGL 4.6 (Core Profile) - May 5, 2022

E.2. DEPRECATED AND REMOVED FEATURES

680

MultMatrix, MultTransposeMatrix, Ortho, PopMatrix, PushMatrix, Rotate, Scale, and Translate; Enable/Disable targets RESCALE_NORMAL and NORMALIZE; TexGen* and Enable/Disable targets TEXTURE_GEN_*, Material*, Light*, LightModel*, and ColorMaterial, ShadeModel, and Enable/Disable targets LIGHTING, VERTEX_PROGRAM_TWO_SIDE, LIGHTi, and COLOR_MATERIAL; ClipPlane; and all associated fixed-function vertex array, multitexture, matrix and matrix stack, normal and texture coordinate, lighting, and clipping state. A vertex shader must be defined in order to draw primitives. Language referring to edge flags in the current specification is modified as though all edge flags are TRUE. Note that the FrontFace and ClampColor commands are not deprecated, as they still affect other non-deprecated functionality; however, the ClampColor targets CLAMP_VERTEX_COLOR and CLAMP_FRAGMENT_COLOR are deprecated. • Client vertex and index arrays - all vertex array attribute and element array index pointers must refer to buffer objects. The default vertex array object (the name zero) is also deprecated. Calling VertexAttribPointer when no buffer object or no vertex array object is bound will generate an INVALID_OPERATION error, as will calling any array drawing command when no vertex array object is bound. • Rectangles - Rect*. • Current raster position - RasterPos* and WindowPos*, and all associated state. • Two-sided color selection - Enable target VERTEX_PROGRAM_TWO_SIDE; OpenGL Shading Language built-ins gl_BackColor and gl_BackSecondaryColor; and all associated state. • Non-sprite points - Enable/Disable targets POINT_SMOOTH and POINT_SPRITE, and all associated state. Point rasterization is always performed as though POINT_SPRITE were enabled. • Wide lines and line stipple - LineWidth is not deprecated, but values greater than 1.0 will generate an INVALID_VALUE error; LineStipple and Enable/Disable target LINE_STIPPLE, and all associated state.

OpenGL 4.6 (Core Profile) - May 5, 2022

E.2. DEPRECATED AND REMOVED FEATURES

681

• Quadrilateral and polygon primitives - vertex array drawing modes POLYGON, QUADS, and QUAD_STRIP, related descriptions of rasterization of non-triangle polygons, and all associated state. • Separate polygon draw mode - PolygonMode face values of FRONT and BACK; polygons are always drawn in the same mode, no matter which face is being rasterized. • Polygon Stipple - PolygonStipple and Enable/Disable target POLYGON_STIPPLE, and all associated state. • Pixel transfer modes and operations - all pixel transfer modes, including pixel maps, shift and bias, color table lookup, color matrix, and convolution commands and state, and all associated state and commands defining that state. • Pixel drawing - DrawPixels and PixelZoom. However, the language describing pixel rectangles in section 8.4 is retained as it is required for TexImage* and ReadPixels. • Bitmaps - Bitmap and the BITMAP external format. • Legacy OpenGL 1.0 pixel formats - the values 1, 2, 3, and 4 are no longer accepted as internal formats by TexImage* or any other command taking an internal format argument. The initial internal format of a texel array is RGBA instead of 1. TEXTURE_COMPONENTS is deprecated; always use TEXTURE_INTERNAL_FORMAT. • Legacy pixel formats - all ALPHA, LUMINANCE, LUMINANCE_ALPHA, and INTENSITY external and internal formats, including compressed, floatingpoint, and integer variants; all references to luminance and intensity formats elsewhere in the specification, including conversion to and from those formats; and all associated state, including state describing the allocation or format of luminance and intensity texture or framebuffer components. • Depth texture mode - DEPTH_TEXTURE_MODE. Section 8.23.1 is to be changed so that r is returned to texture samplers directly, and the OpenGL Shading Language 1.30 Specification is to be changed so that (r, 0, 0, 1) is always returned from depth texture samplers in this case. • Texture wrap mode CLAMP - CLAMP is no longer accepted as a value of texture parameters TEXTURE_WRAP_S, TEXTURE_WRAP_T, or TEXTURE_WRAP_R. OpenGL 4.6 (Core Profile) - May 5, 2022

E.2. DEPRECATED AND REMOVED FEATURES

682

• Texture borders - the border value to TexImage* must always be zero, or an INVALID_VALUE error is generated (section 8.5); all language in section 8 referring to nonzero border widths during texture image specification and texture sampling; and all associated state. • Automatic mipmap generation - TexParameter* target MIPMAP, and all associated state.

GENERATE_-

• Fixed-function fragment processing - AreTexturesResident, PrioritizeTextures, and TexParameter target TEXTURE_PRIORITY; TexEnv target TEXTURE_ENV, and all associated parameters; TexEnv target TEXTURE_FILTER_CONTROL, and parameter name TEXTURE_LOD_BIAS; Enable targets of all dimensionalities (TEXTURE_1D, TEXTURE_2D, TEXTURE_3D, TEXTURE_1D_ARRAY, TEXTURE_2D_ARRAY, and TEXTURE_CUBE_MAP); Enable target COLOR_SUM; Enable target FOG, Fog, and all associated parameters; the implementation-dependent values MAX_TEXTURE_UNITS and MAX_TEXTURE_COORDS; and all associated state. • Alpha test - AlphaFunc and Enable/Disable target ALPHA_TEST, and all associated state. • Accumulation buffers - ClearAccum, and ACCUM_BUFFER_BIT is not valid as a bit in the argument to Clear (section 17.4.3); Accum; the ACCUM_*_BITS framebuffer state describing the size of accumulation buffer components; and all associated state. Window system-binding APIs such as GLX and WGL may choose to either not expose window configs containing accumulation buffers, or to ignore accumulation buffers when the default framebuffer bound to a GL context contains them. • Pixel copying - CopyPixels (the comments also applying to CopyTexImage will be moved to section 8.6). • Auxiliary color buffers, including AUXi targets of the default framebuffer. • Context framebuffer size queries - RED_BITS, GREEN_BITS, BLUE_BITS, ALPHA_BITS, DEPTH_BITS, and STENCIL_BITS. • Evaluators - Map*, EvalCoord*, MapGrid*, EvalMesh*, EvalPoint*, and all evaluator map enables, and all associated state. • Selection and feedback modes - RenderMode, InitNames, PopName, PushName, LoadName, and SelectBuffer; FeedbackBuffer and PassThrough; and all associated state. OpenGL 4.6 (Core Profile) - May 5, 2022

E.2. DEPRECATED AND REMOVED FEATURES

683

• Display lists - NewList, EndList, CallList, CallLists, ListBase, GenLists, IsList, and DeleteLists; all references to display lists and behavior when compiling commands into display lists elsewhere in the specification; and all associated state. • Hints - the PERSPECTIVE_CORRECTION_HINT, POINT_SMOOTH_HINT, FOG_HINT, and GENERATE_MIPMAP_HINT targets to Hint (section 21.5). • Attribute stacks - PushAttrib, PushClientAttrib, PopAttrib, PopClientAttrib, the MAX_ATTRIB_STACK_DEPTH, MAX_CLIENT_ATTRIB_STACK_DEPTH, ATTRIB_STACK_DEPTH, and CLIENT_ATTRIB_STACK_DEPTH state, the client and server attribute stacks, and the values ALL_ATTRIB_BITS and CLIENT_ALL_ATTRIB_BITS. • Unified extension string - EXTENSIONS target to GetString. • Token names and queries - all token names and queries not otherwise mentioned above for deprecated state, as well as all query entry points where all valid targets of that query are deprecated state (chapter 22 and the state tables).

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix F

Version 4.2 OpenGL version 4.2, released on August 8, 2011, is the fourteenth revision since the original version 1.0. Separate versions of the OpenGL 4.2 Specification exist for the core and compatibility profiles described in appendix E, respectively subtitled the “Core Profile” and the “Compatibility Profile”. This document describes the Core Profile. An OpenGL 4.2 implementation must be able to create a context supporting the core profile, and may also be able to create a context supporting the compatibility profile. Material specific to the compatibility profile specification is marked in a distinct color to clearly call out differences between the two profiles. The OpenGL 4.2 compatibility and core profiles are upward compatible with the OpenGL 4.1 compatibility and core profiles, respectively. Following are brief descriptions of changes and additions to OpenGL 4.2.

F.1

New Features

New features in OpenGL 4.2, including the extension or extensions if any on which they were based, include: • Support for BPTC compressed textures (ARB_texture_compression_bptc). • Allow pixel storage parameters to affect packing and unpacking of compressed textures (ARB_compressed_texture_pixel_storage). • Shader atomic counters (ARB_shader_atomic_counters). • Immutable texture images (ARB_texture_storage). 684

F.2. DEPRECATION MODEL

685

• Instanced transformed feedback drawing (ARB_transform_feedback_instanced). • Allow the offset within buffer objects used for instanced rendering to be specified (ARB_base_instance). • OpenGL Shading Language built-in functions allowing loads from and stores to texture images from any shader stage, and application control over the ordering of image load/store operations relative to other OpenGL pipeline operations accessing the same memory (ARB_shader_image_load_store). • New OpenGL Shading Language features with no OpenGL API impact (ARB_conservative_depth and ARB_shading_language_420pack see the OpenGL Shading Language Specification for details). • Queries for sample counts available for a given internal format and usage (ARB_internalformat_query). • More restrictive alignment constraints for mapped buffers (ARB_map_buffer_alignment).

F.2

Deprecation Model

The following features are newly deprecated by the OpenGL 4.2 core profile: • The

query targets NUM_COMPRESSED_TEXTURE_FORMATS COMPRESSED_TEXTURE_FORMATS (see section 8.5).

and

Features deprecated by OpenGL 4.1 remain deprecated, but have not yet been removed.

F.3

Changed Tokens

New token names are introduced to be used in place of old, less general names. However, the old token names continue to be supported, for backwards compatibility with code written for previous versions of OpenGL. The new names, and the old names they replace, are shown in table F.1. Note that COPY_READ_BUFFER and COPY_WRITE_BUFFER continue to be used as buffer targets for e.g. BindBuffer; the _BINDING forms are used only when querying the buffer object bound to those targets. OpenGL 4.6 (Core Profile) - May 5, 2022

F.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

686

New Token Name

Old Token Name

COPY_READ_BUFFER_BINDING COPY_WRITE_BUFFER_BINDING TRANSFORM_FEEDBACK_ACTIVE TRANSFORM_FEEDBACK_PAUSED

COPY_READ_BUFFER COPY_WRITE_BUFFER TRANSFORM_FEEDBACK_BUFFER_ACTIVE TRANSFORM_FEEDBACK_BUFFER_PAUSED

Table F.1: New token names and the old names they replace.

F.4

Change Log for Released Specifications

Changes in the specification update of January 19, 2012: • Corrections to figure 3.1 (Bug 7997). • Minor bugfixes and typos in sections 3, 10.3, 11.1, 11.1.1, 11.1.3.5, 4.2, 13.3.3, 14.5.2 (restored description of non-antialiased wide line rendering to the core profile since they are deprecated, but not yet removed), 8.2 (fixed prototypes for SamplerParameter commands), 15.2.1, 17.3.10 (specify that multisample buffer is only resolved at this time if the default framebuffer is bound), 9.2.8 (correct limits on layer for different types of attached textures), 9.4.2, 8.11 (remove redundant description by IsTexture that unbound object names created by GenTextures are not the names of texture objects), 23 (add GetInteger64v as a supported state query), chapter 5, and tables 23.31, 23.32, 23.55, and 23.72 (Bug 7895). • Add missing automatic unbinding of previously bound buffer objects for BindBufferRange and BindBufferBase in section 6.1.1 (Bug 8196). • More clearly specify interface matching rules for shader inputs and outputs in section 7.4.1, for cases where both sides of an interface are found in the same program and where they are in different programs (Bug 7030). • Clarify in section 11.1.1 that dvec3 and dvec4 vertex shader inputs consume only a single attribute location for the purpose of matching inputs to generic vertex attributes, but may consume two vectors for the purposes of determining if too many attribute vectors are used (Bug 7809). Also, add missing language describing the set of attributes consumed by matrix vertex attributes, with fixes to explicitly address dmat* types. • Remove dangling references to nonexistent gl_VerticesOut in section 11.2.1.2.3 (Bug 8357). OpenGL 4.6 (Core Profile) - May 5, 2022

F.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

687

• Fix names of cube map sampler type tokens in table 7.3 (Bug 8303). • Fix behavior of DeleteTransformFeedbacks in section 13.3.1 to generate an error if any of the objects being deleted has transform feedback active (Bug 8323). • Remove ambiguity in the order of operations and which vertices are appended by transform feedback when it is resumed in section 13.3.2 (Bug 8202). • Updated description of errors resulting from specifying texture images of level 1 or greater which exceed implementation-dependent limits, in sections 8.5 and 8.17.3 (Bug 8210). • Remove clamping of Dt and Dref prior to depth texture comparison in section 8.23.1, since it doesn’t reflect hardware reality (Bug 7975). • Update description of texture access from shadow samplers in section 15.2.1 to interact with texture swizzle (Bug 7962) and clarify that swizzling is not performed on the results of incomplete texture lookups (Bug 7917). • Add buffer clearing to the list of operations affected by scissor rectangle zero in section 14.9.2 (Bug 8368). • Remove error (from the core profile only) for querying CURRENT_VERTEX_ATTRIB for attribute zero with GetVertexAttrib* in section 7.14 (Bug 8352). • Clarify that the initial state of SAMPLE_MASK_VALUE is for all bits to be set in table 23.11 (Bug 8441). • Add missing PROGRAM_SEPARABLE state to table 23.32 (Bug 8442). • Numerous minor fixes to state table type fields and formatting (Bugs 8430, 8431). • Clarified that automatic unbinding of deleted objects, as described in section 5.1.2, does not affect attachments to unbound container objects the deleted objects are themselves attached to (Bug 8233). • Add version in which several extensions were introduced to core GL in section K.3 (Bug 8418). Changes in the specification update of August 22, 2011: OpenGL 4.6 (Core Profile) - May 5, 2022

F.5. CREDITS AND ACKNOWLEDGEMENTS

688

• More clearly specify interface matching rules for shader inputs and outputs in section 7.4.1, for cases where both sides of an interface are found in the same program and where they are in different programs (Bug 7030). • Clarify in section 11.1.1 that dvec3 and dvec4 vertex shader inputs consume only a single attribute location for the purpose of matching inputs to generic vertex attributes, but may consume two vectors for the purposes of determining if too many attribute vectors are used (Bug 7809). Also, add missing language describing the set of attributes consumed by matrix vertex attributes, with fixes to explicitly address dmat* types. Changes in the released specification of August 8, 2011: • Update name of MIN_MAP_BUFFER_ALIGNMENT to follow GL conventions in section 6.3 and table 23.55 (Bug 7825). • Change query object state description in section 4.2 so the initial state of the query result available flag agrees with the state table (Bug 7823). • Minor cleanups to atomic counter language in section 7.6 and to atomic counter token names in tables 23.57, 23.58, 23.60, and 23.61 (Bug 7834). • Clarify that completeness affects texture lookup and fetch operations in all shader stages in section 8.17 (Bug 7856). • Change BindImageTexture parameter name from index to unit and fix minor language issues in section 8.26 (Bugs 7744, 7850, 7851). • Fix typos in section 22.3 (Bug 7843). • Fix minimum maximums for MAX_FRAGMENT_IMAGE_UNIFORMS and MAX_COMBINED_IMAGE_UNIFORMS in table 23.65 (Bug 7805). • Change minimum maximum for MAX_ATOMIC_COUNTER_BUFFER_SIZE to 32 in table 23.64 (Bug 7855).

F.5

Credits and Acknowledgements

OpenGL 4.2 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ARB Working Group during the development of OpenGL 4.2, 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, including specific functionality developed in the form of OpenGL 4.6 (Core Profile) - May 5, 2022

F.5. CREDITS AND ACKNOWLEDGEMENTS

689

new ARB extensions together with OpenGL 4.2. In addition, many people participated in developing earlier vendor and EXT extensions on which the OpenGL 4.2 functionality is based in part; those individuals are listed in the respective extension specifications in the OpenGL Extension Registry. Acorn Pooley, NVIDIA Barthold Lichtenbelt, NVIDIA (Chair, Khronos OpenGL ARB Working Group) Benji Bowman, Imagination Technologies Bill Licea-Kane (Chair, ARB OpenGL Shading Language TSG, ARB_shader_atomic_counters) Bruce Merry, ARM (Detailed specification review, ARB_texture_storage) Chris Dodd, NVIDIA Christophe Riccio, Imagination Technologies Daniel Koch (ARB_internalformat_query) Eric Werness, NVIDIA (ARB_texture_compression_bptc) Graham Sellers, AMD (ARB_base_instance, ARB_conservative_depth, ARB_transform_feedback_instanced) Greg Roth, NVIDIA Ian Romanick, Intel (ARB_texture_storage) Jacob Ström, Ericsson AB Jan-Harald Fredriksen (ARB_internalformat_query) Jeannot Breton, NVIDIA Jeff Bolz, NVIDIA Corporation (ARB_shader_image_load_store) Jeremy Sandmel, Apple John Kessenich, Independent (OpenGL Shading Language Specification Editor, ARB_shading_language_420pack) Jon Leech, Independent (OpenGL API Specification Editor) Lingjun (Frank) Chen, Qualcomm Mark Callow, HI Corporation Maurice Ribble, Qualcomm Nick Haemel, AMD Pat Brown, NVIDIA Corporation (ARB_shader_image_load_store, ARB_shading_language_packing) Patrick Doane, Blizzard Pierre Boudier, AMD Piers Daniell, NVIDIA Corporation (ARB_compressed_texture_pixel_storage, ARB_map_buffer_alignment) Robert Simpson, Qualcomm Tom Olson, ARM (Chair, Khronos OpenGL ES Working Group)

OpenGL 4.6 (Core Profile) - May 5, 2022

F.5. CREDITS AND ACKNOWLEDGEMENTS

690

The ARB 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 4.6 (Core Profile) - May 5, 2022

Appendix G

Version 4.3 OpenGL version 4.3, released on August 6, 2012, is the fifteenth revision since the original version 1.0. Separate versions of the OpenGL 4.3 Specification exist for the core profile and compatibility profile described in appendix E, respectively subtitled the “Core Profile” and the “Compatibility Profile”. This document describes the Core Profile. An OpenGL 4.3 implementation must be able to create a context supporting the core profile, and may also be able to create a context supporting the compatibility profile. Material specific to the compatibility profile specification is marked in a distinct color to clearly call out differences between the two profiles. The OpenGL 4.3 compatibility and core profiles are upward compatible with the OpenGL 4.2 compatibility and core profiles, respectively (see appendix F). Following are brief descriptions of changes and additions to OpenGL 4.3. Descriptions of changes and additions in versions of OpenGL prior to 4.2 are omitted in this Specification, but may be found in the OpenGL 3.0 Specification (for features in versions 1.0 - 3.0, inclusive) and the OpenGL 4.2 Specification (for features in versions 3.1 - 4.1, inclusive). These Specifications are available in the OpenGL Registry.

G.1

Restructuring

The Specification has been substantially restructured to introduce high-level concepts and describe objects before their use, and more cleanly split descriptions of programmable and fixed-function processing. Chapter and section numbering has been aligned between the two profile Specifications so that a section number will always refer to the same concept in both profiles (although that section may be 691

G.2. NEW FEATURES

692

empty in the core profile).

G.2

New Features

New features in OpenGL 4.3, including the extension or extensions if any on which they were based, include: • ARB_arrays_of_arrays (OpenGL Shading Language only) • ARB_ES3_compatibility • ARB_clear_buffer_object • ARB_compute_shader • ARB_copy_image • ARB_debug_output • ARB_explicit_uniform_location • ARB_fragment_layer_viewport (OpenGL Shading Language only) • ARB_framebuffer_no_attachments • ARB_internalformat_query2 • ARB_invalidate_subdata • ARB_multi_draw_indirect • ARB_program_interface_query • ARB_robust_buffer_access_behavior • ARB_shader_image_size (OpenGL Shading Language only) • ARB_shader_storage_buffer_object • ARB_stencil_texturing • ARB_texture_buffer_range • ARB_texture_query_levels • ARB_texture_storage_multisample OpenGL 4.6 (Core Profile) - May 5, 2022

G.3. DEPRECATION MODEL

693

• ARB_texture_view • ARB_vertex_attrib_binding • KHR_debug • Add VERTEX_ATTRIB_ARRAY_LONG query for whether a vertex attribute is stored as an unconverted double (Bug 8272). • Add queries for #version strings of all OpenGL Shading Language versions supported by the GL (Bug 7811). • Increase required number of uniform blocks per program stage from 12 to 14 (Bug 8891).

G.3

Deprecation Model

The following features are deprecated by the OpenGL 4.3 core profile. • Bitmap pack/unpack state for bitmaps - the pixel pack parameters UNPACK_LSB_FIRST PACK_LSB_FIRST and (see sections 8.4.1 and 18.2). The following features which were previously deprecated have been reintroduced to the OpenGL 4.3 core profile: • The GetPointerv command (see section 22.2) and the STACK_OVERFLOW and STACK_UNDERFLOW errors (see table 2.3). These features are used by the debug functionality in chapter 20. Other features deprecated by OpenGL 4.2 remain deprecated, but have not yet been removed.

G.4

Changed Tokens

OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

694

New Token Name

Old Token Name

MAX_COMBINED_SHADER_OUTPUT_RESOURCES

MAX_COMBINED_IMAGE_UNITS_AND_FRAGMENT_OUTPUTS

Table G.1: New token names and the old names they replace.

G.5

Change Log for Released Specifications

Changes in the specification update of February 14, 2013: • Do not perform validity checks on the BindBufferRange size and offset arguments when a zero buffer is specified to unbind a buffer, in section 6.1.1 (Bug 9765). • Clean up descriptions of BindBufferBase in section 6.1.1 so it is described without reference to BindBufferRange, and note in section 6.7.1 that a zero size query result for a buffer binding is a sentinel indicating the entire buffer is bound (Bug 9513). • Fix typo in error descriptions in section 6.7 (Bug 9720). • Update section 6.8 to reference the tables of buffer binding state of different types, and move uniform buffer binding state from table 23.35 to new table 23.49 to match (Bug 9566). • Clarify that Uniform*d cannot be used to load uniforms with boolean types in section 7.6.1 (Bug 9345). • Added double-precision matrix types to the description of uniform buffer object storage layouts in section 7.6.2.1, and cleaned up description of the matrix stride and how to query it (Bug 9375). • Correct off-by-one error for valid range of sampler values in introduction to section 7.11 (Bug 8905). • Clarify in section 7.15 that table 23.43 is not part of program object state, and update the table caption to match (Bug 9781). • Clarify description of the data argument to TexSubImage* in section 8.6 so that it may not be NULL, unlike TexImage* (Bug 9750). • Fix typo in description of TexParameter* in section 8.10 (Bug 9625). OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

695

• Add a color-renderable column to table 8.12 and modify section 9.4 to define color-renderable formats with respect to the table, rather than with respect to base formats. This results in the RGB9_E5 format no longer being colorrenderable, which was an error (Bug 9338). • Allow vector forms of TexParameter* to be used to set scalar parameters in section 8.10, reversing an old spec change made in error (vector parameters, however, still cannot be set with the scalar calls) (Bug 7346). • Restore missing clamp for Dt and Dref (depth texture comparison mode parameters) in section 8.23.1 when using a fixed-point texture (Bug 7975). • Correct expshared to exps terminology and include missing N term when describing shared exponent texture color conversion and final conversion in sections 8.25 and 18.2.8 (Bug 9486). • Specify that FRAMEBUFFER is equivalent to DRAW_FRAMEBUFFER for GetFramebufferParameteriv in section 9.2.3 (Bug 9344). • Added STENCIL_INDEX8 as a required stencil-only renderbuffer format in sections 9.2.5 and 9.4.3, for compatibility with OpenGL ES 3.0 (Bug 9418). • Fixes to description of isoline tessellation in section 11.2.2.3 to describe use of outer tessellation levels in the correct order (Bug 9607). • Clamp values at specification time for DepthRange* (section 13.8.1) and ClearDepth (section 17.4.3), to avoid subtle issues when using floatingpoint depth buffers. However, this change does not reintroduce use of the clampf and clampd types eliminated in OpenGL 4.2 (Bug 9517). • Change DrawBuffer error for COLOR_ATTACHMENTm out of range from INVALID_VALUE to INVALID_OPERATION in section 17.4.1, to match DrawBuffers and OpenGL ES 3.0 (Bug 8568). • Modify language describing buffer writes in section 17.4.1 so that fragment colors are not written only to draw buffers with no color attachment, or with NONE as the draw buffer, allowing writes to other draw buffers to succeed. Specify that when only some output variables are written, only the fragment colors corresponding to unwritten variables are undefined (Bug 9494). • Allow attachment parameters to InvalidateSubFramebuffer in section 17.4.4 to include DEPTH_STENCIL_ATTACHMENT (Bug 9480).

OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

696

• Specify that the BlitFramebuffer mask may be zero in section 18.3.1 (Bug 9748). • Cleaned up language describing parameters to DebugMessageControl in section 20.4 to avoid triple negatives (Bug 9392). • Increase minimum value for MAX_UNIFORM_BUFFER_BINDINGS to 84 in table 23.63 to account for correct number of bindings/stage (14) (Bug 9424). Changes in the released Specification of August 6, 2012: • Restructured as described in section G.1. • Added new features as described in section G.2. • Add title image page using the “pipeline metro” diagram. • Miscellaneous minor typos and fixes to better match OpenGL ES 3.0 spec language (Bugs 7885, 7904, 7919). • Changed “rectangular texture” to “rectangle texture” throughout the spec for consistency (Bug 9262). Other consistency changes including using “equivalent to” consistently for pseudocode samples defining the operation of a command. • Many cleanups and additions to error language throughout the spec to add previously implicit errors explicitly (however, this is still a work in progress). In particular, added explicit errors for all commands taking program or shader arguments as described at the start of section 7.1, and for commands taking shadertype arguments (Bug 9145); and added explicit INVALID_VALUE errors for negative values of sizei and sizeiptr arguments (Bug 9320). • Cleaned up description of function prototypes from the old T notation to T * or const T * as appropriate for the actual C binding of the corresponding command. • Added

NUM_SHADING_LANGUAGE_VERSIONS and SHADING_LANGUAGE_VERSION queries for supported GLSL #version strings in sec-

tions 1.3.1, 1.3.3 and 22.2, and in table 23.56 (see Bug 7811). Still need enum assignments for these. • Remove assertion that draw and read framebuffers must be of the same class in section 2.1 (Bug 9134). OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

697

• Clarify in the caption to table 2.2 that sync is defined as a pointer type in the C binding (Bug 9140). • Reintroduced STACK_OVERFLOW and STACK_UNDERFLOW errors to the core profile in table 2.3, since they are used by the debug group APIs (Bug 9158). • Describe new, more complete error summary and typesetting style in section 2.3.1. Convert (most) error summaries beginning with section 4.1, adding implicit error conditions that have not been described with the commands they apply to before. This is a work in progress. • Clean up query objects in section 4.2 to clarify that TIME_ELAPSED and TIMESTAMP queries are different type of queries, and remove an inapplicable error condition for TIMESTAMP queries (Bug 9268). • Add language to DeleteBuffers in section 6 and BufferData in section 6.2 specifying that these commands cause any existing mappings of a buffer being operated on in any context to be unmapped, per a rather offhand reference in section 6.3.1 (Bug 9323). • Restore COPY_READ_BUFFER and COPY_WRITE_BUFFER as buffer target names in sections 6.1 and 6.6. The _BINDING aliases are used only when querying those binding points (Bugs 8475, 9115). • Bring compute shader language in sync with changes to the extension spec. In particular, add DISPATCH_INDIRECT_BUFFER binding section 6.1, describe it in section 10.3.11, update DispatchComputeIndirect to use it in section 19, add new table 23.52, and update aggregate shader limits in section 23.63 (Bug 9130). • Add create-on-bind behavior for BindBufferRange and BindBufferBase in section 6.1.1, mirroring BindBuffer (Bug 9216). • Clarify that offset and alignment constraints for ClearBufferSubData in section 6.2.1 are based on the total size of a texel of type internalformat (size of base type times no. of components) (Bug 9211). • Update errors for ClearBuffer*Data and mention them and InvalidateBuffer*Data among the commands that can modify buffer object storage in sections 6.2.1, 7.6, 7.8, and 7.13 (Bug 9154). • Clarify that buffer mappings are not affected by whether or not a context is current in section 6.3.1 (Bug 9323). OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

698

• Add language in section 6.3.2 specifying that commands which write to (as well as read from) mapped buffers are also supposed to generate errors (Bug 9115). • Make InvalidateBuffer*Data generate errors for invalid object handles in section 6.5 (Bug 9341). • Merge description of different types of indexed array buffer bindings into section 6.7.1, and move description of target-specific BindBufferRange errors into section 6.1.1 with reference to section 6.7.1 (Bug 9115 and general cleanup). • Extend ShaderBinary in section 7.2 to allow support for shader binary formats including all shader types, not just vertex and fragment shaders (Bug 9282). • Add description of “top-level arrays” to active shader storage block discussion in section 7.3.1 (Bug 9115). This probably needs to migrate back to the extension as well, along with a few other language changes in this section which Pat suggested in his PDF review but hasn’t put into the extension yet. • Clarify error descriptions for UseProgramStages and ActiveShaderProgram (section 7.4), UseProgram (section 7.6.1), and ProgramUniform* (section 7.6.1) to generate an INVALID_OPERATION error “if program has not been linked, or was last linked unsucessfully” rather than “if program has not been successfully linked” (Bug 8640, tracking similar changes to other commands previously). • Merge similar descriptions of uniform variable component limits for each separate shader stage into section 7.6. • Fix nonexistent token ATOMIC_COUNTER_ARRAY_STRIDE to UNIFORM_ARRAY_STRIDE in section 7.7.1 (Bug 9346). • Removed redundant definition of GetSubroutineUniformLocation from the beginning of section 7.10. • Added INVALID_VALUE error in section 7.11 if Uniform1i{v} is used to set a sampler to a value less than zero or greater than or equal to the value of MAX_COMBINED_TEXTURE_IMAGE_UNITS, matching the similar error for setting image uniforms. • Add VERTEX_ATTRIB_ARRAY_LONG state in section 7.14 and table 23.3 (Bug 8272). OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

699

• Add all specific compressed texture formats to the required format list in section 8.5.1. Include EAC and ETC2 format in specific format language in table 8.14 and sections 8.6, 8.7, and 8.24 (Bug 9156). • Add additional INVALID_OPERATION errors depending on odd combinations of read buffer and FBO state for CopyTexImage* and CopyTexSubImage* in section 8.6 (Bug 8559). • Disallow CopyTexImage* between sRGB and linear formats in section 8.6, and define BlitFramebuffer to linearize sRGB data from the read buffer in section 18.3.1 (Bug 8560). • Allow multisample texture targets as arguments to TexParameter* in section 8.10, with additional error conditions when attempting to set a disallowed min filter or base level parameter value. • Replace listings of all six cube map face selection targets with references to tables 8.19 or 9.3, in several places throughout the spec. • Fix error generated for invalid texture handle passed to TextureView in section 8.18 (Bug 9337). • Tweaked descriptions of transferring vertices in sections 10.3.4 and 10.4 to more closely match OpenGL ES 3.0 (Bug 8686). • Restore missing description of DrawElementsInstanced in section 10.4. • Renamed the formal parameter primcount to instancecount for DrawArraysInstancedBaseInstance, DrawArraysInstanced, DrawElementsInstancedBaseInstance, DrawElementsInstanced, DrawElementsInstancedBaseVertex, and DrawElementsInstancedBaseVertexBaseInstance (section 10.4), and for DrawTransformFeedbackInstanced and DrawTransformFeedbackStreamInstanced (section 13.3.3). Used equivalent terminology in the pseudocode descriptions of DrawElementsIndirect and DrawArraysIndirect (section 10.4). Renamed the formal parameter primcount to drawcount for MultiDrawArrays, MultiDrawArraysIndirect, MultiDrawElements, MultiDrawElementsIndirect, MultiDrawElementsBaseVertex (section 10.4) (Bug 9230). • Moved description of GetVertexAttrib* into section 10.5 (Bug 9115).

OpenGL 4.6 (Core Profile) - May 5, 2022

G.5. CHANGE LOG FOR RELEASED SPECIFICATIONS

700

• Specify in section 11.1.1 that special built-in inputs and outputs such as gl_VertexID should be enumerated in the PROGRAM_INPUT and PROGRAM_OUTPUT interfaces, as well as the legacy function GetActiveAttrib. Add spec language counting the built-ins gl_VertexID and gl_InstanceID against the active attribute limit (Bug 9201). • Swap order of tessellation levels in describing isoline tessellation in section 11.2.2.3, to match actual hardware (Bug 9195). • Remove language about deferred deletion for DeleteTransformFeedbacks in section 13.3.1 (Bug 8948). • Add transform feedback-related error for ProgramBinary, matching existing error for LinkProgram in section 13.3.2 when program is the name of a program being used by one or more transform feedback objects (Bug 7928). • Add description of MAX_COMPUTE_SHARED_MEMORY_SIZE in section 19.1, lifted from GLSL spec (Bug 9069). • Add description of the type of the debug callback function, including platform-dependent calling conventions, in section 20.2. • Remove inaccurate description of GLSL version string sort order in section 22.2. Instead, ensure that the most recent GLSL version corresponding to the context profile is returned first, and other entries have no defined ordering (Bug 7811). • Change Zn terminology used in state tables to describe enumerated state with n possible values to E throughout, since maintaining the n was always tricky as features were added and the possible values are fully described in the spec body. This affects hundreds of state table entries as well as adding a description of E in table 23.1. • Move

IMPLEMENTATION_COLOR_READ_FORMAT and IMPLEMENTATION_COLOR_READ_TYPE from table 23.73 to table 23.53

since they are not framebuffer-dependent values, unlike OpenGL ES (Bug 8561). • Increased minimum values for 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 to 14 in tables 23.57, 23.58, 23.59, 23.60, 23.61, and 23.62 respectively, and of MAX_COMBINED_UNIFORM_BLOCKS to 70 in table 23.63 (Bug 8891). OpenGL 4.6 (Core Profile) - May 5, 2022

G.6. CREDITS

701

• Added UNPACK_LSB_FIRST and PACK_LSB_FIRST state to the deprecated features list in section G.3 (Bug 7865).

G.6

Credits

OpenGL 4.3 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ARB Working Group during the development of OpenGL 4.3, 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, including specific functionality developed in the form of new ARB extensions together with OpenGL 4.3. In addition, many people participated in developing earlier vendor and EXT extensions on which the OpenGL 4.3 functionality is based in part; those individuals are listed in the respective extension specifications in the OpenGL Registry. Aaron Plattner, NVIDIA Acorn Pooley, NVIDIA Ahmet Oguz Akyuz, AMD Alex Eddy, Apple Inc Anton Staaf, Google Barthold Lichtenbelt, NVIDIA (Chair, Khronos OpenGL ARB Working Group) Benj Lipchak, Apple Benjamin Morris, NVIDIA Bill Licea-Kane (Chair, ARB OpenGL Shading Language TSG) Brent Wilson, NVIDIA Bruce Merry, Independent Chris Marrin, Apple Chris Niederauer, Apple Inc Christophe Riccio, AMD (ARB_debug_group, ARB_debug_label, ARB_shader_image_size, ARB_texture_query_levels, KHR_debug_output) Dan Omachi, Apple Inc Daniel Koch, TransGaming Inc. (ARB_internalformat_query2) Daniel Rakos, AMD Eric Werness, NVIDIA Georg Kolling, Imagination Technologies Graham Sellers, AMD (ARB_multi_draw_indirect, ARB_clear_buffer_object, ARB_compute_shader, ARB_copy_image, ARB_texture_buffer_range, ARB_texture_storage_multisample) OpenGL 4.6 (Core Profile) - May 5, 2022

G.6. CREDITS

702

Greg Roth, NVIDIA Henri Verbeet, CodeWeavers Jaakko Konttinen, AMD (ARB_debug_output) James Jones, NVIDIA Jan-Harald Fredriksen, ARM Jason Green, TransGaming Jean-François Roy, Apple Jeff Bolz, NVIDIA (ARB_invalidate_subdata, ARB_texture_view, ARB_vertex_attrib_binding) Joe Kain, NVIDIA John Kessenich, Independent (OpenGL Shading Language Specification Editor, ARB_arrays_of_arrays) Jon Leech, Independent (OpenGL API Specification Editor) Kenneth Russell, Google (ARB_robustness_isolation) Kent Miller, Apple Lingjun (Frank) Chen, Qualcomm Mark Callow, HI Corporation Mark Kilgard, NVIDIA (ARB_robustness) Mark Young, AMD Mathias Schott, NVIDIA Matt Collins, Apple Maurice Ribble, Qualcomm Michael Gold, NVIDIA (ARB_copy_image) Michael Morrison, NVIDIA Pat Brown, NVIDIA (ARB_framebuffer_no_attachments, ARB_program_interface_query, ARB_shader_storage_buffer_object) Pierre Boudier, AMD Piers Daniell, NVIDIA (ARB_ES3_compatibility, ARB_debug_output2, ARB_explicit_uniform_location, ARB_fragment_layer_viewport, ARB_robust_buffer_access_behavior, ARB_stencil_texturing) Richard Schreyer, Apple Seth Sowerby, Apple Thomas Volk, NVIDIA Tim Johansson, Opera Vladimir Vukicevic, Mozilla Yaki Tebeka, Graphic Remedy Yuan Wang, IMG

OpenGL 4.6 (Core Profile) - May 5, 2022

G.7. ACKNOWLEDGEMENTS

G.7

703

Acknowledgements

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix H

Version 4.4 OpenGL version 4.4, released on July 22, 2013, is the sixteenth revision since the original version 1.0. Separate versions of the OpenGL 4.4 Specification exist for the core profile and compatibility profile described in appendix E, respectively subtitled the “Core Profile” and the “Compatibility Profile”. This document describes the Core Profile. An OpenGL 4.4 implementation must be able to create a context supporting the core profile, and may also be able to create a context supporting the compatibility profile. Material specific to the compatibility profile specification is marked in a distinct color to clearly call out differences between the two profiles. The OpenGL 4.4 compatibility and core profiles are upward compatible with the OpenGL 4.3 compatibility and core profiles, respectively (see appendix G). Following are brief descriptions of changes and additions to OpenGL 4.4. Descriptions of changes and additions in versions of OpenGL prior to 4.2 are omitted in this Specification, but may be found in the OpenGL 3.0 Specification (for features in versions 1.0 - 3.0, inclusive) and the OpenGL 4.2 Specification (for features in versions 3.1 - 4.1, inclusive). These Specifications are available in the OpenGL Registry.

H.1

New Features

New features in OpenGL 4.4, including the extension or extensions if any on which they were based, include: • ARB_buffer_storage • ARB_clear_texture 704

H.2. DEPRECATION MODEL

705

• ARB_enhanced_layouts • ARB_multi_bind • ARB_query_buffer_object • ARB_texture_mirror_clamp_to_edge • ARB_texture_stencil8 • ARB_vertex_type_10f_11f_11f_rev • New implementation-dependent state MAX_VERTEX_ATTRIB_STRIDE, which constrains the maximum value of stride parameters to vertex array pointer-setting commands.

H.2

Deprecation Model

No new features are deprecated, and no previously deprecated features are reintroduced by the OpenGL 4.4 core profile. Features deprecated by OpenGL 4.3 remain deprecated, but have not yet been removed.

H.3

Change Log for Released Specifications

Changes in the released Specification update of March 19, 2014: • Fix typos and apply minor non-semantic language changes in several places to more closely match the OpenGL ES 3.1 Specification. • Tweak Enable / Disable language for different targets throughout the spec for more consistency. • Modify language in section 5.1.2 so that binding-related state is restored to default values after automatic unbinds (Bugs 10076, 11630). • Add samplers to the list of object types in section 5.1.3 for which deletion is delayed until the object is no longer in use, and remove transform feedback objects since deletion is defined to generate an error while they are active (Bugs 11374, 10079). • Add atomic counter binding, offset and stride assignments to the values reset by ProgramBinary in section 7.5 (bugfix from ES spec). OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

706

• Move Uniform* errors to section 7.6.1 where the Uniform* commands are defined, from sections 7.11 and 7.12 • Correct typo MAX_ATOMIC_COUNTER_BUFFERS to MAX_COMBINED_ATOMIC_COUNTER_BUFFERS in section 7.7, and specify new program interface query for atomic counter buffer data size query in section 7.7.2 • Restrict error for UniformSubroutinesuiv in section 7.10 to the case where no program stage is active for the shader stage identified by shadertype (Bug 11306). • Update DeleteSamplers language in section 8.2 to allow for the case where a sampler is bound to multiple texture units (bugfix from ES spec). • Fix conversion equation reference for SamplerParameteriv in section 8.2. Reorganize descriptions of exceptions to default data conversion rules here and in section 8.10 for TexParameter* (Bug 11185). • Add error in section 8.2 for calling scalar SamplerParameter{if} entry points with non-scalar param tokens, matching TexParameter* (Bug 11186). • Add error for TexImage3D in section 8.5 when specifying invalid border or target values with specific compressed texture formats, matching CompressedTexImage3D (Bug 11239). • Update table 8.12 in section 8.5.2 to make all remaining sized color formats texture-renderable that were not already, since the component size promotion rules mean there are already required texture formats with the same format and as least as many bits/component (Bug 11097). We discussed but did not also make these formats required renderbuffer formats, because doing this might imply a format change when a format supported at exactly the required component size for textures is only supported at a larger component size for renderbuffers (e.g. create texture at RGB4, use as a texture, then use as a renderbuffer and find it has to be converted to RGB565). • Change maximum allowed texture size for TexImage3DMultisample in section 8.8 to the value of MAX_ARRAY_TEXTURE_LAYERS (Bug 11135). • Remove explicit OUT_OF_MEMORY errors from TexImage*Multisample in section 8.8 and RenderbufferStorageMultisample in section 9.2.4. (public Bug 952). OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

707

• Add subsection headings and Errors sections for GetTexParameter* and GetTexLevelParameter* in section 8.11. • Change formal parameter names value and data to pname and params, following the headers and man pages, for GetTexParameter* and GetTexLevelParameter* (see section 8.11). (Bug 11523). • Added per-target maximum level-of-detail values to definition of GetTexLevelParameter* in section 8.11.3 (Bug 11136) • Add error for GetTexLevelParameter* in section 8.11 when querying multisample-specific parameters for non-multisample textures (this was imported from the OpenGL ES spec and it’s not certain it should be here). • Define behavior of GetTexLevelParameter* in section 8.11.3 for queries of multisample state from non-multisampled textures (Bug 11814). • 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 (Bug 11399). • Add cube map array texel arrays to the enumerated state in section 8.22, and remove redundant description of the bound buffer texture object name. • Add description of DEPTH_STENCIL_TEXTURE_MODE in section 8.22, and correct its type in table 23.15 (Bug 11770). • Add section 8.26.1 summarizing image unit binding state (Bug 10076). • Restructure error condition for FramebufferParameteri in section 9.2.1 to avoid ambiguity (Bug 11831). • 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 (Bug 10908). • Clean up language describing the GetFramebufferAttachmentParameteriv query for FRAMEBUFFER_ATTACHMENT_TEXTURE_LAYER in section 9.2.3 (Bug 11102). • Remove bogus error condition for FramebufferRenderbuffer, and add errors for invalid layer depending on texture type in section 9.2.7 (bugfix from ES spec). OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

708

• Set the vertex attribute array pointer state explicitly in the pseudocode for VertexAttrib*Pointer in section 10.3.2, and define queries of VERTEX_ATTRIB_ARRAY_BUFFER_BINDING and VERTEX_ATTRIB_ARRAY_DIVISOR to look up the corresponding state via the vertex attribute binding in section 10.5 (Bug 11789). • Rewrite section 10.3.11 to merge language for indirect drawing and indirect compute dispatch commands (Bug 11601). • Fix texel fetch limit on layer for array textures in section 11.1.3.2 (Bug 11770). • Fix typo in section 11.1.3.7 (Bug 11553). • Relax restrictions on required shader stages in sections 11.2 and 11.3 so that separable program objects containing tessellation and geometry shaders are not also required to contain a vertex shader (Bug 11508). • Modify tessellation primitive generation language in section 11.2.2 to discard patches only when relevant outer tessellation levels are NaNs, but not for non-relevant levels (Bug 11484). • Add missing error for invalid pname to GetMultisamplefv in section 14.3.1 (Bug 11134). • Change error for invalid mode* parameters to BlendEquation* in section 17.3.6.1 to INVALID_ENUM (bug 11354). • Fix error for invalid blending function arguments in section 17.3.6.2 to INVALID_ENUM (Bug 11770). • Add errors for ClearBuffer* in section 17.4.3.1 when the wrong type of buffer is passed to different forms of the command (Bug 11139). • Fix table reference in description of ClearBuffer* in section 17.4.3.1 from 17.5 to 17.4 so it’s clear that DRAW_BUFFERi can contain any legal value and the corresponding buffers will be cleared (Bug 11463). • Modify caption for table 17.4 in section 17.4.1 so it’s clear table entries refer only to selected buffers for the draw buffer, and remove redundant description of initial draw buffers state (Bugs 11462, 11463). • Moved language allowing undefined behavior for overlapping copies to the beginning of section 18.3, extended it to the case of overlapping copies via OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

709

texture views, applied it to all pixel copy operations in the section, and restricted the effects to undefined pixel values rather than general undefined behavior (Bug 11355). • Replace the “Get Value” of LABEL with - for debug labels of objects in tables 23.3, 23.6, 23.15, 23.18, 23.24, 23.27, 23.30, 23.31, 23.32, 23.44, 23.48, 23.50, and 23.74 (Bug 11131). • Rewrote descriptions of the vertex binding-specific state in table 23.4, added missing VERTEX_BINDING_DIVISOR and VERTEX_BINDING_BUFFER state, and marked all such state as part of the vertex-array attribute group (in the compatibility profile only) (Bug 10736). • Add missing state for framebuffers with no attachment to tables 23.24 and 23.69 (Bug 11187). • Add missing COMPUTE_SHADER state to program pipeline objects in table 23.31 (Bug 11539). Changes in the released Specification update of October 18, 2013: • Add footnote to section 2.3.1 noting that OUT_OF_MEMORY errors are not explicitly shown in command-specific Error sections because all GL commands can potentially generate them. Remove explicit OUT_OF_MEMORY errors from BufferStorage, BufferData, and MapBufferRange in section 6, and the TexStorage* commands in section 8.19 (public Bug 952). • Add new section 2.4 defining the term rendering commands, and modify language using this term in sections 7.11, 9.4.4, 10.3.9, 10.9, and 11.1.3.11 to use the new definition, sometimes narrowed to drawing commands, or to rendering commands which invoke shaders (Bug 10403). • Add footnote to section 2.6.9 specifying that undefined behavior results when mixing non-shared core API framebuffer objects and shared EXT extension framebuffer objects (Bug 10738). • Change description of BufferStorage MAP_COHERENT_BIT in section 6.2 to say that changes are “visible to any subsequently issued GL commands” rather than “immediately visible” (Public Bug 935). • Remove error for BufferStorage in section 6.2 which did not allow flags to contain MAP_WRITE_BIT while not also containing DYNAMIC_STORAGE_BIT (Bug 10561, public Bug 925). OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

710

• Rewrite section 6.3.2 describing the effects of mapped buffers on other commands and possible resulting errors or undefined behavior (Bug 10684). • Update GetProgramResourceiv in section 7.3.1 to return one for ARRAY_SIZE queries of non-arrays (for compatibility with GetActiveUniform) and zero for explicitly unsized arrays, and describe circumstances in which queries of ARRAY_SIZE and TOP_LEVEL_ARRAY_SIZE for unknown array sizes may return zero (Bugs 10641, 10647). • Allow ActiveShaderProgram and UseProgramStages to accept zero program values in section 7.4, to reset the corresponding program pipeline object state to its initial value (Public Bug 871). • Add additional const qualifier to parameter type of uniformNames for GetUniformIndices in section 7.6 (Bug 10703). • Fix typo from SHADER_STORAGE_BLOCK to SHADER_STORAGE_BUFFER for ShaderStorageBlockBinding in section 7.8 (Bug 10795). • Fix list of supported texture targets for stencil, depth, and depth+stencil formats in section 8.5 to include multisample targets (Bug 10558). • Move descriptions of required texture and renderbuffer/texture formats into tables 8.12 and 8.13, and update language in sections 8.5.1, 9.2.5, 9.4, and 9.4.3 to refer to those tables and to describe required framebuffer formats as color-renderable. Also, merged the no-longer-existent table 9.1 of renderbuffer-only stencil formats into table 8.13, and modified the definition of stencil-renderable in section 9.4 accordingly (Bug 9338). • Tag type of stencil fields in table 8.13 as ui (Bug 10748). • Add error for TexImage*DMultisample in section 8.8 when internalformat is not a valid format (Bug 11018). • Change type of internalformat argument to TexImage2DMultisample and TexImage3DMultisample in section 8.8 to enum, since the legacy internal formats are never accepted by these commands (Bug 10496). • Note that during computation of scale factors in section 8.14.1, implementations have chosen to perform clamping of intermediate terms in the sum of level-of-detail biases differently (Bug 9779). • Specify how border color values are clamped for compressed texture image formats in table 8.14 and section 8.14.2 (Bug 9476). OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

711

• Specify behavior of GetFramebufferAttachmentParameteriv in section 9.2.3 when querying texture attachments which have not yet specified a texture image or which do not yet have an allocated image store, and fix the error generated when querying a combined depth+stencil attachment to apply to the component type query, rather than the color encoding query (Bugs 9170, 10357). • Clean up errors for FramebufferTexture*D in section 9.2.8 (Bug 10674). • Merge rows of table 10.3 for VertexAttrib*Format and VertexAttrib*Pointer commands which are otherwise identical (Bug 10692). • Rewrite description of BindVertexBuffer in section 10.3.2 to clarify that it may create buffers in the same fashion as BindBuffer (Bug 10693). • Explicitly specify in section 10.3.2 that the VertexAttrib*Pointer commands may generate any of the errors defined by VertexAttrib*Format and VertexAttribBinding, since the pseudocode invokes those commands (Bug 10631). This was implicit in the previous spec language and has simply been called out more explicitly for clarity without replicating the errors. • Add PRIMITIVE_RESTART_FOR_PATCHES_SUPPORTED to section 10.3.6 and table 23.53 to allow querying primitive restart support for patch primitives (Bug 10364). Note that some implementations already could not support this, but had no way to indicate this. • Drop bogus references to “disabled attributes” in section 10.3.7 (Bug 10695). • Generate INVALID_VALUE errors consistently for misaligned indirect parameters to DrawArraysIndirect and DrawElementsIndirect in section 10.4, and DispatchComputeIndirect in section 19 (Bug 10385). • Clarify in sections 11.1.3.5 and 15.2.1 that swizzling during texture lookups of textures with depth component data is always performed, whether depth comparision is disabled or enabled (Bug 10702, Public Bug 749). • Restrict description in section 11.1.3.11 of checks required to be performed by ValidateProgram to only the errors described in that section (Bug 10650). • Remove a sentence fragment accidentally left in the description of ValidateProgramPipeline in section 11.1.3.11.

OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

712

• Specify in section 14.3.1 that the value of SAMPLE_BUFFERS is framebufferdependent, like SAMPLES (Bug 10688). • Remove redundant language describing depth texture lookups in section 15.2.1 and replace with a link to section 11.1.3.5, which also has some additional bugfixes that were not present here (Bug 10997). • Move a paragraph in section 15.2.2 to near the end of the section for better flow (Bug 10687). • Restrict language disallowing writing to multiple “classes” of fragment shader outputs in section 15.2.3 to compatibility profile only (Bug 10126). • Improve language describing undefined behavior when different color values are written to the same multiply-attached color buffer, and move it from section 17.3.9 to section 17.4.1 (Bug 10983). • Specify in section 18.3.1 that linearization of sRGB formats during reads performed by blending and blitting operations is controlled by the FRAMEBUFFER_SRGB enable (Bug 8560). • Change the type of FRAMEBUFFER_ATTACHMENT_LAYERED in table 23.25 from n × B to B. • Restore missing TEXTURE_BUFFER_BINDING query in table 23.74 for the buffer object bound to the corresponding bind point (Bug 4353, public Bug 844). Changes in the released Specification of July 22, 2013: • Added new features as described in section H.1. • Changed references throughout the spec to the value of SAMPLE_BUFFERS being “greater than zero” to “one”, since it can only take on values of zero and one. • Added introductory subsection 1.4 describing how to file bug reports against the GL and GLSL Specifications in the Khronos public Bugzilla (public bug 379). • Added a note that querying QUERY_RESULT_AVAILABLE for a query object will eventually succed in section 4.2.3 (Bug 9766).

OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

713

• Add views of an object’s data store to the list of conditions in sections 5 and 5.1.3 under which an object is considered in use for purposes of determining object lifetimes (Bug 10511). • Minor clarifications in sections 5.3.1, 8.21, 7.6.3, 7.7.2, 7.8, 13.1, and 15.2.1 (Bug 10346). • Added INVALID_VALUE error for BindBufferRange in section 6.1.1 if offset is negative (Bug 9873). • Add BufferSubData error for immutable storage without dynamic draw flag in section 6.2 (Bug 10326). • Modified prototypes for vector forms of the commands GetTexParameter* and GetTexLevelParameter* (section 8.11.1), and VertexAttrib4Nub (section 10.2.1), so that they are passed pointers rather than scalars as suggested by the ’T’ notation (public bug 273). • Clarify that no specific compressed one-dimensional texture formats are supported by the API, but may be by extensions, in section 8.7 (Bug 10388). • Restore errors for setting invalid rectangle texture parameters in section 8.10 (Bug 10208). • Move language constraining levelbase and levelmax for immutable textures from section 8.10 to section 8.14.3 (Bug 9342). • Added alternate mipmap level selection computation in equation 8.14 (Bug 10119). • Add forward references to pixel packing and PACK_* pixel storage modes from descriptions of texture queries and PixelStore in sections 8.11, 8.4.1, and 18.4 (Bug 10380). • Describe wrap mode application for MIRROR_CLAMP_TO_EDGE consistently with other modes in table 8.20, and use mod math operator instead of f mod() function since the inputs are described in non-scaled integer coordinates (Bug 10432). • Use actual VIEW_CLASS_* compatibility class names in table 8.22 instead of abstract class names, and refer to these tokens from the description of VIEW_COMPATIBILITY_CLASS in section 22.3.2 instead of enumerating them again (Bug 10518).

OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

714

• Change completeness condition for stencil index textures and stencil texturing from depth+stencil textures to match other integer textures in section 8.17 (Bug 10372). • Add number of samples to proxy texture state in section 8.22 and specify that its value is not checked for multisample textures (Bug 10171). • Add error for querying component type of a combined depth+stencil buffer with GetFramebufferAttachmentParameteriv in section 9.2.3, and note that the color encoding for non-color buffers is returned as LINEAR (Bug 9170). • Add conditions in section 9.4.1 for texture image attachments, making framebuffer attachment completeness dependent on valid mipmap level and mipmap completness of the image and including the case where texture image attachment is part of a cubemap texture, which must be mipmap cube complete (Bug 9689). • Remove redundant paragraph from section 10.2 (Bug 10311). • Limit stride parameters to BindVertexBuffer* and the generic vertex array specification commands in section 10.3 to the value of the new implementation-dependent state MAX_VERTEX_ATTRIB_STRIDE in table 23.55 (Bug 10229). • Move non-local error descriptions for *Indirect* rendering and dispatch commands to live with the commands themselves in sections 10.3.11, 10.4, and 19 (Bug 10385). • Only allow generated buffer object names in the core profile for BindVertexBuffer in section 10.3.2 (Bug 10486). • Added BlitFramebuffer to commands affected by conditional rendering in section 10.9 (Bug 9562). • Restore error for BindAttribLocation attribute variable names starting with the reserved "gl_" prefix to core profile in section 11.1.1 (Bug 10203). • Clarify that GetTransformFeedbackVarying may be used to query any transform feedback varying variable in section 11.1.2.1 (Bug 10472). • Add a non-local error in section 13.3.2 for GL commands that attempt to read or write to an active and unpaused transform feedback buffer (Bug 10193).

OpenGL 4.6 (Core Profile) - May 5, 2022

H.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

715

• Changed type of GetPolygonStipple argument to ubyte in section 14.6.2 to match shipping header files and be typesafe (Bug 10110). • Clean up description of GetFragData* in section 15.2.3 (Bug 10127). • Made default value for ALPHA_TEST_REF floating-point in section 17.3.2, and changed query in table 23.20 to GetFloatv (Bug 10128). • Strike redundant language describing default/FBO bindings under DrawBuffer (see section 17.4.1) and ReadBuffer (section 18.2.1). Bring language for the two sections in sync, and restore errors for ReadBuffer corresponding to DrawBuffer (Bug 10172). • Fix error specified for ColorMaski in section 17.4.2 to specify the correct formal parameter name, buf, and be restricted to that command (public bug 256). • Minimize use of generic ClearBuffer* terminology, since ClearBuffer*Data behave differently than the similarly named commands in section 17.4.3 with respect to conditional rendering and rasterizer discard (Bug 10312). • Clarify that ClearBuffer* can clear any valid draw buffers in section 17.4.3.1, and that the commands have no effect when the selected draw buffer has the value NONE (Bug 10537). • Relax BlitFramebuffer in section 18.3.1 so that format conversion can take place during multisample blits, since drivers already allow this and some apps depend on it. Simplified references to SAMPLE_BUFFERS by using the term “multisampled”. Cleaned up the Errors section and moved errors from spec body language into it (Bugs 9692, 10219). • Add const qualifier to userParam argument of DebugMessageCallback and DEBUGPROC in section 20.2. Specify that unrecognized message IDs in DebugMessageControl ids array are ignored in section 20.4. Specify in section 20.9 that GetDebugMessageLog messageLog parameter must be NULL when bufSize is less than zero, to allow an early out. (Bug 10083). • Reorganize section 22.3 to group internal format query pnames by the type of query and sort them, and add CLEAR_TEXTURE pname to match extension spec. • Add missing PRIMITIVE_RESTART_FIXED_INDEX to table 23.5 (Bug 10250). OpenGL 4.6 (Core Profile) - May 5, 2022

H.4. CREDITS

716

• Clean up state table entries to indicate enumerated types in tables 23.14, 23.18, 23.19, 23.25, and 23.27 (Bug 10251). • Fixed types of CLIP_DISTANCEi and DEPTH_RANGE in table 23.7 (Bugs 10106, 10107). • Add missing UNIFORM_BLOCK_NAME_LENGTH state to table 23.36 (Bug 8136). • Change minimum number of compressed texture formats to 18 in table 23.55, comprising the required specific formats in table 8.14 (Bug 7235).

H.4

Credits

OpenGL 4.4 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ARB Working Group during the development of OpenGL 4.4, 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, including specific functionality developed in the form of new ARB extensions together with OpenGL 4.4. In addition, many people participated in developing earlier vendor and EXT extensions on which the OpenGL 4.4 functionality is based in part; those individuals are listed in the respective extension specifications in the OpenGL Registry. Alon Or-bach, Samsung Aras Pranckevicius, Unity3D Barthold Lichtenbelt, NVIDIA (Chair, Khronos OpenGL ARB Working Group) Benj Lipchak, Apple Benji Bowman, Imagination Technologies Bill Licea-Kane, AMD Brent Insko, Intel Bruce Merry, Independent Christoph Kubisch, NVIDIA Christophe Riccio, AMD Daniel Koch, NVIDIA (GL_ARB_clear_texture, GL_ARB_texture_mirror_clamp_to_edge, GL_ARB_vertex_type_10f_11f_11f_rev) Daniel Rakos, AMD (GL_ARB_query_buffer_object) Graham Connor, Imagination Technologies Graham Sellers, AMD (GL_ARB_buffer_storage) Ian Romanick, Intel OpenGL 4.6 (Core Profile) - May 5, 2022

H.5. ACKNOWLEDGEMENTS

717

James Helferty, NVIDIA James Jones, NVIDIA Jeff Bolz, NVIDIA (GL_ARB_texture_stencil8) John Kessenich, Independent (OpenGL Shading Language Specification Editor, GL_ARB_enhanced_layouts) Jon Leech, Independent (OpenGL API Specification Editor) Jonathan Putsman, Imagination Technologies Karol Gasinski, Intel Larry Seiler, Intel Lingjun (Frank) Chen, Qualcomm Mark Callow, Artspark Nick Penwarden, Epic Games Pat Brown, NVIDIA (GL_ARB_multi_bind) Pierre Boudier, AMD Piers Daniell, NVIDIA Rob Barris, NVIDIA Robert Simpson, Qualcomm Slawomir Grajewski, Intel Tim Foley, Intel Tom Olson, ARM

H.5

Acknowledgements

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix I

Version 4.5 OpenGL version 4.5, released on August 11, 2014, is the seventeenth revision since the original version 1.0. Separate versions of the OpenGL 4.5 Specification exist for the core profile and compatibility profile described in appendix E, respectively subtitled the “Core Profile” and the “Compatibility Profile”. This document describes the Core Profile. An OpenGL 4.5 implementation must be able to create a context supporting the core profile, and may also be able to create a context supporting the compatibility profile. Material specific to the compatibility profile specification is marked in a distinct color to clearly call out differences between the two profiles. The OpenGL 4.5 compatibility and core profiles are upward compatible with the OpenGL 4.4 compatibility and core profiles, respectively (see appendix H). Following are brief descriptions of changes and additions to OpenGL 4.5. Descriptions of changes and additions in versions of OpenGL prior to 4.2 are omitted in this Specification, but may be found in the OpenGL 3.0 Specification (for features in versions 1.0 - 3.0, inclusive) and the OpenGL 4.2 Specification (for features in versions 3.1 - 4.1, inclusive). These Specifications are available in the OpenGL Registry.

I.1

New Features

New features in OpenGL 4.5, including the extension or extensions if any on which they were based, include: • GL_ARB_clip_control • GL_ARB_cull_distance 718

I.2. DEPRECATION MODEL

719

• GL_ARB_ES3_1_compatibility • GL_ARB_conditional_render_inverted • GL_KHR_context_flush_control • GL_ARB_derivative_control (OpenGL Shading Language Only) • GL_ARB_direct_state_access • GL_ARB_get_texture_sub_image • GL_KHR_robustness • GL_ARB_shader_texture_image_samples (OpenGL Shading Language Only) • GL_ARB_texture_barrier

I.2

Deprecation Model

No new features are deprecated, and no previously deprecated features are reintroduced by the OpenGL 4.5 core profile. Features deprecated by OpenGL 4.4 remain deprecated, but have not yet been removed.

I.3

Change Log for Released Specifications

Changes in the released Specification update of June 29, 2017 • Updated document copyright to new standard Khronos specification copyright. • 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). • Clarify that program object state queries in section 7.14 return the state presently in effect, which may be different than most recently set state (Bug 9702). • Clarify in section 7.14 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). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

720

• 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 (Bug 12249). • 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 8.14 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 (Bug 11147). • Add a description in section 9.2.3 of the case when querying Get*FramebufferAttachmentParameteriv for FRAMEBUFFER_ATTACHMENT_OBJECT_TYPE returns NONE, and the default framebuffer attachment is not allocated (Bug 16189). • Add an error to FramebufferTexture*D and *FramebufferTextureLayer in section 9.2.8 when attempting to attach a non-existent level of an immutable texture, and make the language in these sections closer to OpenGL ES (Bugs 15946,16214). • Add clarification that one of the error conditions for *DrawBuffers in section 17.4.1 applies only to the default framebuffer (Bug 16153). • Change the initial QUERY_RESULT_AVAILABLE state for query objects to TRUE in table 23.44, consistent with other changes made in Public Bug 1213 (Bug 16089). • Update the extension summaries in appendix K.3.3 to reflect all currently published ARB extensions. Changes in the released Specification update of October 24, 2016 • 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). • Remove required clamping of texture border colors in section 8.14.2 and make out-of-range values undefined, since not all hardware supports this (Bug 14442). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

721

• Add missing error for VertexAttrib*Pointer in section 10.3.2 of the core profile when no vertex array object is bound (Bug 10694). • Add validation errors in section 11.1.3.11 for multiple program objects exceeding any of the COMBINED resource limits, not just shader storage blocks (Bug 8834). • 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 update of July 7, 2016 • Specify that queries returning unsigned integers will clamp negative state values in section 2.2.1 (Bug 14444). • Clean up error language around reuse of query objects in section 4.2 to make clear that occlusion query objects created with targets ANY_SAMPLES_PASSED or ANY_SAMPLES_PASSED_CONSERVATIVE may be specified and reused with either of those query targets (Bug 13342). • Use consistent phrasing of “has/has been linked successfully” to describe the link status of programs where relevant in the descriptions of conditions for 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, GetActiveAttrib in section 11.1.1, and GetFragDataLocation and GetFragDataIndex in section 15.2.3. This replaces a variety of previous usages including the rather wordy “has not been linked, or was linked unsuccessfully” (Bug 8640). • 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). • Fixed name of formal parameter of SamplerParameter*v in section 8.2 to params (Bug 14158). • Relax error that prevented CompressedTex*SubImage3D from accepting TEXTURE_CUBE_MAP and TEXTURE_CUBE_MAP_ARRAY in section 8.7 (Bug 14366). • Simplify error language for TexParameter* and TextureParameter* in section 8.10 (Bug 15457). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

722

• Fix name of formal parameter target to GetnTexImage in section 8.11.4. • 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 Get*FramebufferAttachmentParameteriv in section 9.2.3, *FramebufferRenderbuffer in section 9.2.7, *FramebufferTexture* in section 9.2.8, *DrawBuffers in section 17.4.1, and InvalidateSubFramebuffer and InvalidateNamedFramebufferSubData in section 17.4.4 (Bug 13858). • Add missing validation error for the target argument of BindRenderbuffer in section 9.2.4 (Bug 14283). • Add missing constraint for two-dimensional multisample array textures on the level argument of *FramebufferTextureLayer in section 9.2.8 (Bug 14189). • Rewrite layer validation errors for *FramebufferTextureLayer in section 9.2.8 to break out validation by texture dimensionality and specify the correct constraint for cube map array textures, matching the OpenGL ES specification. • Remove description of component layout qualifiers affecting values assigned to attribute variables of matrix type in section 11.1.1 (Bug 15326). • Remove description of OpenGL Shading Language as not supporting multidimensional arrays 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 clause from section 13.3.2 specifying an error for ResumeTransformFeedback if the active program object has been relinked since transform feedback became active for the current transform feedback object (Bug 15414). • Cleanup the description of fragment coordinate wf in section 15.2.2 (Bug 5434). • Update the description of gl_SampleMaskIn in section 15.2.2 and gl_SampleMask in section 15.2.3 to refer to the maximum number of samples supported for any renderable internal format, and make corresponding OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

723

changes to the definitions of the MAX_*SAMPLES queries in table 23.69 (Bug 15792, based on ES bug 15122). • Fix work group size limit error for DispatchCompute in section 19 (Bug 15069). • Cleanup introduction to section 23, fix type of SYNC_FLAGS in table 23.50, and add CONTEXT_PROFILE_MASK in table 23.56 (Bug 5581). • Rewrite appendix D to refer to the Khronos Data Format Specification for the definition of formats (Bug 15255). Changes in the released Specification update of May 28, 2015 • Fix minor typos found during OpenGL ES spec updates. • Add language to section 4.1.2 tightening the behavior of SYNC_FLUSH_COMMANDS_BIT to only guarantee that commands are flushed up to and including the FenceSync command (Bug 11525). • Change description of MAP_COHERENT_BIT for buffer storage in section 6.2 so that barriers with CLIENT_MAPPED_BUFFER_BARRIER_BIT does not need to make CPU writes visible to the GPU in this case without an explicit flush (Bug 13578). • Add language for arrays of arrays to the generation of active resource lists in section 7.3.1 (Bug 13004). • Add language in sections 7.3.1, 7.3.1.1, and 11.1.2.1 allowing program interface queries to distinguish between entire arrays and array elements for transform feedback resources (Bug 12787). • Specify that shared and std140 layout uniform blocks and their members are always active in section 7.6 (Bug 10182). • Clarify the interaction of multisample write masks, shader side effects, and early per-fragment tests. Update the description of rasterization at the start of chapter 14 and in figure 14.1, and add language in sections 7.13.1 and 14.9. Move pixel ownership, scissor, and multisample fragment operations from section 17.3 into sections 14.9.1 and 14.9.2, note that stencil test, depth test, and occlusion query operations may be performed prior to fragment shading if requeseted by the shader, and update figure 17.1 to match. Move description of the application of gl_SampleMask from section 14.9.3 to OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

724

section 15.2.3. Split alpha to coverage from multisample fragment operations, leaving the former in section 17.3.1. Update section references in table 23.11 (Bug 12936). • Strike INVALID_ENUM error for invalid packed depth-stencil format/type combinations in section 8.4.4; it is covered by the generic special interpretation language in section 8.4.4.2. Also change the error for invalid integerfloat format/type mismatches to INVALID_OPERATION in section 8.4.4 (Bug 11167). • Add missing error conditions in section 8.5 when specifying texture images with specific compressed internal formats and there’s a texture dimensionality and/or target mismatch with the format (Bug 11239). • Added a definition of the effective internal format corresponding to base formats in section 8.5 (without going into detail on how it is determined, unlike the OpenGL ES Specification), and modified the descriptions of format matching for TextureView in section 8.18 and CopyImageSubData in section 18.3.2 to use the effective internal format where appropriate (Bug 13111). • Fix error for invalid target to GetTexSubImage* in section 8.6 to generate INVALID_ENUM instead of INVALID_VALUE (Bug 13563). • Add errors to *Tex*SubImage3D (section 8.6 and to GetTexture*Image and GetCompressedTextureImage (section 8.11.4) in the cases where the target or effective target of the texture is TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY, and the texture is not cube complete or cube array complete, respectively (Bug 13223). • Define initial state for sampler objects min filter and wrap modes in table 23.18, instead of deferring it to inapplicable language describing these parameters for texture objects as dependent on the texture type. Add a footnote noting this for CreateSamplers in section 8.2 (Bug 13499). • Add missing language from ARB_gpu_program5 describing comp argument to textureGather*, and textureGatherOffsets commands in section 8.14.2. This language should have been included in the OpenGL 4.0 Specification, but was left out by mistake (Bug 5910). • Specify that the values returned for RGBA texture sampling and fetches from incomplete textures are in floating-point format, in sections 8.14.2 and 11.1.3.5 (Bug 13525). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

725

• Drop redundant bullet point in the list of texture incompleteness conditions at the end of section 8.17 (Bug 12791). • Fix typo for DEPTH_STENCIL_TEXTURE_MODE (was improperly shown as DEPTH_TEXTURE_STENCIL_MODE in section 8.22) (Bug 13590). • Restore the initial value of internal format for buffer textures in the compatibility profile to LUMINANCE8 in section 8.22 and table 23.16, fixing an error introduced in earlier GL specs (Bug 10185). • Restore cube map arrays, which were accidentally removed, to the list of framebuffer attachment completeness conditions in section 9.4.1 (Bug 11201). • Rewrite the beginning of section 11.1.3.2 to clarify which operations are and are not performed during texelFetch operations (Bug 13833). • Allow errors resulting from user-defined tessellation control inputs and outputs, and tessellation evaluation inputs that are not specified to be arrays exactly gl_MaxPatchVertices long to be detected at compile time, as well as link time in sections 11.2.1.2, 11.2.1.2.3 and 11.2.3.3 (Bug 12185). • Clarify in section 14.3.1 that querying SAMPLE_POSITION returns a shading sample location, not a rasterization or coverage sample location (Bug 13484). • Align language with GL_OES_sample_shading in section 14.3.1.1. • Rearrange description of gl_SampleMaskIn in section 15.2.2 following variables whose use can kick off per-sample shading, and mention that (by symmetry with GL_OES_sample_variables). • Add explicit errors list for GetPointerv and GetString in section 22.2. • Fix description of MAX_TESS_CONTROL_TOTAL_OUTPUT_COMPONENTS in table 23.58 to reflect that it is the maximum for all outputs, not per-patch outputs (Bug 13765). Changes in the released Specification update of February 2, 2015 • Remove reference to nonexistent ARRAY_SIZE language in the discussion of array trimming for active variables in section 7.3.1 (Bug 13445).

OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

726

• Remove unimplementable TEXTURE_BINDING query from description of BindTexture in section 8.1, section 22.5, and table 23.13 (Bug 13278). Also removed in GL_ARB_direct_state_access specification and the specfile and headers in the Registry. Changes in the released Specification update of October 30, 2014 • Restore language describing non-sequentiality of resource locations for consecutive active array elements in section 7.3.1 (Bug 12318). • Restore description of MAX_UNIFORM_BLOCK_SIZE in section 7.6.2, which was lost in the restructuring and program interface query language introduced in OpenGL 4.3, but change behavior so that exceeding the limit will cause link failure, compared to may cause link failure in the 4.2 language (Bug 12897). • Add NEAREST_MIPMAP_NEAREST to the allowed filter modes for STENCIL_INDEX textures in section 8.17 (Bug 12791). • Modify behavior of primitive restart in section 10.3.6 to only apply to *DrawElements* commands for both forms of primitive restart, matching shipping drivers (Bug 12893). • Rearrange descriptions of DrawArraysOneInstance, DrawElementsOneInstance, and the actual DrawElements* commands in section 10.4 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 (Bug 12756). • Add description of conditions for which multisample texel fetch operations are undefined in section 11.1.3.3 (Bug 12255). • Modify the end of section 13.7.1 to remove language requiring that integer outputs be qualified as flat, since fragment shader inputs now control as described in the OpenGL Shading Language Specification (Bug 12623). • In the description of the source of the derived point size when program point size mode is enabled in section 14.4, remove the unreachable case where a tessellation evaluation shader is active, but no tessellation control shader or geometry shader is active (Bug 12865). • Add missing error for invalid target argument to InvalidateSubFramebuffer in section 17.4.4 (Bug 12727). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

727

• Restore description of TEXTURE_BUFFER_OFFSET_ALIGNMENT in table 23.55, UNIFORM_BUFFER_OFFSET_ALIGNMENT in table 23.63, and SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT in table 23.64 to refer to them as minimum require alignments, while the footnote in the caption continues to note that the numeric limits are the maximum allowed values (Bug 11962). • Increase the minimum values for MAX_GEOMETRY_UNIFORM_COMPONENTS and MAX_COMPUTE_UNIFORM_COMPONENTS from 512 to 1024 in tables 23.60 and 23.62, respectively. This makes the minimums consistent across all shader stages (Bug 12731). Changes in the released Specification update of September 19, 2014 • Clarify description of the initial state of query objects for CreateQueries and BeginQueryIndexed in section 4.2 (Public Bug 1213). • Add DSA-style GetQueryBufferObject* queries along with GetQueryObject* in section 4.2.3 (Public Bug 1214). • Add missing MapNamedBufferRange to description of MAP_COHERENT_BIT in section 6.2 (Public Bug 1208). • 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). • Clarify behavior of rendering to multiple framebuffer object attachments of different sizes in section 9.2 (Bug 10403). • Add FRONT and BACK as valid attachment names for Get*FramebufferAttachmentParameteriv in section 9.2.3 (Bug 12695). Changes in the released Specification of August 11, 2014: • Added new features as described in section I.1. • Cleanup many bugs and typos (public Bug 1186). • Clean up reference to texture borders in the core profile, where borders must always be zero width, simplifying equations (Bug 10507).

OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

728

• Replace all uses of “texel array” and “image array” with “texture image” or “image” as appropriate, to avoid confusion with array texture types. • Use “max.”, “min.”, and “no.” consistently throughout state tables instead of “maximum”, “minimum”, and “number”. • Make description of WebGL in section 1.3.5 consistent with the OpenGL ES 3.1 Specification. • Change required size of boolean from “1 or more” bits to exactly 8 bits in table 2.2, matching OpenGL ES 3.1 (Bug 11847). • Clarify in equations 2.3 and 2.4 that either truncation or rounding are acceptable when converting from floating-point to normalized fixed-point (Bug 9976). • Simplify and generalize language about object and name lifetimes in section 5.1.3 to make clear that an active query object behaves like an object attachment (Bug 12161). • Add missing validation error for invalid target parameters to MapBufferRange in section 6.3. • Simplify language in section 6.3.2 to remove confusing reference to “invalid reads, writes, or state changes” to mapped buffer objects (Bug 12300). • Modify error condition for invalid buffer binding target arguments to CopyBufferSubData in section 6.6 to INVALID_ENUM instead of INVALID_VALUE. • Make LinkProgram fail with an empty program object (one with no shaders attached) in the core profile only, in section 7.3 (Bug 12215). • Fix error condition for UseProgram in section 7.3 (Bug 12281). • Fix description of LINK_STATUS for GetProgramiv in section 7.14 (Bug 9698). • Change the variable representing compressed texture block size to blocksize in section 8.7, to avoid confusion with the texture border width bs used in many other places. • Clarify that filter state is ignored for multisample texture access in sections 8.8 and 11.1.3.3 (Bug 12171). OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

729

• Include MIRROR_CLAMP_TO_EDGE in the wrap modes not allowed for rectangle texture targets in the description of TexParameter* in section 8.10 (public Bug 1186). • Make the minification vs. magnification switch-over point always zero in sections 8.14.3 and 8.15 (Bug 9997). • Strike irrelevant clause from description of error condition for numlayers argument to TextureView in section 8.18 (Bug 11891). • Add an error for TextureView in section 8.18 if the computed number of levels or layers for the new texture view is not positive (Bug 12256). • Update errors for TexStorage*DMultisample in section 8.19 to include an appropriate subset of the generic errors for Tex*Storage*D commands defined in section 8.19, and remove redundant errors from the latter commands (Bug 11937). • Allow GetFramebufferParameteriv and GetNamedFramebufferParameteriv in section 9.2.3 to be used to query the framebuffer-dependent state in table 23.73, including queries for the default framebuffers (Bug 12360). • Add missing error for invalid attachment arguments of GetFramebufferAttachmentParameteriv and GetNamedFramebufferAttachmentParameteriv in section 9.2.3. • Fix error condition for GetFramebufferAttachmentParameteriv in section 9.2.3 (Bug 12180). • 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, 14.3.1, 18.2.2, and 18.3.1 (Bug 12360). • Add missing cube map array textures to list of layered texture types for FramebufferTexture in section 9.2.8 (Bug 12336). • Restore missing error condition for CheckFramebufferStatus target argument in section 9.4.2. • Clarify in section 10.4 that DrawArrays transfers no elements when count is zero (Bug 10015).

OpenGL 4.6 (Core Profile) - May 5, 2022

I.3. CHANGE LOG FOR RELEASED SPECIFICATIONS

730

• Specify the values of gl_VertexID in the descriptions of drawing pseudocommands DrawArraysOneInstance and DrawElementsOneInstance in section 10.4 (Bug 12202). • Remove count from the list of parameters validated in the pseudocode for MultiDrawElements in section 10.4, since it’s an array, not a value (Bug 7004). • Clarify that MultiDrawArraysIndirect and MultiDrawElementsIndirect in section 10.4 share the same errors as the comparable non-MultiDraw* commands, and throughout the spec, clarify relevant pseudocode examples with “(assuming no errors are generated)” (Bug 12351). • Add description of ELEMENT_ARRAY_BUFFER_BINDING. to section 10.5 (Bug 11042). • Fix list of required state in section 10.6 and table 23.4 (Bug 10283). • Clarify description of BindAttribLocation in section 11.1.1 (Bug 12186). • 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). • Add language for robust buffer access in section 11.1.3.12 expanding the definition of what can be returned from out-of-bounds shader reads within a buffer object (Bugs 10826, 12104). • Modify descriptions of tessellation in sections 11.2.2, 11.2.2.1, and 11.2.2.2 to clarify that tessellation may produce multiple vertices with the same gl_TessCoord values under some conditions (Bug 11979). • Fixed DebugMessageInsert in section 20.5 so error for invalid severity arguments is INVALID_ENUM, matching ARB_debug_output, instead of INVALID_VALUE. • Change title of section 22.2 to more accurately describe the contents (Bug 12352). • Change query commands for buffer storage state in table 23.6 to GetBufferParameteriv (Bug 12307). • Change the description column for TEXTURE_BUFFER_OFFSET_ALIGNMENT in table 23.55, UNIFORM_BUFFER_OFFSET_ALIGNMENT in table 23.63, and SHADER_STORAGE_BUFFER_OFFSET_ALIGNMENT in OpenGL 4.6 (Core Profile) - May 5, 2022

I.4. CREDITS

731

table 23.64 to refer to them as maximum, not minimum, values, consistently with the footnote in the caption (Bug 11962). • Clean up caption for table 23.73 to accurately describe which FBO binding is used for different queries (Bug 12360).

I.4

Credits

OpenGL 4.5 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ARB Working Group during the development of OpenGL 4.5, 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, including specific functionality developed in the form of new ARB extensions together with OpenGL 4.5. In addition, many people participated in developing earlier vendor and EXT extensions on which the OpenGL 4.5 functionality is based in part; those individuals are listed in the respective extension specifications in the OpenGL Registry. Acorn Pooley, NVIDIA Alex Corscadden, VMware Anton Staaf, Google Aras Pranckevicius, Unity Barthold Lichtenbelt, NVIDIA (Chair, Khronos OpenGL ARB Working Group) Bill Licea-Kane, Qualcomm Brian Paul, VMware Inc. (GL_ARB_conditional_render_inverted, GL_ARB_cull_distance, GL_ARB_get_texture_sub_image) Bruce Merry, ARM Chris Marrin, Apple Christoph Kubisch, NVIDIA Christophe Riccio, Unity (GL_ARB_direct_state_access) Daniel Koch, NVIDIA Daniel Rakos, AMD Dominik Witczak, Samsung Electronics (reference pages) Evan Hart, NVIDIA Graham Sellers, AMD (GL_ARB_direct_state_access) Greg Roth, NVIDIA Ian Romanick, Intel Jaroslaw Lipowski (extensive proofreading in public bug 1186) Jason Mitchell, Valve OpenGL 4.6 (Core Profile) - May 5, 2022

I.5. ACKNOWLEDGEMENTS

732

Jeannot Breton, NVIDIA Jeff Bolz, NVIDIA (GL_ARB_texture_barrier) Jesse Hall, Google John Kessenich, Independent (OpenGL Shading Language Specification Editor, GL_ARB_derivative_control) John McDonald, NVIDIA Jon Leech, Independent (OpenGL API Specification Editor, reference pages) Kenneth Russell, Google Kent Miller, Apple Lingjun (Frank) Chen, Qualcomm Mark Callow, ArtSpark Mark Kilgard, NVIDIA (GL_ARB_clip_control, GL_ARB_robustness) Mathias Schott, NVIDIA (GL_ARB_shader_texture_image_samples) Nick Penwarden, Epic Games Pat Brown, NVIDIA Patrick Doane, Blizzard Piers Daniell, NVIDIA (GL_ARB_shader_texture_image_samples, GL_ARB_ES3_1_compatibility) Robert Simpson, Qualcomm Simon Bennett, VMware Stefan Dösinger, CodeWeavers Tim Foley, Intel Timo Suoranta, Broadcom Timothy Lottes, Epic Games Tom Olson, ARM Tristan Lorach, NVIDIA Vladimir Vukicevic, Mozilla

I.5

Acknowledgements

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix J

Version 4.6 OpenGL version 4.6, released on July 31, 2017, is the eighteenth revision since the original version 1.0. Separate versions of the OpenGL 4.6 Specification exist for the core profile and compatibility profile described in appendix E, respectively subtitled the “Core Profile” and the “Compatibility Profile”. This document describes the Core Profile. An OpenGL 4.6 implementation must be able to create a context supporting the core profile, and may also be able to create a context supporting the compatibility profile. Material specific to the compatibility profile specification is marked in a distinct color to clearly call out differences between the two profiles. The OpenGL 4.6 compatibility and core profiles are upward compatible with the OpenGL 4.5 compatibility and core profiles, respectively (see appendix I). Following are brief descriptions of changes and additions to OpenGL 4.6. Descriptions of changes and additions in versions of OpenGL prior to 4.2 are omitted in this Specification, but may be found in the OpenGL 3.0 Specification (for features in versions 1.0 - 3.0, inclusive) and the OpenGL 4.2 Specification (for features in versions 3.1 - 4.1, inclusive). These Specifications are available in the OpenGL Registry.

J.1

New Features

New features in OpenGL 4.6, including the extension or extensions if any on which they were based, include: • GL_ARB_indirect_parameters • GL_ARB_pipeline_statistics_query 733

J.2. DEPRECATION MODEL

734

• GL_ARB_polygon_offset_clamp • GL_KHR_no_error • GL_ARB_shader_atomic_counter_ops (OpenGL Shading Language Only) • GL_ARB_shader_draw_parameters • GL_ARB_shader_group_vote (OpenGL Shading Language Only) • GL_ARB_gl_spirv • GL_ARB_spirv_extensions • GL_ARB_texture_filter_anisotropic • GL_ARB_transform_feedback_overflow_query

J.2

Deprecation Model

No new features are deprecated, and no previously deprecated features are reintroduced by the OpenGL 4.6 core profile. Features deprecated by OpenGL 4.5 remain deprecated, but have not yet been removed.

J.3

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 server; and (Bug 16315) refers to a private Bugzilla issue on Khronos’ internal server.

J.4

Change Log for Released Specifications

Changes in the released Specification of May 5, 2022 • Require shader interface matching to include OpenGL ES precision qualifiers where relevant in section 7.4.1 (gitlab #157).

OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

735

• Change texture image error for mismatching format and base internal format in section 8.5.3 so that if either is STENCIL_INDEX, both must be (github #86). • Disambiguate computation of number of rows to skip when fetching successive compressed image blocks from memory in section 8.7 (github #62). • 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 8.19 when target is TEXTURE_CUBE_MAP_ARRAY and depth is not a multiple of 6 (gitlab #135). • Add alternate order for triangles generated by triangle strips with adjacency in table 10.2 (gitlab #150). • Remove reference to “state for vertex coordinates or generic attribute zero“ from compatibility profile language in section 10.2.4 (gitlab #123). • Add CreateVertexArrays to commands allowed to create an array name for BindVertexArray in section 10.3.1 (github #68). • Fix typo in pseudocode equivalent of DrawElementsInstanced in section 10.4 by removing extra 0 argument (github #69). • Update computation of element array buffer access for DrawElementsOneInstance in section 10.4 to offset in units of the size of the type being accessed (github #85). • Add cube maps to allowed types of texture names for FramebufferTextureLayer in section 9.2.8 (gitlab #134). • Change the minimum value of MAX_FRAGMENT_INTERPOLATION_OFFSET to +0.5 − 1U LP in table 23.70 (gitlab #149). • Add SPIR-V environment validation rule for atomics in section C.4, matching a change in ARB_gl_spirv (gitlab #132). Changes in the released Specification of October 22, 2019 • Change the name of formal parameter bufSize to count for GetSynciv, GetActiveSubroutineUniformName, GetActiveSubroutineName, GetInternalformativ, GetInternalformati64v, and GetProgramResourceiv (github #52 / #285). OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

736

• Remove query objects from the list of shareable container objects in section 5 (github #60). • Specify that linking OpenGL and OpenGL ES shader objects is allowed in the description of LinkProgram in section 7.3 (gitlab #98). • 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). • 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 7.13 (github #14). • Clarify memory barrier requirements in section 7.13.1 (gitlab #95). • Add a “Copyable” column to table 8.14 specifying if formats may be used with non-Compressed texture image commands, and add errors to the TexImage*D and CopyTexImage*D commands (and thus to the *SubImage* commands which have the same constraints) referring to this table (github #11). • 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). • Add VERTEX_ATTRIB_ARRAY_BUFFER_BINDING, VERTEX_BINDING_STRIDE, VERTEX_BINDING_DIVISOR, and VERTEX_BINDING_BUFFER to the state queriable with GetVertexArrayIndexediv in section 10.5 (github #42 / gitlab #108). • Fix typo in section 11.2.2.2 (gitlab #113). • Update the behavior of point clipping in section 13.7 to allow pop-free clipping, for consistency with OpenGL ES (github #51).

OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

737

• Loosen the line interpolation requirements in section 14.5.2.2 to properly handle gl_FragCoord and allow diagonal interpolation of input attributes, bring the specification in line with Vulkan and shipping hardware (gitlab #116). • Change default values of VERTEX_BINDING_DIVISOR and VERTEX_BINDING_BUFFER to zero in table 23.4 (github #44). • Specify that the image format of OpImageWrite must not be unknown unless the StorageImageWriteWithoutFormat capability was declared, in section C.4 (gitlab #112). • Specify that user interface variables must not contain a value of type OpTypeBool in section C.4 (gitlab #111). Changes in the released Specification of February 2, 2019 • Fix minor a/an grammar issues (gitlab #101). • 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). OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

738

• Make clarifications on uniform and shader storage buffer object layout requirements in section 7.6.2.3 (gitlab #92). • Add TEXTURE_CUBE_MAP or TEXTURE_CUBE_MAP_ARRAY as valid targets for CompressedTexImage3D in section 8.7 (Bug 14713) • Update the texture level-of-detail equation to better approximate ellipse major and minor axes for anisotropic sampling in section 8.14, and use the same terminology as the isotropic sampling case (gitlab #5). 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). • Add a footnote to section 11.1.3.11 clarifying that unlike OpenGL ES, OpenGL does not require validation to detect mismatched interfaces between shader stages (Bug 15331). • Modified BeginTransformFeedback language about simultaneous use of buffers for transform feedback and other operations in section 13.3.2 to ensure this behavior is undefined (gitlab #83). • Specify std430 as a base layout for compute shader uniform storage requirements in section 19.1 (github #40). Changes in the released Specification of May 14, 2018 • Clarify undefined GL behavior with no current context in section 2.1 (Bug 11522). • 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, and include texture views as ways to create feedback loops (Bug 11934). • 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). OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

739

• Clarify for Clear*BufferSubData in section 6.2.1 that there are no compatibility restrictions between the format and type of the specified clear color, and the internal format to be stored in the buffer (Bug 16232). • Clarify error for glSpecializeShader in section 7.2.1, and add a new error if shader types mismatch (github #16). • Add

descriptions of COMPATIBLE_SUBROUTINES and NUM_COMPATIBLE_SUBROUTINES to the GetProgramResourceiv queries

described in section 7.3.1.1 (gitlab #40). • 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). • Require Binding for uniform and shader storage blocks in section 7.4.2 and add whitelist for accepted storage classes in section C.4 (github #5, gitlab #55) • Specify in section 7.7.1 that the value of UNIFORM_ARRAY_STRIDE) for atomic_uint arrays is always 4 (gitlab #31). • Added description of storage block activeness in section 7.8, based on that of uniform blocks (Bug 13745, gitlab #63). • Add new section 7.9 defining invocation groups, as used in the OpenGL Shading Language Specification (github #15). • Add MultiDraw*IndirectCount and PARAMETER_BUFFER to the list of commands whose buffer bindings can be affected by COMMAND_BARRIER_BIT in the discussion of MemoryBarrier in section 7.13.2 (github #17). • Use the term “work group size” consistently in section 7.14 and chapter 19, instead of a mix of similar terms usually starting with “local” (Bug 11723). • Remove error in section 8.6 for the *TexSubImage3D commands when target is TEXTURE_CUBE_MAP_ARRAY, and the target image is not cube array complete. This was inappropriately added and should only exist for the *TextureSubImage3D commands (Bug 13223). • Fix typo (missing comma) in prototype for TextureParameterI* in section 8.10 (Public Bug 1432). OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

740

• Drop reference in section 8.14.1 to the no-longer-used minification switchover value c and replace with 0.0 (github #34). • Specify in section 8.14.4 that mimap generation has no effect when any levelbase dimension is zero (gitlab #72). • Add a requirement to the Generate*Mipmap commands in section 8.14.4 that any synchronization required before performing mipmap reduction will be done within the commands themselves (Public Bug 1429). • Add error for Tex*Storage3D in section 8.19 when target is TEXTURE_CUBE_MAP_ARRAY and depth is not a multiple of 6 (Public Bug 1439). • Clarify meaning of FULL_SUPPORT for FRAMEBUFFER_RENDERABLE and FRAMEBUFFER_RENDERABLE_LAYERED in section 9.4.3 (Public Bug 689). • Modify the description of indirect commands in buffer objects in section 10.3.11 to say that indirect parameters are sourced from the indirect buffer in the core profile, and may be sourced from it in the compatibility profile (Public Bug 1082). • Remove assertions that OpenGL Shading Language does not support multidimensional arrays in sections 11.2.1.2.3 and 11.3.4.4 (gitlab #62). • Specify that interpolation of vertex outputs is controlled by fragment shader input qualifiers at the beginning of section 11.1.3.10 (github #10). • Move definition of upstream shader for geometry shaders from section 11.3.4.4 to section 11.3.3 (Bug 14158). • 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 13.2 and in sections 13.3, 13.3.2, 13.6, 13.7, 13.7.1, and 14.4 (github #9). • Add a footnote to the description of BindTransformFeedback in section 13.3.1 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 23.48 to the context transformation state table 23.7 (gitlab #66). OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

741

• Clarify near the end of section 13.3.2 that a buffer object must be bound for transform feedback only if the buffer is active (gitlab #38). • Remove duplicate definition of “fragment” near the beginning of section 14 (Public Bug 1388). • Ensure discussion of polygon vertex lighting in section 14.6.4 appears in the compatibility profile (github #18). • Simplify description of per-sample shading in section 15.2.2 and clarify the requirements on the interaction between gl_SampleMaskIn and MinSampleShading (gitlab #45). • Clarify occlusion query description in section 17.3.5 to refer to fragments that are not discarded by earlier stages, rather than fragments that pass the depth test (github #19). • Don’t require that textures passed to CopyImageSubData in section 18.3.2 be mipmap complete, by ignoring format-based completeness rules (Bug 16224). • Fix error for DispatchComputeIndirect in section 19 to refer to DISPATCH_INDIRECT_BUFFER instead of DRAW_INDIRECT_BUFFER (gitlab #41). • Clarify comparison of internal format queries to GetTexLevelParameter* in section 22.3.2, so that internal formats not supported by that command are still valid queries (gitlab #65). • Specify in section C.4 that OpImageQuerySizeLod and OpImageQueryLevels only work with Sampled images (gitlab #73). • Fix typo in table D.2 (Public Bug 1920). Changes in the released Specification of July 30, 2017: • Added new features as described in section J.1. • Update cover image to OpenGL 4.6 (gitlab #16). • Numerous minor typographical cleanups (gitlab #25), and use “level-ofdetail” consistently, replacing occasional usage of “level of detail”. These are not marked in color. OpenGL 4.6 (Core Profile) - May 5, 2022

J.4. CHANGE LOG FOR RELEASED SPECIFICATIONS

742

• Direct bug reports to API and GLSL Github repositories, instead of Bugzilla, in section 1.4 (gitlab #2). • Add SPIR-V to the Related APIs in section 1.3.4, and refer to this section from the SPIR-V Execution Environment appendix in section C (gitlab #2). • Added clarifying language to section 2.3.1 specifying when INVALID_OPERATION errors are generated vs. INVALID_ENUM (gitlab #7). • Change reporting of OUT_OF_MEMORY errors in section 2.3.1.1 (gitlab #14). • Add the pipeline statistics query targets to those for which the index parameter of BeginQueryIndexed and EndQueryIndexed in section 4.2.2, and of GetQueryIndexediv in section 4.2.3, must be zero (gitlab #26). • Add new subsections 4.2.1 and 4.2.2, with existing content. Refer to the table of query object targets in section 4.2.1 for error purposes in BeginQueryIndexed and related commands in the descriptions of CreateQueries, BeginQueryIndexed, EndQueryIndexed, and GetQueryIndexediv (gitlab #18). • Fix reference to the BuiltIn decoration (not variable) in section 7.4.2 (gitlab #32). • Minor cleanup of SPIR-V interface language in section 7.4.2 and table C.6, and clarify OpTypeRuntimeArray restriction in section C.4 (gitlab #20). • Clarify RECTANGLE_TEXTURE and compressed texture interactions in errors for *Tex*SubImage2D in section 8.6, CompressedTexImage2D in section 8.7, and Tex*Storage2D in section 8.19 (gitlab #10). • Change errors for invalid effective targets of TextureBufferRange in section 8.9, TextureParameter* in section 8.10, GetTextureParameter* in section 8.11.2, GetTextureLevelParameter* in section 8.11.3, and GetTextureImage in section 8.11.4 from INVALID_VALUE to INVALID_OPERATION (gitlab #22). • Change some errors for GetTextureSubImage in section 8.18 and InvalidateTexSubImage in section 8.20 from INVALID_VALUE to INVALID_OPERATION (gitlab #17). • Tweak language for TextureStorage1D errors in section 8.19 to update generated error for the wrong effective target (gitlab #8).

OpenGL 4.6 (Core Profile) - May 5, 2022

J.5. CREDITS

743

• Change level limit for *FramebufferTexture*D and *FramebufferTextureLayer in section 9.2.8 from the value of TEXTURE_IMMUTABLE_LEVELS to that of TEXTURE_VIEW_NUM_LEVELS, to account for texture views (gitlab #19). • Remove incorrectly-applied language for gl_DrawID that had been applied to DrawArraysInstancedBaseInstance in section 10.4 (gitlab #21). • Add TEXTURE_MAX_ANISOTROPY parameter to sampler state in table 23.18, as well as texture object state (gitlab #15). • Change required state for QUERY_COUNTER_BITS in table 23.69 and CURRENT_QUERY in table 23.74 from 18 × E to n × E to avoid having to track the total number of otherwise unrelated pieces of queryable state (gitlab #11). • Refer to the tables in section C instead of saying “above” or “below”, since these tables are floats that may shift their position in the document as other material is added and removed, and clean up related language in section C.6 (gitlab #2).

J.5

Credits

OpenGL 4.6 is the result of the contributions of many people and companies. Members of the Khronos OpenGL ARB Working Group during the development of OpenGL 4.6, 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, including specific functionality developed in the form of new ARB extensions together with OpenGL 4.6. In addition, many people participated in developing earlier vendor and EXT extensions on which the OpenGL 4.6 functionality is based in part; those individuals are listed in the respective extension specifications in the OpenGL Registry. Brian

Paul,

VMware

(ARB_pipeline_statistics_query, ARB_transform_feedback_overflow_query) Christophe Riccio, OpenGL Garden / Unity Daniel Kartch, NVIDIA Daniel Koch, NVIDIA (ARB_gl_spirv, ARB_spirv_extensions) Daniel Rakos, AMD (ARB_shader_atomic_counter_ops) David Blythe, SGI OpenGL 4.6 (Core Profile) - May 5, 2022

J.6. ACKNOWLEDGEMENTS

744

Dominik Witczak, AMD Eric Lengyel, Terathon Software Graham Sellers, AMD (ARB_indirect_parameters, ARB_shader_draw_parameters) James Jones, NVIDIA Jan-Harald Fredriksen, ARM John Kessenich, Google (ARB_gl_spirv) Jon Leech, Independent (OpenGL API Specification Editor) Mark Kilgard, NVIDIA (EXT_texture_filter_anisotropic on which the ARB functionality was based) Maurice Ribble, Qualcomm (KHR_no_error) Pat Brown, NVIDIA (ARB_shader_group_vote) Piers Daniell, NVIDIA (Chair, Khronos OpenGL Working Group; ARB_polygon_offset_clamp, ARB_texture_filter_anisotropic) Steve Hill, Broadcom Tobias Hector, Imagination Technologies Walt Donovan, NVIDIA

J.6

Acknowledgements

The ARB gratefully acknowledges technical support by James Riordon, webmaster of Khronos.org and OpenGL.org. The “pipeline metro” cover image was created by Dominic Agoro-Ombaka.

OpenGL 4.6 (Core Profile) - May 5, 2022

Appendix K

OpenGL Registry, Header Files, and ARB Extensions K.1

OpenGL Registry

Many extensions to the OpenGL API have been defined by vendors, groups of vendors, and the OpenGL ARB. In order not to compromise the readability of the OpenGL Specification, such extensions are not integrated into the core language; instead, they are made available online in the OpenGL Registry, together with extensions to window system binding APIs, such as GLX and WGL, and with specifications for OpenGL, GLX, and related APIs. Extensions are documented as changes to a particular version of the Specification. The Registry is available on the World Wide Web at URL http://www.opengl.org/registry/

K.2

Header Files

Historically, C and C++ source code calling OpenGL was to #include a single header file, <GL/gl.h>. In addition to the core OpenGL API, the APIs for all extensions provided by an implementation were defined in this header. When platforms became common where the OpenGL SDK (library and header files) were not necessarily obtained from the same source as the OpenGL driver, such as Microsoft Windows and Linux, <GL/gl.h> could not always be kept in sync with new core API versions and extensions supported by drivers. At this time the OpenGL ARB defined a new header, <GL/glext.h>, which could be obtained directly from the OpenGL Registry (see section K.1). The combination 745

K.3. ARB AND KHRONOS EXTENSIONS

746

of <GL/gl.h> and <GL/glext.h> always defines all APIs for all profiles of the latest OpenGL version, as well as for all extensions defined in the Registry. <GL/glcorearb.h> defines APIs for the core profile of OpenGL, together with ARB extensions compatible with the core profile. It does not include APIs for features only in the compatibility profile or for other extensions. There is currently no Khronos-supported mechanism for using vendor extensions together with <GL/glcorearb.h>, due to lack of demand and lack of knowledge on which vendor extensions are compatible with the core profile. In the future, this may be addressed by a hypothetical header <GL/glcoreext.h> which would define APIs for additional EXT and vendor extensions compatible with the core profile, but not defined in <GL/glcorearb.h>. Most older extensions are not compatible with the core profile. Applications using the compatibility profile (see appendix E) should #include the traditional <GL/gl.h> and <GL/glext.h> headers. Applications using the core profile, and which do not need to use vendor extensions, may instead #include the <GL/glcorearb.h> header. By using <GL/glcorearb.h>, instead of the legacy <GL/gl.h> and <GL/glext.h>, newly developed applications are given increased protection against accidentally using a legacy feature that has been removed from the core profile, and against using a less portable EXT or vendor extension. This can assist in developing applications on a GL implementation that supports the compatibility profile when the application is also intended to run on other platforms supporting only the core profile. Developers should always be able to download <GL/glcorearb.h> from the Registry, with this headers replacing, or being used in place of older versions that may be provided by a platform SDK.

K.3

ARB and Khronos Extensions

OpenGL extensions that have been approved by the Khronos OpenGL Architectural Review Board Working Group (ARB), or jointly approved by the ARB and the Khronos OpenGL ES Working Group (KHR), are summarized in this section. ARB and KHR extensions are not required to be supported by a conformant OpenGL implementation, but are expected to be widely available; they define functionality that is likely to move into the required feature set in a future revision of the specification.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

K.3.1

747

Naming Conventions

To distinguish ARB and KHR extensions from core OpenGL features and from vendor-specific extensions, the following naming conventions are used: • A unique name string of the form ”GL_ARB_name” or ”GL_KHR_name” is associated with each extension. If the extension is supported by an implementation, this string will be among the EXTENSIONS strings returned by GetStringi, as described in section 22.2. • All functions defined by the extension will have names of the form FunctionARB or FunctionKHR, respectively. • All enumerants defined by the extension will have names of the form NAME_ARB. or NAME_KHR, respectively. • In addition to OpenGL extensions, there are also ARB extensions to the related GLX and WGL APIs. Such extensions have name strings prefixed by "GLX_" and "WGL_" respectively. Not all GLX and WGL ARB extensions are described here, but all such extensions are included in the registry.

K.3.2

Promoting Extensions to Core Features

Extensions can be promoted to required core features in later revisions of OpenGL. When this occurs, the extension specifications are merged into the core specification. Functions and enumerants that are part of such promoted extensions will have the ARB, KHR, EXT, or vendor affix removed. Implementations of such later revisions should continue to export the name strings of promoted extensions in the EXTENSIONS strings and continue to support the affixed versions of functions and enumerants as a transition aid. For descriptions of extensions promoted to core features in OpenGL 1.3 and beyond, see the corresponding version of the OpenGL specification, or the descriptions of that version in version-specific appendices to later versions of the specification.

K.3.3

Extension Summaries

This section describes each ARB and KHR extension briefly. In most cases, the functionality of these extensions also was added to a version of the OpenGL Specification, and in these cases only the extension string is described, together with the corresponding OpenGL version.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.1

748

Multitexture

The name string for multitexture is GL_ARB_multitexture. It was promoted to a core feature in OpenGL 1.3. K.3.3.2

Transpose Matrix

The name string for transpose matrix is GL_ARB_transpose_matrix. It was promoted to a core feature in OpenGL 1.3. K.3.3.3

Multisample

The name string for multisample is GL_ARB_multisample. It was promoted to a core feature in OpenGL 1.3. K.3.3.4

Texture Add Environment Mode

The name string for texture add mode is GL_ARB_texture_env_add. It was promoted to a core feature in OpenGL 1.3. K.3.3.5

Cube Map Textures

The name string for cube mapping is GL_ARB_texture_cube_map. It was promoted to a core feature in OpenGL 1.3. K.3.3.6

Compressed Textures

The name string for compressed textures is GL_ARB_texture_compression. It was promoted to a core feature in OpenGL 1.3. K.3.3.7

Texture Border Clamp

The name string for texture border clamp is GL_ARB_texture_border_clamp. It was promoted to a core feature in OpenGL 1.3. K.3.3.8

Point Parameters

The name string for point parameters is GL_ARB_point_parameters. It was promoted to a core features in OpenGL 1.4.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.9

749

Vertex Blend

Vertex blending replaces the single model-view transformation with multiple vertex units. Each unit has its own transform matrix and an associated current weight. Vertices are transformed by all the enabled units, scaled by their respective weights, and summed to create the eye-space vertex. Normals are similarly transformed by the inverse transpose of the model-view matrices. The name string for vertex blend is GL_ARB_vertex_blend. K.3.3.10

Matrix Palette

Matrix palette extends vertex blending to include a palette of model-view matrices. Each vertex may be transformed by a different set of matrices chosen from the palette. The name string for matrix palette is GL_ARB_matrix_palette. K.3.3.11

Texture Combine Environment Mode

The name string for texture combine mode is GL_ARB_texture_env_combine. It was promoted to a core feature in OpenGL 1.3. K.3.3.12

Texture Crossbar Environment Mode

The name string for texture crossbar is GL_ARB_texture_env_crossbar. It was promoted to a core features in OpenGL 1.4. K.3.3.13

Texture Dot3 Environment Mode

The name string for DOT3 is GL_ARB_texture_env_dot3. It was promoted to a core feature in OpenGL 1.3. K.3.3.14

Texture Mirrored Repeat

The name string for texture mirrored repeat is GL_ARB_texture_mirrored_repeat. It was promoted to a core feature in OpenGL 1.4. K.3.3.15

Depth Texture

The name string for depth texture is GL_ARB_depth_texture. It was promoted to a core feature in OpenGL 1.4.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.16

750

Shadow

The name string for shadow is GL_ARB_shadow. It was promoted to a core feature in OpenGL 1.4. K.3.3.17

Shadow Ambient

Shadow ambient extends the basic image-based shadow functionality by allowing a texture value specified by the TEXTURE_COMPARE_FAIL_VALUE_ARB texture parameter to be returned when the texture comparison fails. This may be used for ambient lighting of shadowed fragments and other advanced lighting effects. The name string for shadow ambient is GL_ARB_shadow_ambient. K.3.3.18

Window Raster Position

The name string for window raster position is GL_ARB_window_pos. It was promoted to a core feature in OpenGL 1.4. K.3.3.19

Low-Level Vertex Programming

Application-defined vertex programs may be specified in a new low-level programming language, replacing the standard fixed-function vertex, transformation, lighting, and texture coordinate generation pipeline. Vertex programs enable many new effects and are an important first step towards future graphics pipelines that will be fully programmable in an unrestricted, high-level shading language. The name string for low-level vertex programming is GL_ARB_vertex_program. K.3.3.20

Low-Level Fragment Programming

Application-defined fragment programs may be specified in the same low-level language as GL_ARB_vertex_program, replacing the standard fixed-function vertex texturing, fog, and color sum operations. The name string for low-level fragment programming is GL_ARB_fragment_program. K.3.3.21

Buffer Objects

The name string for buffer objects is GL_ARB_vertex_buffer_object. It was promoted to a core feature in OpenGL 1.5.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.22

751

Occlusion Queries

The name string for occlusion queries is GL_ARB_occlusion_query. It was promoted to a core feature in OpenGL 1.5. K.3.3.23

Shader Objects

The name string for shader objects is GL_ARB_shader_objects. It was promoted to a core feature in OpenGL 2.0. K.3.3.24

High-Level Vertex Programming

The name string for high-level vertex programming is GL_ARB_vertex_shader. It was promoted to a core feature in OpenGL 2.0. K.3.3.25

High-Level Fragment Programming

The name string for high-level fragment programming is GL_ARB_fragment_shader. It was promoted to a core feature in OpenGL 2.0. K.3.3.26

OpenGL Shading Language

The name string for the OpenGL Shading Language is GL_ARB_shading_language_100. The presence of this extension string indicates that programs written in version 1 of the Shading Language are accepted by OpenGL. It was promoted to a core feature in OpenGL 2.0. K.3.3.27

Non-Power-Of-Two Textures

The name string for non-power-of-two textures is GL_ARB_texture_non_power_of_two. It was promoted to a core feature in OpenGL 2.0. K.3.3.28

Point Sprites

The name string for point sprites is GL_ARB_point_sprite. It was promoted to a core feature in OpenGL 2.0. K.3.3.29

Fragment Program Shadow

Fragment program shadow extends low-level fragment programs defined with GL_ARB_fragment_program to add shadow 1D, 2D, and 3D texture targets, and remove the interaction with GL_ARB_shadow. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

752

The name string for fragment program shadow is GL_ARB_fragment_program_shadow.

K.3.3.30

Multiple Render Targets

The name string for multiple render targets is GL_ARB_draw_buffers. It was promoted to a core feature in OpenGL 2.0. K.3.3.31

Rectangle Textures

Rectangle textures define a new texture target TEXTURE_RECTANGLE_ARB that supports 2D textures without requiring power-of-two dimensions. Rectangle textures are useful for storing video images that do not have power-of-two sizes (POTS). Resampling artifacts are avoided and less texture memory may be required. They are also useful for shadow maps and window-space texturing. These textures are accessed by dimension-dependent (aka non-normalized) texture coordinates. Rectangle textures are a restricted version of non-power-of-two textures. The differences are that rectangle textures are supported only for 2D; they require a new texture target; and the new target uses non-normalized texture coordinates. The name string for texture rectangles is GL_ARB_texture_rectangle. It was promoted to a core feature in OpenGL 3.1. K.3.3.32

Floating-Point Color Buffers

Floating-point color buffers can represent values outside the normal [0, 1] range of colors in the fixed-function OpenGL pipeline. This group of related extensions enables controlling clamping of vertex colors, fragment colors throughout the pipeline, and pixel data read back to client memory, and also includes WGL and GLX extensions for creating framebuffers with floating-point color components (referred to in GLX as framebuffer configurations, and in WGL as pixel formats). The name strings for floating-point color buffers are GL_ARB_color_buffer_float, GLX_ARB_fbconfig_float, and WGL_ARB_pixel_format_float. GL_ARB_color_buffer_float was promoted to a core feature in OpenGL 3.0. K.3.3.33

Half-Precision Floating-Point

This extension defines the representation of a 16-bit floating-point data format, and a corresponding type argument which may be used to specify and read back pixel and texture images stored in this format in client memory. Half-precision floats are OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

753

smaller than full precision floats, but provide a larger dynamic range than similarly sized (short) data types. The name string for half-precision floating-point is GL_ARB_half_float_pixel. It was promoted to a core feature in OpenGL 3.0. K.3.3.34

Floating-Point Textures

Floating-point textures stored in both 32- and 16-bit formats may be defined using new internalformat arguments to commands which specify and read back texture images. The name string for floating-point textures is GL_ARB_texture_float. It was promoted to a core feature in OpenGL 3.0. K.3.3.35

Pixel Buffer Objects

The buffer object interface is expanded by adding two new binding targets for buffer objects, the pixel pack and unpack buffers. This permits buffer objects to be used to store pixel data as well as vertex array data. Pixel-drawing and -reading commands using data in pixel buffer objects may operate at greatly improved performance compared to data in client memory. The name string for pixel buffer objects is GL_ARB_pixel_buffer_object. It was promoted to a core feature in OpenGL 2.1. K.3.3.36

Floating-Point Depth Buffers

The name string for floating-point depth buffers is GL_ARB_depth_buffer_float. This extension is equivalent to new core functionality introduced in OpenGL 3.0, based on the earlier GL_NV_depth_buffer_float extension, and is provided to enable this functionality in older drivers. K.3.3.37

Instanced Rendering

The name string for instanced rendering is GL_ARB_draw_instanced. It was promoted to a core feature in OpenGL 3.1. K.3.3.38

Framebuffer Objects

The name string for framebuffer objects is GL_ARB_framebuffer_object. This extension is equivalent to new core functionality introduced in OpenGL 3.0, based

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

754

on the earlier GL_EXT_framebuffer_object, GL_EXT_framebuffer_multisample, and GL_EXT_framebuffer_blit extensions, and is provided to enable this functionality in older drivers. K.3.3.39

sRGB Framebuffers

The name string for sRGB framebuffers is GL_ARB_framebuffer_sRGB. It was promoted to a core feature in OpenGL 3.0. To create sRGB format surfaces for use on display devices, an additional pixel format (config) attribute is required in the window system integration layer. The name strings for the GLX and WGL sRGB pixel format interfaces are GLX_ARB_framebuffer_sRGB and WGL_ARB_framebuffer_sRGB respectively. K.3.3.40

Geometry Shaders

This extension defines a new shader type called a geometry shader. Geometry shaders are run after vertices are transformed, but prior to the remaining fixedfunction vertex processing, and may generate new vertices for, or remove vertices from the primitive assembly process. The name string for geometry shaders is GL_ARB_geometry_shader4. It was promoted to a core feature in OpenGL 3.2. K.3.3.41

Half-Precision Vertex Data

The name string for half-precision vertex data is GL_ARB_half_float_vertex. This extension is equivalent to new core functionality introduced in OpenGL 3.0, based on the earlier GL_NV_half_float extension, and is provided to enable this functionality in older drivers. K.3.3.42

Instanced Rendering

This instanced rendering interface is a less-capable form of GL_ARB_draw_instanced which can be supported on older hardware. The name string for instanced rendering is GL_ARB_instanced_arrays. It was promoted to a core feature in OpenGL 3.3. K.3.3.43

Flexible Buffer Mapping

The name string for flexible buffer mapping is GL_ARB_map_buffer_range. This extension is equivalent to new core functionality introduced in OpenGL 3.0,

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

755

based on the earlier GL_APPLE_flush_buffer_range extension, and is provided to enable this functionality in older drivers. K.3.3.44

Texture Buffer Objects

The name string for texture buffer objects is GL_ARB_texture_buffer_object. It was promoted to a core feature in OpenGL 3.1. K.3.3.45

RGTC Texture Compression Formats

The name string for RGTC texture compression formats is GL_ARB_texture_compression_rgtc. This extension is equivalent to new core functionality introduced in OpenGL 3.0, based on the earlier GL_EXT_texture_compression_rgtc extension, and is provided to enable this functionality in older drivers. It was promoted to a core feature in OpenGL 3.0. K.3.3.46

One- and Two-Component Texture Formats

The name string for one- and two-component texture formats is GL_ARB_texture_rg. It was promoted to a core feature in OpenGL 3.0. This extension is equivalent to new core functionality introduced in OpenGL 3.0, and is provided to enable this functionality in older drivers. K.3.3.47

Vertex Array Objects

The name string for vertex array objects is GL_ARB_vertex_array_object. This extension is equivalent to new core functionality introduced in OpenGL 3.0, based on the earlier GL_APPLE_vertex_array_object extension, and is provided to enable this functionality in older drivers. It was promoted to a core feature in OpenGL 3.0. K.3.3.48

Versioned Context Creation

Starting with OpenGL 3.0, a new context creation interface is required in the window system integration layer. This interface specifies the context version required as well as other attributes of the context. The name strings for the GLX and WGL context creation interfaces are GLX_ARB_create_context and WGL_ARB_create_context respectively.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.49

756

Uniform Buffer Objects

The name string for uniform buffer objects is GL_ARB_uniform_buffer_object. This extension is equivalent to new core functionality introduced in OpenGL 3.1 and is provided to enable this functionality in older drivers. K.3.3.50

Restoration of features removed from OpenGL 3.0

OpenGL 3.1 removes a large number of features that were marked deprecated in OpenGL 3.0. GL implementations needing to maintain these features to support existing applications may do so, following the deprecation model, by exporting an extension string indicating those features are present. Applications written for OpenGL 3.1 should not depend on any of the features corresponding to this extension, since they will not be available on all platforms with 3.1 implementations. The name string for restoration of features deprecated by OpenGL 3.0 is GL_ARB_compatibility. The profile terminology introduced with OpenGL 3.2 eliminates the necessity for evolving this extension. Instead, interactions between features removed by OpenGL 3.1 and new features introduced in later OpenGL versions are defined by the compatibility profile corresponding to those versions. K.3.3.51

Fast Buffer-to-Buffer Copies

The name string for fast buffer-to-buffer copies is GL_ARB_copy_buffer. This extension is equivalent to new core functionality introduced in OpenGL 3.1 and is provided to enable this functionality in older drivers. K.3.3.52

Shader Texture Level-of-Detail Control

The name string for shader texture level-of-detail control is GL_ARB_shader_texture_lod. This extension is equivalent to new core functions introduced in OpenGL Shading Language 1.30 and is provided to enable this functionality in older versions of the shading language. K.3.3.53

Depth Clamp Control

The name string for depth clamp control is GL_ARB_depth_clamp. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.54

757

Base Vertex Offset Drawing Commands

The name string for base vertex offset drawing commands is GL_ARB_draw_elements_base_vertex. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.55

Fragment Coordinate Convention Control

The name string for fragment coordinate convention control is GL_ARB_fragment_coord_conventions. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.56

Provoking Vertex Control

The name string for provoking vertex control is GL_ARB_provoking_vertex. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.57

Seamless Cube Maps

The name string for seamless cube maps is GL_ARB_seamless_cube_map. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.58

Fence Sync Objects

The name string for fence sync objects is GL_ARB_sync. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.59

Multisample Textures

The name string for multisample textures is GL_ARB_texture_multisample. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.60

758

BGRA Attribute Component Ordering

The name string for BGRA attribute component ordering is GL_ARB_vertex_array_bgra. This extension is equivalent to new core functionality introduced in OpenGL 3.2 and is provided to enable this functionality in older drivers. K.3.3.61

Per-Buffer Blend Control

The blending interface is extended to specify blend equation and blend function on a per-draw-buffer basis. The name string for per-buffer blend control is GL_ARB_draw_buffers_blend. It was promoted to a core feature in OpenGL 4.0. K.3.3.62

Sample Shading Control

Sample shading control adds the ability to request that an implementation use a minimum number of unique sets of fragment computation inputs when multisampling a pixel. The name string for sample shading control is GL_ARB_sample_shading. It was promoted to a core feature in OpenGL 4.0. K.3.3.63

Cube Map Array Textures

A cube map array texture is a two-dimensional array texture that may contain many cube map layers. Each cube map layer is a unique cube map image set. The name string for cube map array textures is GL_ARB_texture_cube_map_array. It was promoted to a core feature in OpenGL 4.0. K.3.3.64

Texture Gather

Texture gather adds a new set of texture functions (textureGather) to the OpenGL Shading Language that determine the 2 × 2 footprint used for linear filtering in a texture lookup, and return a vector consisting of the first component from each of the four texels in the footprint. The name string for texture gather is GL_ARB_texture_gather. It was promoted to a core feature in OpenGL 4.0. K.3.3.65

Texture Level-Of-Detail Queries

Texture level-of-detail queries adds a new set of fragment shader texture functions (textureLOD) to the OpenGL Shading Language that return the results of au-

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

759

tomatic level-of-detail computations that would be performed if a texture lookup were to be done. The name string for texture level-of-detail queries is GL_ARB_texture_query_lod. K.3.3.66

Profiled Context Creation

Starting with OpenGL 3.2, API profiles are defined. Profiled context creation extends the versioned context creation interface to specify a profile which must be implemented by the context. The name strings for the GLX and WGL profiled context creation interfaces are GLX_ARB_create_context_profile and WGL_ARB_create_context_profile respectively. K.3.3.67

Shading Language Include

Shading language include adds support for #include directives to shaders, and a named string API for defining the text corresponding to #include pathnames. The name string for shading language include is GL_ARB_shading_language_include. K.3.3.68

BPTC texture compression

BPTC texture compression provides new block compressed specific texture formats which can improve quality in images with sharp edges and strong chrominance transitions, and support high dynamic range floating-point formats. The name string for BPTC texture compression is GL_ARB_texture_compression_bptc. K.3.3.69

Extended Blend Functions

The name string for extended blend functions is GL_ARB_blend_func_extended. This extension is equivalent to new core functionality introduced in OpenGL 3.3, and is provided to enable this functionality in older drivers. K.3.3.70

Explicit Attribute Location

The name string for explicit attribute location is GL_ARB_explicit_attrib_location. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.71

760

Boolean Occlusion Queries

The name string for boolean occlusion queries is GL_ARB_occlusion_query2. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.72

Sampler Objects

The name string for sampler objects is GL_ARB_sampler_objects. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.73

Shader Bit Encoding

The name string for shader bit encoding is GL_ARB_shader_bit_encoding. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.74

RGB10A2 Integer Textures

The name string for RGB10A2 integer textures is GL_ARB_texture_rgb10_a2ui. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.75

Texture Swizzle

The name string for texture swizzle is GL_ARB_texture_swizzle. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.76

Timer Queries

The name string for timer queries is GL_ARB_timer_query. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. K.3.3.77

Packed 2.10.10.10 Vertex Formats

The name string for packed 2.10.10.10 vertex formats is GL_ARB_vertex_type_2_10_10_10_rev. This extension is equivalent to new core functionality introduced in OpenGL 3.3 and is provided to enable this functionality in older drivers. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.78

761

Draw Indirect

The name string for draw indirect is GL_ARB_draw_indirect. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.79

GPU Shader5 Miscellaneous Functionality

The name string for GPU shader5 miscellaneous functionality is GL_ARB_gpu_shader5. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.80

Double-Precision Floating-Point Shader Support

The name string for double-precision floating-point shader support is GL_ARB_gpu_shader_fp64. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.81

Shader Subroutines

The name string for shader subroutines is GL_ARB_shader_subroutine. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.82

Tessellation Shaders

The name string for tessellation shaders is GL_ARB_tessellation_shader. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.83

RGB32 Texture Buffer Objects

The name string for RGB32 texture buffer objects is GL_ARB_texture_buffer_object_rgb32. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.84

Transform Feedback 2

The name string for transform feedback 2 is GL_ARB_transform_feedback2. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.85

762

Transform Feedback 3

The name string for transform feedback 3 is GL_ARB_transform_feedback3. This extension is equivalent to new core functionality introduced in OpenGL 4.0 and is provided to enable this functionality in older drivers. K.3.3.86

OpenGL ES 2.0 Compatibility

The name string for OpenGL ES 2.0 compatibility is GL_ARB_ES2_compatibility. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers. K.3.3.87

Program Binary Support

The name string for program binary support is GL_ARB_get_program_binary. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers. K.3.3.88

Separate Shader Objects

The name string for separate shader objects is GL_ARB_separate_shader_objects. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers. K.3.3.89

Shader Precision Restrictions

The name string for shader precision restritions is GL_ARB_shader_precision. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers. K.3.3.90

Double Precision Vertex Shader Inputs

The name string for double precision vertex shader inputs is GL_ARB_vertex_attrib_64bit. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers. K.3.3.91

Viewport Arrays

The name string for viewport arrays is GL_ARB_viewport_array. This extension is equivalent to new core functionality introduced in OpenGL 4.1 and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.92

763

Robust Context Creation

Robust context creation allows creating an OpenGL context supporting robust buffer access behavior and a specified graphics reset notification behavior exposed through the GL_ARB_robustness extension (see section K.3.3.95). The name strings for GLX and WGL robust context creation are GLX_ARB_create_context_robustness and WGL_ARB_create_context_robustness, respectively. K.3.3.93

OpenCL Event Sharing

OpenCL event sharing allows creating OpenGL sync objects linked to OpenCL event objects, potentially improving efficiency of sharing images and buffers between the two APIs. The name string for OpenCL event sharing is GL_ARB_cl_event. K.3.3.94

Debug Output Notification

Debug output notification enables GL to inform the application when various events occur that may be useful during development and debugging. The name string for debug output notification is GL_ARB_debug_output. K.3.3.95

Context Robustness

Context robustness provides “safe” APIs that limit data written to application memory to a specified length, provides a mechanism to learn about graphics resets affecting the context, and defines guarantee that out-of-bounds buffer object accesses will have deterministic behavior precluding instability or termination. Some of these behaviors are controlled at context creation time via the companion GLX_ARB_create_context_robustness or WGL_ARB_create_context_robustness extensions (see section K.3.3.92). The name string for context robustness is GL_ARB_robustness. K.3.3.96

Shader Stencil Export

Sharder stencil export enables shaders to generate a stencil reference value, allowing stencil testing to be performed against per-shader-invocation values. The name string for shader stencil export is GL_ARB_shader_stencil_export.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.97

764

Base Instanced Rendering

The name string for base instanced rendering is GL_ARB_base_instance. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.98

OpenGL Shading Language 4.20 Feature Pack

The name string for the OpenGL Shading Language 4.20 feature pack is GL_ARB_shading_language_420pack. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.99

Instanced Transform Feedback

The name string for instanced transform feedback is GL_ARB_transform_feedback_instanced. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.100

Compressed Texture Pixel Storage

The name string for compressed texture pixel storage is GL_ARB_compressed_texture_pixel_storage. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.101

Conservative Depth

The name string for conservative depth is GL_ARB_conservative_depth. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.102

Internal Format Query

The name string for internal format query is GL_ARB_internalformat_query. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.103

765

Map Buffer Alignment

The name string for map buffer alignment is GL_ARB_map_buffer_alignment. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.104

Shader Atomic Counters

The name string for shader atomic counters is GL_ARB_shader_atomic_counters. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.105

Shader Image Load/Store

The name string for shader image load/store is GL_ARB_shader_image_load_store. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.106

Shading Language Packing

The name string for shading language packing is GL_ARB_shading_language_packing. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.107

Texture Storage

The name string for texture storage is GL_ARB_texture_storage. This extension is equivalent to new core functionality introduced in OpenGL 4.2 and is provided to enable this functionality in older drivers. K.3.3.108

ASTC Texture Compression

The name string for ASTC texture compression is GL_KHR_texture_compression_astc_ldr. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.109

Debug Contexts

This KHR extension defines debugging features and combines the functionality of GL_ARB_debug_output, GL_ARB_debug_output2, GL_ARB_debug_group,

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

766

and GL_ARB_debug_label. It is intended primarily to bring this debug functionality to OpenGL ES implementations. The name string for debug contexts is GL_KHR_debug. K.3.3.110

Shader Array of Arrays

The name string for shader array of arrays is GL_ARB_arrays_of_arrays. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.111

Clear Buffer Object

The name string for clear buffer object is GL_ARB_clear_buffer_object. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.112

Compute Shaders

The name string for compute shaders is GL_ARB_compute_shader. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.113

Copy Image

The name string for copy image is GL_ARB_copy_image. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.114

Texture Views

The name string for texture views is GL_ARB_texture_view. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.115

Vertex Attribute Binding

The name string for vertex attribute binding is GL_ARB_vertex_attrib_binding. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.116

767

Robustness Isolation

The name string for robustness isolation is GL_ARB_robustness_isolation. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.117

OpenGL ES 3.0 Compatibility

The name string for OpenGL ES 3.0 compatibility is GL_ARB_ES3_compatibility. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.118

Shader Explicit Uniform Location

The name string for shader explicit uniform location is GL_ARB_explicit_uniform_location. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.119

Fragment Layer Viewport

The name string for fragment layer viewport is GL_ARB_fragment_layer_viewport. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.120

Binding a Framebuffer Without Attachments

The name string for binding a framebuffer without attachments is GL_ARB_framebuffer_no_attachments. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.121 The

name

Extended Internal Format Query string

for

extended internal format query is GL_ARB_internalformat_query2. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.122

768

Invalidate SubData

The name string for invalidate subdata is GL_ARB_invalidate_subdata. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.123

Multi Draw Indirect

The name string for multi draw indirect is GL_ARB_multi_draw_indirect. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.124

Program Interface Queries

The name string for program interface queries is GL_ARB_program_interface_query. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.125

Robust Buffer Access Behavior

The name string for robust buffer access behavior is GL_ARB_robust_buffer_access_behavior. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.126

Shader Image Size Query

The name string for shader image size query is GL_ARB_shader_image_size. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.127

Shader Storage in Buffer Objects

The name string for shader storage in buffer objects is GL_ARB_shader_storage_buffer_object. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.128

Stencil Texturing

The name string for stencil texturing is GL_ARB_stencil_texturing. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.129

769

Texture Buffer Range

The name string for texture buffer range is GL_ARB_texture_buffer_range. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.130

Texture Query Levels

The name string for texture query levels is GL_ARB_texture_query_levels. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.131

Texture Storage Multisample

The name string for texture storage multisample is GL_ARB_texture_storage_multisample. This extension is equivalent to new core functionality introduced in OpenGL 4.3, and is provided to enable this functionality in older drivers. K.3.3.132

Robustness Application Isolation Context Creation

These extensions allow creation of OpenGL contexts which support robustness isolation through OpenGL 4.3 or the equivalent functionality in the GL_ARB_robustness_isolation extension (see K.3.3.116), and may also define additional constraints around how OpenGL context reset notification affects other contexts in the share group, or other applications on the system. There are equivalent sets of extensions for both GLX and WGL window-system binding layers. The name strings for GLX robustness application isolation context creation are GLX_ARB_robustness_application_isolation and GLX_ARB_robustness_share_group_isolation. The name strings for WGL robustness application isolation context creation are WGL_ARB_robustness_application_isolation and WGL_ARB_robustness_share_group_isolation. K.3.3.133

Buffer Storage

The name string for buffer storage is GL_ARB_buffer_storage. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.134

770

Clear Texture

The name string for clear texture is GL_ARB_clear_texture. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.135

Enhanced Layouts

The name string for enhanced layouts is GL_ARB_enhanced_layouts. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.136

Multiple Bind

The name string for multiple bind is GL_ARB_multi_bind. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.137

Query Buffer Object

The name string for query buffer object is GL_ARB_query_buffer_object. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.138

Texture Mirror Clamp To Edge

The name string for texture mirror clamp to edge is GL_ARB_texture_mirror_clamp_to_edge. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.139

Texture Stencil8

The name string for texture stencil8 is GL_ARB_texture_stencil8. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers. K.3.3.140

Vertex Type 10f 11f 11f

The name string for vertex type 10f 11f 11f is GL_ARB_vertex_type_10f_11f_11f_rev. This extension is equivalent to new core functionality introduced in OpenGL 4.4, and is provided to enable this functionality in older drivers.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.141

771

Bindless Textures

Bindless textures allows access to texture objects in shaders without first binding each texture to one of a limited number of texture image units. The name string for bindless textures is GL_ARB_bindless_texture. K.3.3.142

Compute Variable Group Size

Compute variable groups size allows writing generic compute shaders that operate on workgroups with arbitrary dimensions. The name string for compute variable group size is GL_ARB_compute_variable_group_size. K.3.3.143

Indirect Parameters

Indirect parameters allows buffer objects to store drawing parameters for some drawing commands, and introduces new drawing commands that source some of their parameters from buffers. The name string for indirect parameters is GL_ARB_indirect_parameters. K.3.3.144

Seamless Cubemap per Texture

This allows providing a per-texture setting for enabling seamless sampling from cube maps. The name string for seamless cubemap per texture is GL_ARB_seamless_cubemap_per_texture. K.3.3.145

Shader Draw Parameters

This adds two new built-in variables to the OpenGL Shading Language containing values passed in the basevertex and baseinstance parameters of drawinng commands, and an index of the draw command being processed by MultiDraw* commands. The name string for shader draw parameters is GL_ARB_shader_draw_parameters. K.3.3.146

Shader Group Vote

Shader group vote provides new built-in functions to compute the composite of a set of boolean conditions across a group of shader invocations. These composite results may be used to execute shaders more efficiently on a single-instruction multiple-data (SIMD) processor. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

772

The name string for shader group vote is GL_ARB_shader_group_vote. K.3.3.147

Sparse Textures

This extension allows the separation of the graphics processor’s address space (reservation) from the requirement that all textures must be physically backed (commitment). This exposes a limited form of virtualization for textures. Use cases include sparse (or partially resident) textures, texture paging, on-demand and delayed loading of texture assets and application controlled level-of-detail. The name string for sparse textures is GL_ARB_sparse_texture. K.3.3.148

OpenGL ES 3.1 Compatibility

This extension adds supported for features of OpenGL ES 3.1 that are missing from OpenGL 4.5. The name string for OpenGL ES 3.1 compatibility is GL_ARB_ES3_1_compatibility. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.149

Clip Control

This extension provides additional clip control modes to configure how clip space is mapped to window space. The name string for clip control is GL_ARB_clip_control. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.150

Inverted Conditional Rendering

This extension adds new modes to BeginConditionalRender which invert the condition used to determine whether to draw or not. The name string for inverted conditional rendering is GL_ARB_conditional_render_inverted. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.151

Cull Distance Shader Output

This extension adds a new OpenGL Shading Languagegl_CullDistance shader output, similar to gl_ClipDistance, but used for whole primitive culling.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

773

The name string for cull distance shader output is GL_ARB_cull_distance. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.152

Derivative Computation Granularity

This extension provides control over the spacial granularity at which the underlying implementation computes derivatives. The name string for derivative computation granularity is GL_ARB_derivative_control. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.153

Direct State Access

This extension allows state for most objects to be queried and modified without the object needing to be bound to a context. The name string for direct state access is GL_ARB_direct_state_access. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.154

Texture Subregion Readback

This extension adds commands allowing readback of subregions of texture images. The name string for texture subregion readback is GL_ARB_get_texture_sub_image. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.155

Shading Language Texture Sample Queries

This extension provides OpenGL Shading Language built-in functions allowing shaders to query the number of samples of a texture. The name string for shading language texture sample queries is GL_ARB_shader_texture_image_samples. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.156

Texture Barriers

This extension relaxes the restrictions on rendering to a currently bound texture and provides a mechanism to avoid read-after-write hazards. OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS

774

The name string for texture barriers is GL_ARB_texture_barrier. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.157

Context Flush Control

This extension allows querying the context flush behavior. There are associated EGL, GLX and WGL extensions which add new context creation parameters allowing specification of the desired behavior when a context is made non-current, as well as opting out of the implicit flush behavior of unextended OpenGL. The name string for context flush control is GL_KHR_context_flush_control. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.158

Robust Buffer Access Behavior

This extension specifies the behavior of out-of-bounds buffer and array accesses. The name string for robust buffer access behavior is GL_KHR_robust_buffer_access_behavior. K.3.3.159

Context Robustness

This extension adds APIs for safe access to application buffers by the GL implementation; mechanisms to learn about how graphics hardware resets affect the GL context; specification of behavior of GL commands after a graphics reset; and guarantees that out-of-bounds buffer object accesses by the GPU will have deterministic behavior. The name string for context robustness is GL_KHR_robustness. This extension is equivalent to new core functionality introduced in OpenGL 4.5, and is provided to enable this functionality in older drivers. K.3.3.160

Pipelin Statistics Queries

This extension introduces new query types that allow applications to get statistics information about different parts of the pipeline: The name string for pipeline statistics queries is GL_ARB_pipeline_statistics_query.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.161

775

Sparse Buffers

This extension adds a mechanism to decouple the virtual and physical storage requirements of buffer objects, and allows an application to create partially populated buffers that would over-subscribe available graphics memory if made fully resident. The name string for sparse buffers is GL_ARB_sparse_buffer. K.3.3.162

Transform Feedback Overflow Queries

This extension adds new query types which can be used to detect overflow of transform feedback buffers, and are accepted by conditional rendering commands. The name string for transform feedback overflow queries is GL_ARB_transform_feedback_overflow_query. K.3.3.163

Advanced Blend Equations

These extensions add a number of “advanced” blend equations. A base extension provides the new equations, but guarantees defined results only if each sample is touched no more than once in any single rendering pass, and defines a command to indicate pass boundaries. A layered extension guarantees that blending is done coherently and in API primitive order. The name strings for advanced blend equations are GL_KHR_blend_equation_advanced and GL_KHR_blend_equation_advanced_coherent. K.3.3.164

Undefined Error Behavior

This extension causes the GL implementation to have undefined behavior in any situation that would otherwise cause a GL error to be generated, allowing increased performance and decreased power usage for error-free applications. The extension is only supported for contexts created using associated EGL, GLX and WGL extensions which add corresponding new context creation parameters. The name string for undefined error behavior is GL_KHR_no_error. K.3.3.165

OpenGL ES 3.2 Compatibility

This extension adds supported for features of OpenGL ES 3.2 that are missing from OpenGL 4.5. The name string for OpenGL ES 3.2 compatibility is GL_ARB_ES3_2_compatibility.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.166

776

Fragment Shader Interlock

This extension provides new OpenGL Shading Language built-in functions that are used to delimit critical sections of fragment shader code. For pairs of shader invocations with “overlapping” coverage in a given pixel, the GL implementation will guarantee that the critical section of the fragment shader will be executed for only one fragment at a time. The name string for fragment shader interlock is GL_ARB_fragment_shader_interlock. K.3.3.167

64-bit Integer Shader Operations

This extension adds support for 64-bit integer data types to OpenGL Shading Language shaders, as well as related packing, conversion, relational, and mahtematical functions. The name string for 64-bit integer shader operations is GL_ARB_gpu_shader_int64. K.3.3.168

Parallel Shader Compilation

This extension provides a mechanism to allow compilation of OpenGL Shading Language shaders in multiple separate CPU threads, limit the number of threads to be used, and query if compilation is complete. The name string for parallel shader compilation is GL_ARB_parallel_shader_compile. K.3.3.169

Post Depth Coverage

This extension allows the fragment shader to control whether values in gl_SampleMaskIn[] reflect coverage after application of the early depth and stencil tests. The name string for post depth coverage is GL_ARB_post_depth_coverage. K.3.3.170

Explicit Sample Locations

This extension allows an application to modify the locations of samples within a pixel used in multisample rasterization, and to specify different sample locations for each pixel in a group of adjacent pixels The name string for explicit sample locations is GL_ARB_sample_locations.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.171

777

Shader Atomic Counter Operations

This extension extends the list of OpenGL Shading Language built-in functions that can operate on atomic counters. The name string for shader atomic counter operations is GL_ARB_shader_atomic_counter_ops. K.3.3.172

Shader Ballot

This extension provides the ability for a group of invocations which execute in lockstep to do limited forms of cross-invocation communication via a group broadcast of a invocation value, or broadcast of a bitarray representing a predicate value from each invocation in the group. The name string for shader ballot is GL_ARB_shader_ballot. K.3.3.173

Shader Clock

This extension exposes a 64-bit monotonically incrementing shader counter which may be used to derive local timing information within a single shader invocation. The name string for shader clock is GL_ARB_shader_clock. K.3.3.174

Shader Viewport Layer Array

This extension allows the gl_Layer and gl_ViewportIndex outputs to be written directly from the vertex shader with no geometry shader present. The name string for shader viewport layer array is GL_ARB_shader_viewport_layer_array. K.3.3.175

Sparse Texture 2

This extension adds additional shader functions and behaviors for use of sparse textures. The name string for sparse texture 2 is GL_ARB_sparse_texture2. K.3.3.176

Sparse Texture Clamp

This extension adds additional shader texture lookup functions that specify a minimum level-of-detail to use for lookups where the level-of-detail is computed automatically. The name string for sparse texture clamp GL_ARB_sparse_texture_clamp.

OpenGL 4.6 (Core Profile) - May 5, 2022

K.3. ARB AND KHRONOS EXTENSIONS K.3.3.177

778

Texture Filter Minmax

This extension provides a new texture and sampler parameter which allows applications to produce a filtered texel value by computing a component-wise minimum or maximum of the texels that would normally be averaged. The name string for texture filter minmax is GL_ARB_texture_filter_minmax. K.3.3.178

ASTC Sliced 3D Texture Compression

This extension supports slice-based 3D textures using the ASTC LDR profile, in the same fashion as the ASTC HDR profile when using the GL_KHR_texture_compression_astc_ldr extension. The name string for astc sliced 3d texture compression is GL_KHR_texture_compression_astc_sliced_3d. K.3.3.179

SPIR-V Shaders

This extension allows a SPIR-V module to be specified for use in a shader stage, rather than using OpenGL Shading Language, and modifies OpenGL Shading Language to be usable as a source language for creating SPIR-V modules, outside of the GL runtime. The name string for SPIR-V shaders is GL_ARB_gl_spirv. K.3.3.180

Create Context Undefined Error

These extensions allow the creation of an OpenGL or OpenGL ES context that doesn’t generate errors, and has undefined behavior on errors. They are paired with GL_KHR_no_error. The name strings for create context no error are GLX_ARB_create_context_no_error and WGL_ARB_create_context_no_error. There is a similar EGL_KHR_create_context_no_error extension for creating contexts through EGL, located in the EGL Registry.

OpenGL 4.6 (Core Profile) - May 5, 2022

Index #version, 4, 5, 568, 693, 696 Accum, 682 ACCUM BUFFER BIT, 682 ACTIVE ATOMIC COUNTER BUFFERS, 169, 621 ACTIVE ATTRIBUTE MAX LENGTH, 167, 615 ACTIVE ATTRIBUTES, 167, 615 ACTIVE PROGRAM, 170, 613 ACTIVE RESOURCES, 107, 108, 622 ACTIVE SUBROUTINE MAX LENGTH, 175, 620 ACTIVE SUBROUTINE UNIFORM LOCATIONS, 154, 156, 157, 174, 175, 620 ACTIVE SUBROUTINE UNIFORM MAX LENGTH, 175, 620 ACTIVE SUBROUTINE UNIFORMS, 175, 620 ACTIVE SUBROUTINES, 154, 157, 175, 620 ACTIVE TEXTURE, 179, 183, 242, 601 ACTIVE UNIFORM BLOCK MAX NAME LENGTH, 168, 617 ACTIVE UNIFORM BLOCKS, 168, 617 ACTIVE UNIFORM MAX LENGTH, 168, 615 ACTIVE UNIFORMS, 168, 615

ACTIVE VARIABLES, 111, 112, 139, 140, 623 ActiveShaderProgram, 125, 141, 698, 710, 721 ActiveTexture, 157, 179, 181 ALIASED LINE WIDTH RANGE, 636 ALL ATTRIB BITS, 683 ALL BARRIER BITS, 164, 166 ALL SHADER BITS, 124 ALPHA, 240, 285, 490, 511, 596, 598, 607, 681 ALPHA BITS, 682 ALPHA TEST, 682 ALPHA TEST REF, 715 AlphaFunc, 682 ALREADY SIGNALED, 38 ALWAYS, 239, 288, 504, 505, 602 AND, 515 AND INVERTED, 515 AND REVERSE, 515 Antialiasing, 476 ANY SAMPLES PASSED, 42, 45, 48, 381, 382, 505, 506, 721 ANY SAMPLES PASSED CONSERVATIVE, 42, 45, 48, 381, 382, 505, 506, 721 ARB arrays of arrays, 692, 702 ARB base instance, 685, 689 ARB buffer storage, 704 ARB clear buffer object, 692, 701 ARB clear texture, 704

779

INDEX

780

ARB compressed texture pixel storage, 684, 689 ARB compute shader, 692, 701 ARB conservative depth, 685, 689 ARB copy image, 692, 701, 702 ARB debug group, 701 ARB debug label, 701 ARB debug output, 692, 702, 730 ARB debug output2, 702 ARB enhanced layouts, 705 ARB ES3 compatibility, 692, 702 ARB explicit uniform location, 692, 702 ARB fragment layer viewport, 692, 702 ARB framebuffer no attachments, 692, 702 ARB gl spirv, 735, 743, 744 ARB gpu program5, 724 ARB indirect parameters, 744 ARB internalformat query, 685, 689 ARB internalformat query2, 692, 701 ARB invalidate subdata, 692, 702 ARB map buffer alignment, 685, 689 ARB multi bind, 705 ARB multi draw indirect, 692, 701 ARB pipeline statistics query, 743 ARB polygon offset clamp, 744 ARB program interface query, 692, 702 ARB query buffer object, 705 ARB robust buffer access behavior, 692, 702 ARB robustness, 702 ARB robustness isolation, 702 ARB shader atomic counter ops, 743 ARB shader atomic counters, 684, 689 ARB shader draw parameters, 744 ARB shader group vote, 744

ARB shader image load store, 685, 689 ARB shader image size, 692, 701 ARB shader storage buffer object, 692, 702 ARB shading language 420pack, 685, 689 ARB shading language packing, 689 ARB spirv extensions, 743 ARB stencil texturing, 692, 702 ARB texture buffer range, 692, 701 ARB texture compression bptc, 684, 689 ARB texture filter anisotropic, 744 ARB texture mirror clamp to edge, 705 ARB texture query levels, 692, 701 ARB texture stencil8, 705 ARB texture storage, 684, 689 ARB texture storage multisample, 692, 701 ARB texture view, 693, 702 ARB transform feedback instanced, 685, 689 ARB transform feedback overflow query, 743 ARB vertex attrib binding, 693, 702 ARB vertex type 10f 11f 11f rev, 705 AreTexturesResident, 682 ARRAY BUFFER, 61, 358, 363 363, ARRAY BUFFER BINDING, 587 ARRAY SIZE, 106, 111, 113, 137, 138, 388, 396, 623, 710, 725 ARRAY STRIDE, 111, 113, 138, 623 ArrayStride, 147, 148 ATOMIC COUNTER BARRIER BIT, 163, 165 ATOMIC COUNTER BUFFER, 61, 62,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

781

102, 106, 108–112, 139, 150 ATOMIC COUNTER BUFFER ACTIVE ATOMIC COUNTER INDICES, 140, 621 ATOMIC COUNTER BUFFER ACTIVE ATOMIC COUNTERS, 140, 621 ATOMIC COUNTER BUFFER BINDING, 85, 140, 621, 628 ATOMIC COUNTER BUFFER DATA SIZE, 140, 621 ATOMIC COUNTER BUFFER INDEX, 111, 113, 138, 623 ATOMIC COUNTER BUFFER REFERENCED BY COMPUTE SHADER, 140, 621 ATOMIC COUNTER BUFFER REFERENCED BY FRAGMENT SHADER, 140, 621 ATOMIC COUNTER BUFFER REFERENCED BY GEOMETRY SHADER, 140, 621 ATOMIC COUNTER BUFFER REFERENCED BY TESS CONTROL SHADER, 140, 621 ATOMIC COUNTER BUFFER REFERENCED BY TESS EVALUATION SHADER, 140 ATOMIC COUNTER BUFFER REFERENCED BY TESS EVALUTION SHADER, 621 ATOMIC COUNTER BUFFER REFERENCED BY VERTEX SHADER, 140, 621 ATOMIC COUNTER BUFFER SIZE, 85, 628 ATOMIC COUNTER BUFFER START, 85, 628

atomic uint, 119, 150, 739 AtomicCounter, 667, 668 atomicCounter, 662 atomicCounterDecrement, 662 atomicCounterIncrement, 662 AtomicCounterMemory, 668 AtomicStorage, 672 AtomicStorageOps, 672 ATTACHED SHADERS, 167, 170, 614 AttachShader, 95 ATTRIB STACK DEPTH, 683 BACK, 309, 478, 503, 507, 517–522, 530, 592, 681, 727 BACK LEFT, 309, 518, 519, 527 BACK RIGHT, 309, 518, 519, 527 barrier, 415 BaseInstance, 671 BaseVertex, 671 Begin, 679 BeginConditionalRender, 381, 381– 383, 772 BeginQuery, 43, 46, 51, 383, 409, 429, 439, 457, 498, 505, 548 BeginQueryIndexed, 43, 45, 46, 47, 451, 452, 727, 742 BeginTransformFeedback, 445, 445– 449, 738 BGR, 194, 534, 536 BGR INTEGER, 194 BGRA, 194, 197, 202, 353, 354, 357, 362, 534 BGRA INTEGER, 194, 197 BindAttribLocation, 132, 387, 387, 714, 730 BindBuffer, 28, 59, 60, 62, 238, 355, 364, 685, 697, 711 BindBufferBase, 62, 63, 64, 84, 447, 449, 686, 694, 697 BindBufferRange, 54, 62, 63, 64, 84,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

782

149, 151, 153, 447–449, 686, 694, 697, 698, 713 BindBuffersBase, 63, 64 BindBuffersRange, 64, 64, 65 BindFragDataLocation, 132, 495, 495, 496 BindFragDataLocationIndexed, 494, 496, 512 BindFramebuffer, 301, 302, 304, 330 BindImageTexture, 54, 290, 291–293, 298, 572, 576, 688 BindImageTextures, 291, 291 Binding, 129, 739 binding, 152 BindProgramPipeline, 99, 123, 123– 125, 157, 170, 407, 449 BindRenderbuffer, 312, 313, 314, 722 BindSampler, 28, 184, 184, 186, 187 BindSamplers, 185, 185 BindTexture, 157, 179, 180, 180, 181, 183, 287, 726 BindTextures, 181, 181 BindTextureUnit, 182 BindTransformFeedback, 443, 444, 740 BindVertexArray, 350, 351, 735 BindVertexBuffer, 354, 355, 356, 363, 711, 714 BindVertexBuffers, 355, 356 BITMAP, 681 Bitmap, 681 BLEND, 507, 513, 514, 603 BLEND COLOR, 603 BLEND DST ALPHA, 603 BLEND DST RGB, 603 BLEND EQUATION ALPHA, 603 BLEND EQUATION RGB, 603 BLEND SRC ALPHA, 603 BLEND SRC RGB, 603 BlendColor, 509, 512 BlendEquation, 507, 507

BlendEquationi, 507, 507 BlendEquationSeparate, 507, 507 BlendEquationSeparatei, 507, 507 BlendFunc, 509, 510 BlendFunci, 510, 510 BlendFuncSeparate, 509, 510 BlendFuncSeparatei, 510, 510 BlitFramebuffer, 26, 326, 529, 538, 538, 539, 541, 696, 699, 714, 715 BlitNamedFramebuffer, 538, 538, 541 Block, 147 BLOCK INDEX, 111, 113, 138, 623 BLUE, 194, 240, 285, 490, 534, 536, 596, 598, 607 BLUE BITS, 682 BLUE INTEGER, 194 BOOL, 116 bool, 116, 145 BOOL VEC2, 116 BOOL VEC3, 116 BOOL VEC4, 116 boolean, 142, 694 BUFFER, 557 BUFFER ACCESS, 61, 68, 75, 588 BUFFER ACCESS FLAGS, 61, 68, 75, 78, 588 BUFFER BINDING, 111, 113, 139, 140, 623 BUFFER DATA SIZE, 111, 113, 139, 140, 151, 152, 623, 736 BUFFER IMMUTABLE STORAGE, 61, 68, 70, 71, 588 BUFFER MAP LENGTH, 61, 68, 75, 78, 588 BUFFER MAP OFFSET, 61, 68, 75, 78, 588 BUFFER MAP POINTER, 61, 68, 75, 78, 82, 83, 588 BUFFER MAPPED, 61, 68, 75, 78, 588 BUFFER SIZE, 61, 68, 71–73, 76, 80,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

783

ClampColor, 535, 680 CLEAR, 515 Clear, 26, 335, 464, 523, 524–526, 682 CLEAR BUFFER, 571 CLEAR TEXTURE, 571, 715 ClearAccum, 682 ClearBuffer{if ui}v, 524 ClearBufferData, 72, 576 ClearBufferfi, 525, 526 ClearBufferfv, 525, 526 ClearBufferiv, 525, 526 ClearBufferSubData, 66, 71, 71, 72, 576, 697 ClearBufferuiv, 526 ClearColor, 523, 525 ClearDepth, 523, 525, 695 ClearDepthf, 523 ClearNamedBufferData, 72, 72 ClearNamedBufferSubData, 71, 71 ClearNamedFramebuffer, 524 callback, 552 ClearNamedFramebufferfi, 525, 526 CallList, 683 ClearNamedFramebufferfv, 526 CallLists, 683 ClearNamedFramebufferiv, 526 CAVEAT SUPPORT, 570 ClearNamedFramebufferuiv, 526 CCW, 168, 477, 592, 619 ClearStencil, 524, 525 ccw, 418 ClearTexImage, 284 Centroid, 668 ClearTexImageSubData, 71 centroid, 489 ClearTexSubImage, 282, 283, 284 centroid in, 489 CLIENT ALL ATTRIB BITS, 683 CheckFramebufferStatus, 330, 331, CLIENT ATTRIB STACK DEPTH, 331, 729 683 CheckNamedFramebufferStatus, 331, CLIENT MAPPED BUFFER BAR331, 332 RIER BIT, 67, 163, 723 CLAMP, 681 CLIENT STORAGE BIT, 67 CLAMP FRAGMENT COLOR, 680 ClientActiveTexture, 679 CLAMP READ COLOR, 535, 590 ClientWaitSync, 36, 37, 38, 38–40, 55 CLAMP TO BORDER, 240, 254, 260 CLIP DEPTH MODE, 589 CLAMP TO EDGE, 240, 254, 260, CLIP DISTANCEi, 589 285, 539 CLIP DISTANCE0, 455 CLAMP VERTEX COLOR, 680 CLIP ORIGIN, 589 83, 153, 235, 236, 588 BUFFER STORAGE FLAGS, 61, 68, 71, 73, 74, 588 BUFFER UPDATE BARRIER BIT, 163 BUFFER USAGE, 61, 68, 74, 588 BUFFER VARIABLE, 104, 111, 112, 114 BufferBlock, 147, 148, 668 BufferData, 56, 65, 68, 68–70, 78, 697, 709 BufferStorage, 65, 65, 67–70, 709 BufferSubData, 56, 66, 70, 70, 161, 164, 713 BuiltIn, 128, 742 bvec2, 116, 142 bvec3, 116 bvec4, 116 BYTE, 193, 297, 298, 352, 537, 538

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

784

ClipControl, 453 ClipDistance, 671, 672 CLIPPING INPUT PRIMITIVES, 43, 48, 457 CLIPPING OUTPUT PRIMITIVES, 43, 48, 457 ClipPlane, 680 coherent, 164 COLOR, 218, 524, 526, 527 COLOR ATTACHMENTm, 722 COLOR ATTACHMENT0, 303, 521, 530 COLOR BUFFER BIT, 523, 526, 539– 541 COLOR CLEAR VALUE, 604 COLOR COMPONENTS, 573 COLOR ENCODING, 573 COLOR INDEX, 679 COLOR LOGIC OP, 514, 603 COLOR MATERIAL, 680 COLOR RENDERABLE, 573 COLOR SUM, 682 COLOR WRITEMASK, 522, 604 ColorMask, 521, 521–523 ColorMaski, 521, 521, 522, 715 ColorMaterial, 680 ColorPointer, 679 COMMAND BARRIER BIT, 162, 739 COMPARE REF TO TEXTURE, 239, 287 COMPATIBLE SUBROUTINES, 111, 120, 156, 620, 623, 739 COMPILE STATUS, 90, 92, 101, 166, 167, 612 CompileShader, 90, 90, 91 Component, 129, 390, 495, 497 component, 114, 126, 385, 390, 495, 722 COMPRESSED R11 EAC, 212, 676 COMPRESSED RED, 212

COMPRESSED RED RGTC1, 212, 273, 544, 675 COMPRESSED RG, 212 COMPRESSED RG11 EAC, 212, 676 COMPRESSED RG RGTC2, 212, 273, 544, 675 COMPRESSED RGB, 212 COMPRESSED RGB8 ETC2, 212, 676 COMPRESSED RGB8 PUNCHTHROUGH ALPHA1 ETC2, 212, 676 COMPRESSED RGB BPTC SIGNED FLOAT, 212, 273, 544, 675 COMPRESSED RGB BPTC UNSIGNED FLOAT, 212, 273, 544, 675 COMPRESSED RGBA, 212 COMPRESSED RGBA8 ETC2 EAC, 212, 676 COMPRESSED RGBA BPTC UNORM, 212, 273, 544, 675 COMPRESSED SIGNED R11 EAC, 212, 676 COMPRESSED SIGNED RED RGTC1, 212, 273, 544, 675 COMPRESSED SIGNED RG11 EAC, 212, 676 COMPRESSED SIGNED RG RGTC2, 212, 273, 544, 675 COMPRESSED SRGB, 212, 289 COMPRESSED SRGB8 ALPHA8 ETC2 EAC, 212, 289, 676 COMPRESSED SRGB8 ETC2, 212, 289, 676 COMPRESSED SRGB8 PUNCHTHROUGH ALPHA1 ETC2, 212, 289, 676

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

785

COMPRESSED SRGB ALPHA, 212, 289 COMPRESSED SRGB ALPHA BPTC UNORM, 212, 273, 289, 544, 675 COMPRESSED TEXTURE FORMATS, 206, 637, 679, 685 CompressedTexImage1D, 215, 226, 227, 229–231, 576 CompressedTexImage2D, 226, 227, 229–231, 576 CompressedTexImage3D, 226, 227– 231, 251, 576, 706, 738 CompressedTexSubImage*D, 232 CompressedTexSubImage1D, 230 CompressedTexSubImage2D, 230 CompressedTexSubImage3D, 230, 252 CompressedTextureSubImage*D, 232 CompressedTextureSubImage1D, 230 CompressedTextureSubImage2D, 230 CompressedTextureSubImage3D, 230 COMPUTE SHADER, 89, 155, 546, 613, 709 COMPUTE SHADER BIT, 124 COMPUTE SHADER INVOCATIONS, 43, 48, 548 COMPUTE SUBROUTINE, 103, 155 COMPUTE SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 COMPUTE TEXTURE, 571 COMPUTE WORK GROUP SIZE, 169, 547, 614 CONDITION SATISFIED, 38 CONSTANT ALPHA, 511 CONSTANT COLOR, 511 CONTEXT COMPATIBILITY PROFILE BIT, 567 CONTEXT CORE PROFILE BIT, 567 CONTEXT FLAG DEBUG BIT, 549,

567 CONTEXT FLAG FORWARD COMPATIBLE BIT, 567 CONTEXT FLAG NO ERROR BIT, 568 CONTEXT FLAG ROBUST ACCESS BIT, 568 CONTEXT FLAGS, 549, 567, 568, 638 CONTEXT LOST, 16–18, 20 CONTEXT PROFILE MASK, 567, 638, 723 CONTEXT RELEASE BEHAVIOR, 568, 635 CONTEXT RELEASE BEHAVIOR FLUSH, 568 COPY, 515, 603 COPY INVERTED, 515 COPY READ BUFFER, 61, 81, 685, 686, 697 COPY READ BUFFER BINDING, 656, 686 COPY WRITE BUFFER, 61, 81, 685, 686, 697 COPY WRITE BUFFER BINDING, 656, 686 CopyBufferSubData, 66, 80, 81, 728 CopyImageSubData, 206, 542, 542– 544, 724, 741 CopyNamedBufferSubData, 80, 81 CopyPixels, 218, 682 CopyTexImage, 283, 682 CopyTexImage1D, 219, 220, 222, 265, 576 CopyTexImage2D, 218, 219, 220, 222, 265, 576 CopyTexSubImage1D, 220, 221 CopyTexSubImage2D, 220, 221 CopyTexSubImage3D, 220, 221 CopyTextureSubImage1D, 221, 221 CopyTextureSubImage2D, 221, 221

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

786

CopyTextureSubImage3D, 220, 221, 222 CreateBuffers, 28, 59, 59, 355 CreateFramebuffers, 302, 302 CreateProgram, 28, 94 CreateProgramPipelines, 123, 123 CreateQueries, 44, 44, 45, 47, 51, 727, 742 CreateRenderbuffers, 313, 313 CreateSamplers, 184, 184, 724 CreateShader, 88, 89 CreateShaderProgramv, 100, 101 CreateTextures, 28, 182, 182, 243 CreateTransformFeedbacks, 444, 444 CreateVertexArrays, 28, 351, 351, 735 CULL FACE, 478, 592 CULL FACE MODE, 592 CullDistance, 671, 672 CullFace, 478, 478, 482 CURRENT PROGRAM, 614 CURRENT QUERY, 47, 48, 656, 743 CURRENT VERTEX ATTRIB, 379, 625, 687 CW, 168, 477 cw, 418

DEBUG SEVERITY HIGH, 551 DEBUG SEVERITY LOW, 551, 552 DEBUG SEVERITY MEDIUM, 551 DEBUG SEVERITY NOTIFICATION, 551, 556 DEBUG SOURCE API, 17, 550, 571 DEBUG SOURCE APPLICATION, 550, 555, 556 DEBUG SOURCE OTHER, 550 DEBUG SOURCE SHADER COMPILER, 550 DEBUG SOURCE THIRD PARTY, 550, 555, 556 DEBUG SOURCE WINDOW SYSTEM, 550 DEBUG TYPE DEPRECATED BEHAVIOR, 551 DEBUG TYPE ERROR, 17, 551 DEBUG TYPE MARKER, 551 DEBUG TYPE OTHER, 551 DEBUG TYPE PERFORMANCE, 551, 571 DEBUG TYPE POP GROUP, 551, 556 DEBUG TYPE PORTABILITY, 551 DEBUG TYPE PUSH GROUP, 551, 556 DEBUG CALLBACK FUNCTION, DEBUG TYPE UNDEFINED BE553, 559, 566, 649 HAVIOR, 551 DEBUG CALLBACK USER DebugMessageCallback, 552, 553, 559, PARAM, 559, 566, 649 715 DEBUG GROUP STACK DEPTH, DebugMessageControl, 552, 554, 696, 649 715 DEBUG LOGGED MESSAGES, 553, DebugMessageInsert, 555, 555, 730 649 DECR, 504 DEBUG NEXT LOGGED MESDECR WRAP, 504 SAGE LENGTH, 553, 649 DELETE STATUS, 91, 166, 167, 612, DEBUG OUTPUT, 405, 549, 553–555, 614 649 DeleteBuffers, 28, 54, 60, 62, 63, 65, DEBUG OUTPUT SYNCHRONOUS, 355, 697 558, 559, 649 DeleteFramebuffers, 302, 304 OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

787

DeleteLists, 683 DeleteProgram, 100, 100 DeleteProgramPipelines, 122, 123–125, 170, 408 DeleteQueries, 44, 45, 52 DeleteRenderbuffers, 54, 313, 314, 330 DeleteSamplers, 185, 187, 706 DeleteShader, 91, 91 DeleteSync, 37, 38, 41 DeleteTextures, 54, 183, 291, 330 DeleteTransformFeedbacks, 443, 444, 687, 700 DeleteVertexArrays, 350, 351 DEPTH, 218, 309, 525–527, 598, 607 Depth, 667 DEPTH24 STENCIL8, 211 DEPTH32F STENCIL8, 211 DEPTH ATTACHMENT, 303, 319, 329 DEPTH BITS, 682 DEPTH BUFFER BIT, 523, 526, 539, 541 DEPTH CLAMP, 455, 589 DEPTH CLEAR VALUE, 604 DEPTH COMPONENT, 194, 205, 211, 215, 239, 247, 248, 283–285, 287, 327, 401, 402, 532–534, 597 DEPTH COMPONENT16, 211 DEPTH COMPONENT24, 211 DEPTH COMPONENT32, 211 DEPTH COMPONENT32F, 211 DEPTH COMPONENTS, 574 DEPTH FUNC, 602 DEPTH RANGE, 589, 716 DEPTH RENDERABLE, 574 DEPTH STENCIL, 194, 197, 202–205, 211, 215, 218, 247, 248, 268, 269, 283, 284, 287, 318, 325, 327, 401, 402, 525, 526, 533,

534 DEPTH STENCIL ATTACHMENT, 309, 312, 318, 319, 325, 695 DEPTH STENCIL TEXTURE MODE, 239, 268, 269, 285, 287, 402, 597, 707, 725 DEPTH TEST, 504, 602 DEPTH TEXTURE MODE, 681 DEPTH WRITEMASK, 604 DepthFunc, 505 DepthMask, 522, 523 DepthRange, 15, 459, 459 DepthRangeArrayv, 459, 459 DepthRangef, 459 DepthRangeIndexed, 459, 459 DerivativeControl, 672 DetachShader, 95 Device, 667 dFdx, 563 dFdy, 563 Disable, 253, 361, 455, 464, 467–469, 471, 477, 478, 485, 486, 501, 503, 504, 507, 514, 549, 558, 680–682, 705 DisableClientState, 679 Disablei, 485, 506, 507 DisableVertexArrayAttrib, 359, 359 DisableVertexAttribArray, 359, 359, 379 DISPATCH INDIRECT BUFFER, 61, 162, 365, 547, 548, 697, 741 DISPATCH INDIRECT BUFFER BINDING, 634 DispatchCompute, 546, 547, 723 DispatchComputeIndirect, 162, 364, 365, 547, 697, 711, 741 DITHER, 514, 603 dmatC, 145 dmatCxR, 145

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

788

dmat*, 686, 688 dmat2, 116, 386 dmat2x3, 116, 386, 389 dmat2x4, 116, 386, 389 dmat3, 116, 142, 386, 389 dmat3x2, 116, 386 dmat3x4, 117, 386, 389 dmat4, 116, 386, 389 dmat4x2, 117, 386 dmat4x3, 117, 386, 389 do, 415 DONT CARE, 554, 563, 633 DOUBLE, 116, 352 double, 116, 127, 145, 386, 459, 693 DOUBLE MAT2, 116 DOUBLE MAT2x3, 116 DOUBLE MAT2x4, 116 DOUBLE MAT3, 116 DOUBLE MAT3x2, 116 DOUBLE MAT3x4, 117 DOUBLE MAT4, 116 DOUBLE MAT4x2, 117 DOUBLE MAT4x3, 117 DOUBLE VEC2, 116 DOUBLE VEC3, 116 DOUBLE VEC4, 116 DOUBLEBUFFER, 307, 655 DRAW BUFFER, 521 DRAW BUFFERi, 606 DRAW BUFFER0, 521 DRAW FRAMEBUFFER, 301, 302, 304–308, 311, 318–321, 331, 517, 519, 526, 527, 538, 605, 695 DRAW FRAMEBUFFER BINDING, 264, 305, 333, 605 DRAW INDIRECT, 368, 375 DRAW INDIRECT BUFFER, 61, 162, 365, 367, 374, 741

DRAW INDIRECT BUFFER BINDING, 587 DrawArrays, 337, 339, 351, 360, 363, 366, 366, 367, 404, 729 DrawArraysIndirect, 364, 365, 367, 368, 369, 699, 711 DrawArraysInstanced, 366, 367, 450, 699 DrawArraysInstancedBaseInstance, 366, 366, 371, 699, 743 DrawArraysOneInstance, 364, 365, 730 DrawBuffer, 515, 516, 517, 517, 518, 520, 523, 526, 695, 715 DrawBuffers, 516, 518, 519, 520, 695 DrawElements, 159, 351, 360, 363, 364, 371, 371–373 DrawElementsBaseVertex, 364, 373, 376, 377 DrawElementsIndirect, 364, 365, 374, 375, 699, 711 DrawElementsInstanced, 364, 372, 372, 699, 735 DrawElementsInstancedBaseInstance, 371, 372, 699 DrawElementsInstancedBaseVertex, 364, 373, 699 DrawElementsInstancedBaseVertexBaseInstance, 373, 699 DrawElementsOneInstance, 369, 370, 371, 373, 730, 735 DrawIndex, 671 DrawParameters, 672 DrawPixels, 681 DrawRangeElements, 364, 372, 373, 636 DrawRangeElementsBaseVertex, 364, 373, 373 DrawTransformFeedback, 450, 451 DrawTransformFeedbackInstanced, 450, 450, 699

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

789

DrawTransformFeedbackStream, 450, EndQuery, 47, 51, 505, 506 451 EndQueryIndexed, 46, 46, 742 DrawTransformFeedbackStreamInstanced,EndStreamPrimitive, 434, 439 450, 450, 451, 699 EndTransformFeedback, 56, 57, 445, DST ALPHA, 511 445, 449, 451 EQUAL, 168, 239, 288, 504, 505, 619 DST COLOR, 511 equal spacing, 416, 424 dvec2, 116, 386 dvec3, 116, 386, 389, 686, 688 EQUIV, 515 dvec4, 116, 385, 386, 389, 686, 688 ExecutionMode, 94 DYNAMIC COPY, 61, 69 exp(x), 673 exp2(x), 673 DYNAMIC DRAW, 61, 68, 69 exp2(y * log2(x)), 673 DYNAMIC READ, 61, 69 DYNAMIC STORAGE BIT, 66, 68, EXT texture filter anisotropic, 744 69, 71, 709 EXTENSIONS, 568, 569, 638, 683, 747 early fragment tests, 498 EdgeFlagPointer, 679 EGL KHR create context no error, 778 ELEMENT ARRAY BARRIER BIT, 162 ELEMENT ARRAY BUFFER, 61, 162, 364, 380 ELEMENT ARRAY BUFFER BINDING, 377, 380, 586, 730 EmitStreamVertex, 434 Enable, 253, 361, 455, 464, 467–469, 471, 477, 478, 485, 486, 501, 503, 504, 507, 514, 549, 558, 565, 680–682, 705 EnableClientState, 679 Enablei, 485, 506, 507 EnableVertexArrayAttrib, 359, 359 EnableVertexAttribArray, 351, 359, 359, 379 End, 679 EndConditionalRender, 381, 381, 383 EndList, 683 EndPrimitive, 335, 434, 439

FALSE, 14, 15, 41, 45, 47, 49, 60, 61, 68, 70, 78, 89–92, 94, 96, 99– 101, 122, 131, 141, 142, 166, 167, 169, 175, 176, 183, 188, 189, 285, 290–292, 294, 305, 311, 314, 324, 348, 351, 353, 354, 361, 379, 381, 406, 443, 464, 482, 487, 506, 532, 535, 549, 554, 558, 573, 574, 576, 578, 579, 585, 587–589, 591– 593, 597, 599, 602, 603, 606, 607, 610–614, 619, 621, 625– 627, 630, 649, 656 false, 492 FASTEST, 563 FeedbackBuffer, 682 FenceSync, 28, 36, 36, 37, 40, 56, 67, 723 FILL, 480, 482, 483, 592, 664 FILTER, 571 Finish, 21, 21, 36, 56, 67, 664 FIRST VERTEX CONVENTION, 438, 453 FIXED, 353 FIXED ONLY, 535, 544, 590

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

790

Flat, 668 flat, 433, 453, 726 FLOAT, 116, 193, 245, 294, 297, 310, 352, 381, 535–537, 576, 585 float, 116, 127, 145, 386 Float64, 672 FLOAT 32 UNSIGNED INT 24 8 REV, 192, 193, 195, 197, 201, 533, 537, 538 FLOAT MAT2, 116 FLOAT MAT2x3, 116 FLOAT MAT2x4, 116 FLOAT MAT3, 116 FLOAT MAT3x2, 116 FLOAT MAT3x4, 116 FLOAT MAT4, 116 FLOAT MAT4x2, 116 FLOAT MAT4x3, 116 FLOAT VEC2, 116 FLOAT VEC3, 116 FLOAT VEC4, 116 Flush, 20, 20, 21, 40, 664, 738 FlushMappedBufferRange, 56, 74, 75, 77, 77 FlushMappedNamedBufferRange, 77, 77, 78 fma(), 673 FOG, 682 Fog, 682 FOG HINT, 683 FogCoordPointer, 679 for, 415 FRACTIONAL EVEN, 168 fractional even spacing, 416 FRACTIONAL ODD, 168 fractional odd spacing, 416, 418 FragCoord, 671 FragDepth, 671 FRAGMENT INTERPOLATION OFFSET BITS, 489,

652 FRAGMENT SHADER, 89, 155, 172, 173, 613 FRAGMENT SHADER BIT, 124 FRAGMENT SHADER DERIVATIVE HINT, 563, 633 FRAGMENT SHADER INVOCATIONS, 43, 48, 498 FRAGMENT SUBROUTINE, 103, 155 FRAGMENT SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 FRAGMENT TEXTURE, 571 FRAMEBUFFER, 302, 305–308, 311, 318–321, 331, 526, 527, 557, 695 FRAMEBUFFER ATTACHMENT x SIZE, 607 FRAMEBUFFER ATTACHMENT ALPHA SIZE, 310 FRAMEBUFFER ATTACHMENT BLUE SIZE, 310 FRAMEBUFFER ATTACHMENT COLOR ENCODING, 219, 310, 508, 513, 514, 539, 607 FRAMEBUFFER ATTACHMENT COMPONENT TYPE, 310, 607 FRAMEBUFFER ATTACHMENT DEPTH SIZE, 310 FRAMEBUFFER ATTACHMENT GREEN SIZE, 310 FRAMEBUFFER ATTACHMENT LAYERED, 311, 324, 607, 712 FRAMEBUFFER ATTACHMENT OBJECT NAME, 309–311, 318, 324, 328, 607

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

791

FRAMEBUFFER ATTACHMENT OBJECT TYPE, 309– 311, 318, 324, 328, 333, 607, 720, 738 FRAMEBUFFER ATTACHMENT RED SIZE, 310 FRAMEBUFFER ATTACHMENT STENCIL SIZE, 310 FRAMEBUFFER ATTACHMENT TEXTURE CUBE MAP FACE, 311, 324, 607 FRAMEBUFFER ATTACHMENT TEXTURE LAYER, 311, 324, 334, 607, 707 FRAMEBUFFER ATTACHMENT TEXTURE LEVEL, 264, 311, 324, 607 FRAMEBUFFER BARRIER BIT, 163, 165 FRAMEBUFFER BINDING, 305 FRAMEBUFFER BLEND, 571 FRAMEBUFFER COMPLETE, 331 FRAMEBUFFER DEFAULT, 309 FRAMEBUFFER DEFAULT FIXED SAMPLE LOCATIONS, 305– 307, 606 FRAMEBUFFER DEFAULT HEIGHT, 305–307, 329, 606 FRAMEBUFFER DEFAULT LAYERS, 305–307, 606 FRAMEBUFFER DEFAULT SAMPLES, 305–307, 606 FRAMEBUFFER DEFAULT WIDTH, 305–307, 329, 606 FRAMEBUFFER INCOMPLETE ATTACHMENT, 329 FRAMEBUFFER INCOMPLETE LAYER TARGETS, 330 FRAMEBUFFER INCOMPLETE -

MISSING ATTACHMENT, 329 FRAMEBUFFER INCOMPLETE MULTISAMPLE, 329, 330 FRAMEBUFFER RENDERABLE, 332, 571, 740 FRAMEBUFFER RENDERABLE LAYERED, 332, 571, 740 FRAMEBUFFER SRGB, 508, 513, 514, 539, 573, 603, 712 FRAMEBUFFER UNDEFINED, 329 FRAMEBUFFER UNSUPPORTED, 329, 331 FramebufferParameteri, 305, 305–307, 707 FramebufferRenderbuffer, 318, 318, 319, 330, 707 FramebufferTexture, 54, 320, 320, 322, 323, 729 FramebufferTexture1D, 321, 322 FramebufferTexture2D, 321, 322 FramebufferTexture3D, 321, 322 FramebufferTextureLayer, 322, 735 FRONT, 309, 478, 503, 507, 517, 518, 520–522, 530, 681, 727 FRONT AND BACK, 478, 480, 503, 507, 518, 520–522, 530 FRONT FACE, 592 FRONT LEFT, 309, 518, 519, 527 FRONT RIGHT, 309, 518, 519, 527 FrontFace, 477, 478, 492, 680 FrontFacing, 671 Frustum, 679 FULL SUPPORT, 332, 571, 740 FUNC ADD, 509, 512, 603 FUNC REVERSE SUBTRACT, 509 FUNC SUBTRACT, 509 Function, 667 fwidth, 563

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

792

GenBuffers, 28, 59, 59, 60, 62, 63, 355 GENERATE MIPMAP, 682 GENERATE MIPMAP HINT, 683 GenerateMipmap, 266, 266, 267, 720 GenerateTextureMipmap, 266, 266, 267 GenFramebuffers, 301, 302, 304, 304 GenLists, 683 GenProgramPipelines, 122, 122–125, 170, 407, 408 GenQueries, 43, 43–45, 52 GenRenderbuffers, 312, 313, 314, 314 GenSamplers, 184, 184–188 GenTextures, 180, 180, 181, 183, 274, 686 GenTransformFeedbacks, 442, 442–444 GenVertexArrays, 350, 350, 351 Geometry, 672 GEOMETRY INPUT TYPE, 168, 169, 431, 616 GEOMETRY OUTPUT TYPE, 168, 169, 432, 616 GEOMETRY SHADER, 89, 155, 430, 613 GEOMETRY SHADER BIT, 124 GEOMETRY SHADER INVOCATIONS, 43, 48, 168, 169, 439, 616 GEOMETRY SHADER PRIMITIVES EMITTED, 43, 48, 439 GEOMETRY SUBROUTINE, 103, 155 GEOMETRY SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 GEOMETRY TEXTURE, 571 GEOMETRY VERTICES OUT, 168, 169, 432, 436, 616 GeometryPointSize, 672 GeometryStreams, 672

GEQUAL, 239, 288, 504, 505 GET TEXTURE IMAGE FORMAT, 574 GET TEXTURE IMAGE TYPE, 574 GetActiveSubroutineUniformName, 620 GetActiveSubroutineName, 620 GetActiveSubroutineUniformiv, 620 GetActiveUniformBlockiv, 618, 619 GetActiveAtomicCounterBufferiv, 621 GetActiveAtomicCounterBufferiv, 139, 140 GetActiveAttrib, 388, 388, 615, 700, 721 GetActiveSubroutineName, 155, 735 GetActiveSubroutineUniformiv, 156, 156 GetActiveSubroutineUniformName, 156, 735 GetActiveUniform, 137, 142, 615, 710 GetActiveUniformBlockiv, 138, 139 GetActiveUniformBlockName, 138 GetActiveUniformName, 136 GetActiveUniformsiv, 137, 138, 617, 618, 621 GetAttachedShaders, 170, 614 GetAttribLocation, 387, 389, 615 GetBooleani v, 522, 565, 604, 627 GetBooleanv, 14, 361, 522, 564, 580, 593, 604, 610, 611, 630, 635, 637, 655 GetBufferParameteri64v, 81, 588 GetBufferParameteriv, 81, 588, 730 GetBufferPointerv, 83, 83, 588 GetBufferSubData, 82, 82, 588 GetCompressedTexImage, 229, 231, 245, 250, 250, 252, 532, 563 GetCompressedTexSubImage, 252 GetCompressedTextureImage, 250, 250, 251

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

793

GetCompressedTextureSubImage, 251 GetDebugMessageLog, 553, 559, 559, 560, 715 GetDoublei v, 565, 589 GetDoublev, 15, 564 GetError, 16, 16, 18, 20, 656 GetFloati v, 565, 589 GetFloatv, 11, 15, 461, 468, 564, 584, 591–593, 603, 604, 635, 636, 652, 715 GetFragDataIndex, 497, 721 GetFragDataLocation, 497, 721 GetFramebufferAttachmentParameteriv, 607 GetFramebufferParameteriv, 606 GetFramebufferAttachmentParameteriv, 308, 308, 311, 707, 711, 714, 729, 738 GetFramebufferParameteriv, 307, 307, 308, 312, 695, 729 GetGraphicsResetStatus, 19, 19, 20 GetInteger, 569 GetInteger64i v, 84, 565, 580, 586, 628–631 GetInteger64v, 15, 39, 51, 370, 564, 635, 646, 651, 686 GetIntegeri v, 84, 487, 513, 547, 565, 580, 581, 586, 593, 602, 603, 627–631, 644 GetIntegeriv, 656 GetIntegerv, 15, 19, 51, 133, 144, 148, 150, 152, 179, 185, 206, 305, 312, 313, 372, 380, 438, 454, 510, 513, 519, 521, 530, 532, 547, 564, 567, 568, 580, 581, 584, 586, 587, 589–592, 594, 595, 601–606, 608, 610, 611, 614, 628, 629, 631, 633–652, 654–656 GetInternalformati64v, 569, 577, 735

GetInternalformativ, 234, 315, 332, 569, 653, 735 GetMultisamplefv, 399, 466, 467, 655, 708 GetNamedBufferParameteri64v, 82 GetNamedBufferParameteriv, 81 GetNamedBufferPointerv, 83, 83 GetNamedBufferSubData, 82, 82, 83 GetNamedFramebufferAttachmentParameteriv, 308, 308, 311, 729 GetNamedFramebufferParameteriv, 307, 307, 308, 312, 729 GetNamedRenderbufferParameteriv, 317, 317 GetnCompressedTexImage, 250, 250 GetnTexImage, 246, 246, 247, 250, 722 GetnUniformdv, 173 GetnUniformfv, 173 GetnUniformiv, 174 GetnUniformuiv, 174 GetObjectLabel, 560, 561, 585, 588, 597, 600, 606, 609, 612–614, 626, 630 GetObjectPtrLabel, 561, 632 GetPointerv, 559, 566, 649, 693, 725 GetPolygonStipple, 715 GetProgramInterfaceiv, 622 GetProgramResourceiv, 623, 624 GetProgramStageiv, 620 GetProgramBinary, 130, 131, 132, 614, 721, 737 GetProgramInfoLog, 98, 131, 171, 171, 614, 719 GetProgramInterfaceiv, 107, 110 GetProgramiv, 96, 130, 131, 167, 167, 170, 171, 406, 411, 431, 432, 436, 547, 614–617, 619, 621, 728 GetProgramPipelineInfoLog, 171, 171, 719

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

794

GetProgramPipelineiv, 169, 171, 407, 613 GetProgramPiplineInfoLog, 613 GetProgramResourceIndex, 109 GetProgramResourceiv, 110, 111, 112, 138–140, 145, 151, 496, 710, 720, 727, 735, 739 GetProgramResourceLocation, 120, 120, 121 GetProgramResourceLocationIndex, 120, 120, 121 GetProgramResourceName, 109 GetProgramStageiv, 175 GetQueryBufferObjecti64v, 49 GetQueryBufferObjectiv, 49 GetQueryBufferObjectui64v, 49 GetQueryBufferObjectuiv, 49 GetQueryIndexediv, 47, 742 GetQueryiv, 49, 651, 656 GetQueryObjecti64v, 49 GetQueryObjectiv, 49, 626 GetQueryObjectui64v, 49 GetQueryObjectuiv, 20, 49, 626 GetRenderbufferParameteriv, 609 GetRenderbufferParameteriv, 317, 317, 333 GetSamplerParameter, 188 GetSamplerParameterfv, 600 GetSamplerParameterI{i ui}v, 188 GetSamplerParameterIiv, 188 GetSamplerParameterIuiv, 14, 188 GetSamplerParameteriv, 600 GetShaderInfoLog, 90, 171, 171, 612, 719 GetShaderiv, 90, 91, 166, 171, 172, 612 GetShaderPrecisionFormat, 91, 172 GetShaderSource, 172, 612 GetString, 566, 567, 638, 683, 725 GetStringi, 568, 638, 747 GetSubroutineIndex, 155

GetSubroutineUniformLocation, 156, 698 GetSynciv, 20, 37, 41, 41, 632, 735 GetTexImage, 163, 246, 247, 249, 250, 287, 294, 297, 532, 574, 595 GetTexLevelParameter, 243 GetTexLevelParameteriv, 598, 599 GetTexParameter, 242, 275, 296, 333, 574 GetTexParameterfv, 287, 596, 597 GetTexParameterI, 242 GetTexParameteriv, 287, 596, 597 GetTextureImage, 246, 246, 247, 249, 250, 742 GetTextureLevelParameter, 243 GetTextureLevelParameter*, 245 GetTextureParameter, 242 GetTextureParameter*, 243 GetTextureParameterI, 242 GetTextureParameteriv, 596 GetTextureSubImage, 248, 252, 742 GetTransformFeedbackVarying, 616 GetTransformFeedback*, 581 GetTransformFeedbacki64 v, 580, 581 GetTransformFeedbacki v, 580, 581 GetTransformFeedbackiv, 580, 581 GetTransformFeedbackVarying, 395, 396, 714 GetUniform, 615 GetUniformBlockIndex, 138 GetUniformdv, 173 GetUniformfv, 173 GetUniformIndices, 136, 710 GetUniformiv, 173 GetUniformLocation, 136, 158, 615 GetUniformSubroutineuiv, 174 GetUniformuiv, 173 GetVertexAttribPointerv, 585 GetVertexArrayIndexed64iv, 377, 378

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

795

GetVertexArrayIndexediv, 377, 377, 736 GetVertexArrayiv, 377 GetVertexAttribdv, 378, 379 GetVertexAttribfv, 378, 379, 625 GetVertexAttribIiv, 378, 379 GetVertexAttribIuiv, 378, 379 GetVertexAttribiv, 378, 379, 585, 586 GetVertexAttribLdv, 379, 379 GetVertexAttribPointerv, 380 gl , 107 GL APPLE flush buffer range, 755 GL APPLE vertex array object, 755 GL ARB arrays of arrays, 766 GL ARB base instance, 764 GL ARB bindless texture, 771 GL ARB blend func extended, 759 GL ARB buffer storage, 716 GL ARB buffer storage, 769 GL ARB cl event, 763 GL ARB clear buffer object, 766 GL ARB clear texture, 716 GL ARB clear texture, 770 GL ARB clip control, 718, 732, 772 GL ARB color buffer float, 752 GL ARB compatibility, 678, 756 GL ARB compressed texture pixel storage, 764 GL ARB compute shader, 766 GL ARB compute variable group size, 771 GL ARB conditional render inverted, 719, 731, 772 GL ARB conservative depth, 764 GL ARB copy buffer, 756 GL ARB copy image, 766 GL ARB cull distance, 718, 731, 773 GL ARB debug group, 765 GL ARB debug label, 766 GL ARB debug output, 763, 765

GL ARB debug output2, 765 GL ARB depth buffer float, 753 GL ARB depth clamp, 756 GL ARB depth texture, 749 GL ARB derivative control, 719, 732, 773 GL ARB direct state access, 719, 726, 731, 773 GL ARB draw buffers, 752 GL ARB draw buffers blend, 758 GL ARB draw elements base vertex, 757 GL ARB draw indirect, 761 GL ARB draw instanced, 753, 754 GL ARB enhanced layouts, 717 GL ARB enhanced layouts, 770 GL ARB ES2 compatibility, 762 GL ARB ES3 1 compatibility, 719, 732, 772 GL ARB ES3 2 compatibility, 775 GL ARB ES3 compatibility, 767 GL ARB explicit attrib location, 759 GL ARB explicit uniform location, 767 GL ARB fragment coord conventions, 757 GL ARB fragment layer viewport, 767 GL ARB fragment program, 750, 751 GL ARB fragment program shadow, 752 GL ARB fragment shader, 751 GL ARB fragment shader interlock, 776 GL ARB framebuffer no attachments, 767 GL ARB framebuffer object, 31, 753 GL ARB framebuffer sRGB, 754 GL ARB geometry shader4, 754 GL ARB get program binary, 762 GL ARB get texture sub image, 719,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

796

731, 773 GL ARB gl spirv, 734, 778 GL ARB gpu shader5, 761 GL ARB gpu shader fp64, 761 GL ARB gpu shader int64, 776 GL ARB half float pixel, 753 GL ARB half float vertex, 754 GL ARB indirect parameters, 733 GL ARB indirect parameters, 771 GL ARB instanced arrays, 754 GL ARB internalformat query, 764 GL ARB internalformat query2, 767 GL ARB invalidate subdata, 768 GL ARB map buffer alignment, 765 GL ARB map buffer range, 754 GL ARB matrix palette, 749 GL ARB multi bind, 717 GL ARB multi draw indirect, 768 GL ARB multi bind, 770 GL ARB multisample, 748 GL ARB multitexture, 748 GL ARB occlusion query, 751 GL ARB occlusion query2, 760 GL ARB parallel shader compile, 776 GL ARB pipeline statistics query, 733, 774 GL ARB pixel buffer object, 753 GL ARB point parameters, 748 GL ARB point sprite, 751 GL ARB polygon offset clamp, 734 GL ARB post depth coverage, 776 GL ARB program interface query, 768 GL ARB provoking vertex, 757 GL ARB query buffer object, 716 GL ARB query buffer object, 770 GL ARB robust buffer access behavior, 768 GL ARB robustness, 732, 763 GL ARB robustness isolation, 767, 769

GL ARB sample locations, 776 GL ARB sample shading, 758 GL ARB sampler objects, 760 GL ARB seamless cube map, 757 GL ARB seamless cubemap per texture, 771 GL ARB separate shader objects, 762 GL ARB shader atomic counter ops, 734, 777 GL ARB shader atomic counters, 765 GL ARB shader ballot, 777 GL ARB shader bit encoding, 760 GL ARB shader clock, 777 GL ARB shader draw parameters, 734 GL ARB shader group vote, 734 GL ARB shader image load store, 765 GL ARB shader image size, 768 GL ARB shader objects, 751 GL ARB shader precision, 762 GL ARB shader stencil export, 763 GL ARB shader storage buffer object, 768 GL ARB shader subroutine, 761 GL ARB shader texture image samples, 719, 732, 773 GL ARB shader texture lod, 756 GL ARB shader viewport layer array, 777 GL ARB shader draw parameters, 771 GL ARB shader group vote, 772 GL ARB shading language 100, 751 GL ARB shading language 420pack, 764 GL ARB shading language include, 759 GL ARB shading language packing, 765 GL ARB shadow, 750, 751 GL ARB shadow ambient, 750

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

797

GL ARB sparse buffer, 775 GL ARB sparse texture2, 777 GL ARB sparse texture clamp, 777 GL ARB sparse texture, 772 GL ARB spirv extensions, 734 GL ARB stencil texturing, 768 GL ARB sync, 757 GL ARB tessellation shader, 761 GL ARB texture barrier, 719, 732, 774 GL ARB texture border clamp, 748 GL ARB texture buffer object, 755 GL ARB texture buffer object rgb32, 761 GL ARB texture buffer range, 769 GL ARB texture compression, 748 GL ARB texture compression bptc, 759 GL ARB texture compression rgtc, 755 GL ARB texture cube map, 748 GL ARB texture cube map array, 758 GL ARB texture env add, 748 GL ARB texture env combine, 749 GL ARB texture env crossbar, 749 GL ARB texture env dot3, 749 GL ARB texture filter anisotropic, 734 GL ARB texture filter minmax, 778 GL ARB texture float, 753 GL ARB texture gather, 758 GL ARB texture mirror clamp to edge, 716 GL ARB texture mirrored repeat, 749 GL ARB texture multisample, 757 GL ARB texture non power of two, 751 GL ARB texture query levels, 769 GL ARB texture query lod, 759 GL ARB texture rectangle, 752 GL ARB texture rg, 755 GL ARB texture rgb10 a2ui, 760

GL ARB texture stencil8, 717 GL ARB texture storage, 765 GL ARB texture storage multisample, 769 GL ARB texture swizzle, 760 GL ARB texture view, 766 GL ARB texture mirror clamp to edge, 770 GL ARB texture stencil8, 770 GL ARB timer query, 760 GL ARB transform feedback2, 761 GL ARB transform feedback3, 762 GL ARB transform feedback instanced, 764 GL ARB transform feedback overflow query, 734, 775 GL ARB transpose matrix, 748 GL ARB uniform buffer object, 756 GL ARB vertex array bgra, 758 GL ARB vertex array object, 755 GL ARB vertex attrib 64bit, 762 GL ARB vertex attrib binding, 766 GL ARB vertex blend, 749 GL ARB vertex buffer object, 750 GL ARB vertex program, 750 GL ARB vertex shader, 751 GL ARB vertex type 10f 11f 11f rev, 716 GL ARB vertex type 2 10 10 10 rev, 760 GL ARB vertex type 10f 11f 11f rev, 770 GL ARB viewport array, 762 GL ARB window pos, 750 GL ARB name, 747 gl BackColor, 680 gl BackSecondaryColor, 680 gl BaseInstance, 388, 404 gl BaseVertex, 388, 404

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

798

gl ClipDistance, 405, 412, 413, 427, 429, 437, 772 gl ClipDistance[], 126, 435, 454 gl CullDistance, 405, 412, 413, 427, 429, 437, 454, 772 gl CullDistance[], 126, 435, 454 gl DrawID, 368, 372, 375, 376, 388, 404, 743 GL EXT framebuffer blit, 754 GL EXT framebuffer multisample, 754 GL EXT framebuffer object, 31, 754 GL EXT texture compression rgtc, 755 gl FragCoord, 475, 491, 737 gl FragCoord.z, 659 gl FragDepth, 493, 494, 659 gl FrontFacing, 492 gl in, 412, 427 gl in[], 435 gl InstanceID, 365, 370, 388, 404, 700 gl InvocationID, 411, 412, 414, 434 GL KHR blend equation advanced, 775 GL KHR blend equation advanced coherent, 775 GL KHR context flush control, 719, 774 GL KHR debug, 766 GL KHR no error, 734, 775, 778 GL KHR robust buffer access behavior, 774 GL KHR robustness, 719, 774 GL KHR texture compression astc ldr, 765, 778 GL KHR texture compression astc sliced 3d, 778 GL KHR name, 747 gl Layer, 335, 437, 438, 636, 777 gl MaxPatchVertices, 412, 413, 427,

428 gl NextBuffer, 106, 109, 115, 393 gl NumSamples, 492 gl NumWorkGroups, 546 GL NV depth buffer float, 753 GL NV half float, 754 GL OES sample shading, 725 GL OES sample variables, 725 gl out, 413 gl PatchVerticesIn, 412, 427 gl PerVertex, 125, 127, 128 gl PointCoord, 470 gl PointSize, 125, 127, 405, 412, 413, 427, 429, 435, 437, 468 gl Position, 391, 405, 412, 413, 427, 429, 435, 436, 457, 665 gl PrimitiveID, 412, 427, 437, 492 gl PrimitiveIDIn, 435 gl SampleID, 492, 493 gl SampleMask, 493, 494, 722, 723 gl SampleMaskIn, 493, 722, 725, 741 gl SampleMaskIn[], 776 gl SamplePosition, 492, 493 gl SkipComponents, 448 gl SkipComponents1, 106, 109, 115, 393 gl SkipComponents2, 106, 109, 115, 393 gl SkipComponents3, 106, 109, 115, 393 gl SkipComponents4, 106, 109, 115, 393 gl TessCoord, 417, 427, 662, 730 gl TessLevelInner, 413, 414, 428 gl TessLevelInner[1], 428 gl TessLevelOuter, 413, 414, 428 gl TessLevelOuter[2], 428 gl TessLevelOuter[3], 428 gl VertexID, 365, 370, 388, 404, 492, 700, 730

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

799

gl VerticesOut, 686 gl ViewportIndex, 437, 438, 458, 636, 777 gl MaxPatchVertices, 725 glBlitFramebuffer, 17 GlobalInvocationId, 671 GLSLPacked, 147 GLSLShared, 147 glSpecializeShader, 739 GLX ARB create context, 549, 755 GLX ARB create context no error, 778 GLX ARB create context profile, 678, 759 GLX ARB create context robustness, 763 GLX ARB fbconfig float, 752 GLX ARB framebuffer sRGB, 754 GLX ARB robustness application isolation, 769 GLX ARB robustness share group isolation, 769 GREATER, 239, 288, 504, 505 GREEN, 194, 240, 285, 490, 534, 536, 596, 598, 607 GREEN BITS, 682 GREEN INTEGER, 194 GUILTY CONTEXT RESET, 19

iimage2DMS, 119 iimage2DMSArray, 119 iimage2DRect, 119 iimage3D, 119 iimageBuffer, 119 iimageCube, 119 iimageCubeArray, 119 Image, 667, 668 Image1D, 672 image1D, 118 image1DArray, 118 image2D, 118 image2DArray, 118 image2DMS, 118 image2DMSArray, 118 image2DRect, 118 image3D, 118 IMAGE 1D, 118 IMAGE 1D ARRAY, 118 IMAGE 2D, 118 IMAGE 2D ARRAY, 118 IMAGE 2D MULTISAMPLE, 118 IMAGE 2D MULTISAMPLE ARRAY, 118 IMAGE 2D RECT, 118 IMAGE 3D, 118 IMAGE BINDING ACCESS, 627 IMAGE BINDING FORMAT, 627 IMAGE BINDING LAYER, 627 HALF FLOAT, 193, 297, 352, 535–537 IMAGE BINDING LAYERED, 627 HelperInvocation, 671 IMAGE BINDING LEVEL, 627 HIGH FLOAT, 172 IMAGE BINDING NAME, 627 HIGH INT, 172 IMAGE BUFFER, 118 Hint, 562, 683 IMAGE CLASS 10 10 10 2, 574 IMAGE CLASS 11 11 10, 574 if, 102, 415 IMAGE CLASS 1 X 16, 574 iimage1D, 118 IMAGE CLASS 1 X 32, 574 iimage1DArray, 119 IMAGE CLASS 1 X 8, 574 iimage2D, 118 IMAGE CLASS 2 X 16, 574 iimage2DArray, 119 IMAGE CLASS 2 X 32, 574 OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

800

IMAGE CLASS 2 X 8, 574 IMAGE CLASS 4 X 16, 574 IMAGE CLASS 4 X 32, 574 IMAGE CLASS 4 X 8, 574 IMAGE COMPATIBILITY CLASS, 574 IMAGE CUBE, 118 IMAGE CUBE MAP ARRAY, 118 IMAGE FORMAT COMPATIBILITY BY CLASS, 296, 574 IMAGE FORMAT COMPATIBILITY BY SIZE, 296, 574 IMAGE FORMAT COMPATIBILITY TYPE, 243, 296, 574, 597 IMAGE PIXEL FORMAT, 575 IMAGE PIXEL TYPE, 575 IMAGE TEXEL SIZE, 575 ImageBuffer, 672 imageBuffer, 118 imageCube, 118 ImageCubeArray, 672 imageCubeArray, 118 ImageGatherExtended, 672 ImageMSArray, 672 ImageQuery, 672 ImageRect, 672 IMPLEMENTATION COLOR READ FORMAT, 307, 333, 532, 655, 700 IMPLEMENTATION COLOR READ TYPE, 307, 333, 532, 655, 700 in, 428 INCR, 504 INCR WRAP, 504 Index, 495, 497 index, 495 IndexPointer, 679 INFO LOG LENGTH, 166, 167, 170,

171, 612–614 InitNames, 682 INNOCENT CONTEXT RESET, 19 Input, 128, 390, 667, 668 InstanceId, 671 INT, 116, 193, 245, 294, 297, 298, 310, 352, 537, 538, 576 int, 116, 127, 145, 386 Int64Atomics, 669 INT 2 10 10 10 REV, 348, 353, 354, 362 INT IMAGE 1D, 118 INT IMAGE 1D ARRAY, 119 INT IMAGE 2D, 118 INT IMAGE 2D ARRAY, 119 INT IMAGE 2D MULTISAMPLE, 119 INT IMAGE 2D MULTISAMPLE ARRAY, 119 INT IMAGE 2D RECT, 119 INT IMAGE 3D, 119 INT IMAGE BUFFER, 119 INT IMAGE CUBE, 119 INT IMAGE CUBE MAP ARRAY, 119 INT SAMPLER 1D, 117 INT SAMPLER 1D ARRAY, 117 INT SAMPLER 2D, 117 INT SAMPLER 2D ARRAY, 117 INT SAMPLER 2D MULTISAMPLE, 117 INT SAMPLER 2D MULTISAMPLE ARRAY, 117 INT SAMPLER 2D RECT, 118 INT SAMPLER 3D, 117 INT SAMPLER BUFFER, 118 INT SAMPLER CUBE, 117 INT SAMPLER CUBE MAP ARRAY, 117 INT VEC2, 116

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

801

INT VEC3, 116 INT VEC4, 116 INTENSITY, 681 INTERLEAVED ATTRIBS, 168, 177, 392–394, 616 InterleavedArrays, 679 INTERNALFORMAT ALPHA SIZE, 575 INTERNALFORMAT ALPHA TYPE, 575 INTERNALFORMAT BLUE SIZE, 575 INTERNALFORMAT BLUE TYPE, 575 INTERNALFORMAT DEPTH SIZE, 575 INTERNALFORMAT DEPTH TYPE, 575 INTERNALFORMAT GREEN SIZE, 575 INTERNALFORMAT GREEN TYPE, 575 INTERNALFORMAT PREFERRED, 575 INTERNALFORMAT RED SIZE, 575 INTERNALFORMAT RED TYPE, 575 INTERNALFORMAT SHARED SIZE, 575 INTERNALFORMAT STENCIL SIZE, 575 INTERNALFORMAT STENCIL TYPE, 575 INTERNALFORMAT SUPPORTED, 569, 576 interpolateAtOffset, 652 interpolateAtCentroid, 489 interpolateAtOffset, 489 interpolateAtSample, 489 InterpolationFunction, 672

INVALID ENUM, 17, 18, 37, 41, 44– 46, 48, 50, 51, 62, 63, 65, 70, 72, 75, 77, 78, 81–83, 89, 92, 100, 101, 108–110, 120, 121, 132, 157, 167, 169, 170, 173– 175, 179, 180, 187–189, 214, 215, 218, 224, 227, 229, 232, 234, 235, 241–243, 245, 247, 267, 276, 278–280, 287, 302, 306, 308, 311–313, 315, 317, 319–322, 331, 346, 349, 353, 354, 365, 370, 377, 378, 380, 383, 393, 418, 444, 445, 454, 467, 469, 477, 478, 508, 510, 517, 520, 526, 527, 531, 533, 541, 543, 554–557, 561, 563, 565–567, 569, 580, 581, 708, 724, 728, 730, 742 INVALID FRAMEBUFFER OPERATION, 18, 219, 225, 332, 533, 541 INVALID INDEX, 106, 109, 174 INVALID OPERATION, 17, 18, 45, 46, 50–52, 62, 63, 65, 67, 70– 72, 75–83, 90–95, 97, 99, 100, 108–110, 120, 121, 123–125, 130–132, 142, 144, 149, 153, 157, 158, 167, 169, 170, 172, 174, 175, 180, 182, 185–188, 190, 192, 195, 205, 215, 218, 219, 223–225, 227, 229, 230, 232–235, 241, 243, 245–252, 267, 274–284, 292, 302, 303, 306, 308, 311–313, 315, 317, 319, 320, 322, 323, 331, 351– 360, 366, 367, 369, 370, 374, 376–378, 380, 382, 383, 387– 389, 393, 405–408, 410, 430, 431, 438, 443–451, 495, 497, 512, 517, 520, 525, 527, 530–

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

802

534, 536, 541, 543, 544, 547, 548, 554, 581, 678–680, 695, 698, 699, 722, 724, 737, 742 INVALID VALUE, 17, 18, 37–39, 41, 44–46, 48, 50, 59, 60, 63, 65, 67, 70–72, 76, 78, 80, 81, 83, 84, 90–92, 94, 95, 97, 98, 100, 101, 108–110, 120–125, 130, 132, 142, 143, 149, 153, 157, 164, 166, 167, 169–172, 174, 175, 180, 182–187, 189, 214, 215, 219, 224, 225, 227, 232, 234, 235, 241, 246–248, 250– 252, 274, 276, 282, 283, 291, 292, 302, 304, 306, 313–316, 320, 322, 323, 346, 349–351, 353–355, 357–360, 366, 367, 369, 370, 373–380, 382, 387– 389, 393, 407, 443, 444, 447, 451, 459–461, 467, 469, 471, 484, 485, 487, 495, 507, 508, 510, 515, 520, 522, 523, 526, 527, 531, 541, 543, 544, 547, 548, 554–558, 560, 561, 565, 566, 569, 580, 581, 678, 680, 682, 695, 696, 698, 711, 713, 724, 728, 730, 742 InvalidateBufferData, 80 InvalidateBufferSubData, 80, 80 InvalidateFramebuffer, 527 InvalidateNamedFramebufferData, 527, 528 InvalidateNamedFramebufferSubData, 526, 722 InvalidateSubFramebuffer, 526, 527, 695, 722, 726 InvalidateTexImage, 282 InvalidateTexSubImage, 281, 281, 282, 742 inversesqrt(), 673

INVERT, 504, 515 Invocation, 667 InvocationId, 671 IS PER PATCH, 111, 113, 623 IS ROW MAJOR, 111, 113, 138, 623 isampler1D, 117 isampler1DArray, 117 isampler2D, 117 isampler2DArray, 117 isampler2DMS, 117 isampler2DMSArray, 117 isampler2DRect, 118 isampler3D, 117 isamplerBuffer, 118 isamplerCube, 117 isamplerCubeArray, 117 IsBuffer, 60, 60 IsEnabled, 485, 513, 565, 565, 582, 587, 589, 591–593, 602, 603, 625, 649, 656 IsEnabledi, 485, 507, 513, 566, 566, 602, 603 IsFramebuffer, 304, 305 IsList, 683 ISOLINES, 168 isolines, 415, 416, 424, 427 IsProgram, 100, 100 IsProgramPipeline, 122, 122 IsQuery, 47, 47 IsRenderbuffer, 314, 314 IsSampler, 184, 187, 188 IsShader, 91, 91 IsSync, 41, 41 IsTexture, 183, 183, 686 IsTransformFeedback, 443, 443 IsVertexArray, 351, 351 ivec2, 116, 386 ivec3, 116, 386 ivec4, 116, 294, 295, 386

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

803

KEEP, 504, 602 KHR debug, 693 KHR debug output, 701 KHR no error, 744 LAST VERTEX CONVENTION, 438, 453, 590 Layer, 671 LAYER PROVOKING VERTEX, 438, 636 Layered images, 307 layout, 106, 107, 114, 115, 126, 127, 144, 146, 147, 149, 151, 152, 169, 294–296, 385, 386, 391– 395, 410, 413, 415, 416, 418, 421, 424, 431–433, 492, 495, 496, 498, 547, 722 LEFT, 507, 517, 518, 520, 521, 530 LEQUAL, 239, 285, 288, 504, 505, 597, 600 LESS, 239, 288, 504, 505, 602 LIGHTING, 680 LINE, 480, 482, 592 LINE LOOP, 339, 431, 438, 446 LINE SMOOTH, 471, 476, 591 LINE SMOOTH HINT, 563, 633 LINE STIPPLE, 680 LINE STRIP, 168, 339, 431, 432, 438, 446 LINE STRIP ADJACENCY, 343, 431, 438 LINE WIDTH, 591 LINEAR, 219, 240, 242, 254, 260, 263– 266, 268, 269, 285, 310, 398, 539, 541, 571, 573, 596, 600, 714 LINEAR MIPMAP LINEAR, 240, 264, 266 LINEAR MIPMAP NEAREST, 240, 264, 265

LINES, 168, 339, 431, 438, 445, 446 lines, 431 LINES ADJACENCY, 168, 341, 431, 438 lines adjacency, 431 LineStipple, 680 LineWidth, 471, 678, 680 LINK STATUS, 96, 131, 167, 614, 728 LinkProgram, 93, 96, 97–100, 103, 106, 126, 131, 132, 149, 158, 167, 387, 394, 405, 436, 449, 496, 700, 728, 736 ListBase, 683 LoadIdentity, 679 LoadMatrix, 679 LoadName, 682 LoadTransposeMatrix, 679 LocalInvocationId, 671 LocalInvocationIndex, 671 LocalSize, 669 LOCATION, 112, 114, 121, 496, 623 Location, 129, 384, 390, 495, 497 location, 107, 126, 127, 384, 390, 495, 721 LOCATION COMPONENT, 112, 114, 496, 623 LOCATION INDEX, 112, 114, 121, 496, 623 log(), 673 log2(), 673 LOGIC OP MODE, 603 Logical, 667 LogicOp, 514, 515 LOSE CONTEXT ON RESET, 20 LOW FLOAT, 172 LOW INT, 172 LOWER LEFT, 454, 456, 458, 469, 470, 477, 589 LUMINANCE, 681 LUMINANCE8, 725

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

804

LUMINANCE ALPHA, 681 main, 415 MAJOR VERSION, 567, 638 MANUAL GENERATE MIPMAP, 571 MAP COHERENT BIT, 66, 67, 74, 76, 709, 723, 727 MAP FLUSH EXPLICIT BIT, 74, 76– 78 MAP INVALIDATE BUFFER BIT, 74, 76 MAP INVALIDATE RANGE BIT, 74, 76 MAP PERSISTENT BIT, 66, 67, 71– 73, 76, 77, 79–81, 83 MAP READ BIT, 66–69, 73–77 MAP UNSYNCHRONIZED BIT, 75, 76 MAP WRITE BIT, 66–69, 73, 75–77, 709 MAP COHERENT BIT, 74 MapBuffer, 62, 71, 72, 76, 77, 80, 450 MapBufferRange, 66, 71, 72, 73, 73, 75, 77, 80, 709, 728 MapNamedBuffer, 76, 77 MapNamedBufferRange, 66, 73, 73, 75, 77, 727 matC, 145 matCxR, 145 mat2, 116, 386 mat2x3, 116, 386 mat2x4, 116, 386 mat3, 116, 386 mat3x2, 116, 386 mat3x4, 116, 386 mat4, 116, 386 mat4x2, 116, 386 mat4x3, 116, 386 Matrix, 672

MATRIX STRIDE, 111, 114, 138, 145, 623 MatrixMode, 679 MatrixStride, 147, 148 MAX, 509 MAX 3D TEXTURE SIZE, 213, 244, 321–323, 635 MAX ARRAY TEXTURE LAYERS, 214, 234, 323, 635, 706 MAX ATOMIC COUNTER BUFFER BINDINGS, 85, 150, 646 MAX ATOMIC COUNTER BUFFER SIZE, 646, 688 MAX ATTRIB STACK DEPTH, 683 MAX CLIENT ATTRIB STACK DEPTH, 683 MAX CLIP DISTANCES, 454, 635 MAX COLOR ATTACHMENTS, 301, 312, 319, 320, 322, 332, 518, 527, 531, 652 MAX COLOR TEXTURE SAMPLES, 578, 651 MAX COMBINED ATOMIC COUNTER BUFFERS, 150, 406, 646, 706 MAX COMBINED ATOMIC COUNTERS, 403, 406, 646 MAX COMBINED CLIP AND CULL DISTANCES, 454, 635 MAX COMBINED COMPUTE UNIFORM COMPONENTS, 134, 644 MAX COMBINED DIMENSIONS, 576 MAX COMBINED FRAGMENT UNIFORM COMPONENTS, 134, 648 MAX COMBINED GEOMETRY UNIFORM COMPONENTS,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

805

134, 648 MAX COMBINED IMAGE UNIFORMS, 403, 406, 647, 688 MAX COMBINED IMAGE UNITS AND FRAGMENT OUTPUTS, 694 MAX COMBINED SHADER OUTPUT RESOURCES, 298, 406, 647, 694 MAX COMBINED SHADER STORAGE BLOCKS, 152, 404, 406, 646 MAX COMBINED TESS CONTROL UNIFORM COMPONENTS, 134, 648 MAX COMBINED TESS EVALUATION UNIFORM COMPONENTS, 134, 648 MAX COMBINED TEXTURE IMAGE UNITS, 142, 179, 185, 401, 406, 645, 698 MAX COMBINED UNIFORM BLOCKS, 144, 406, 645, 700 MAX COMBINED VERTEX UNIFORM COMPONENTS, 134, 648 MAX COMPUTE ATOMIC COUNTER BUFFERS, 150, 644 MAX COMPUTE ATOMIC COUNTERS, 403, 644 MAX COMPUTE IMAGE UNIFORMS, 403, 644 MAX COMPUTE SHADER STORAGE BLOCKS, 152, 404, 644 MAX COMPUTE SHARED MEMORY SIZE, 548, 644, 700 MAX COMPUTE TEXTURE IM-

AGE UNITS, 401, 644 MAX COMPUTE UNIFORM BLOCKS, 144, 644, 700 MAX COMPUTE UNIFORM COMPONENTS, 134, 644, 727 MAX COMPUTE WORK GROUP COUNT, 547, 644 MAX COMPUTE WORK GROUP INVOCATIONS, 547, 644 MAX COMPUTE WORK GROUP SIZE, 547, 644 MAX CUBE MAP TEXTURE SIZE, 214, 244, 274, 321, 323, 635 MAX CULL DISTANCES, 454, 635 MAX DEBUG GROUP STACK DEPTH, 556, 650 MAX DEBUG LOGGED MESSAGES, 553, 650 MAX DEBUG MESSAGE LENGTH, 552, 555, 556, 650 MAX DEPTH, 577 MAX DEPTH TEXTURE SAMPLES, 578, 651 MAX DRAW BUFFERS, 495, 496, 507, 508, 510, 519, 520, 522, 526, 652 MAX DUAL SOURCE DRAW BUFFERS, 495, 496, 510, 512, 652 MAX ELEMENT INDEX, 370, 635 MAX ELEMENTS INDICES, 372, 373, 636 MAX ELEMENTS VERTICES, 372, 373, 636 MAX FRAGMENT ATOMIC COUNTER BUFFERS, 150, 643 MAX FRAGMENT ATOMIC COUNTERS, 403, 643

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

806

MAX FRAGMENT IMAGE UNIFORMS, 403, 647, 688 MAX FRAGMENT INPUT COMPONENTS, 493, 643 MAX FRAGMENT INTERPOLATION OFFSET, 489, 652, 735 MAX FRAGMENT SHADER STORAGE BLOCKS, 152, 404, 643 MAX FRAGMENT UNIFORM BLOCKS, 144, 643, 700 MAX FRAGMENT UNIFORM COMPONENTS, 133, 134, 643 MAX FRAGMENT UNIFORM VECTORS, 133, 643 MAX FRAMEBUFFER HEIGHT, 306, 328, 651 MAX FRAMEBUFFER LAYERS, 306, 328, 651 MAX FRAMEBUFFER SAMPLES, 306, 328, 651 MAX FRAMEBUFFER WIDTH, 306, 328, 651 MAX GEOMETRY ATOMIC COUNTER BUFFERS, 150, 642 MAX GEOMETRY ATOMIC COUNTERS, 403, 642 MAX GEOMETRY IMAGE UNIFORMS, 403, 647 MAX GEOMETRY INPUT COMPONENTS, 436, 642 MAX GEOMETRY OUTPUT COMPONENTS, 437, 642 MAX GEOMETRY OUTPUT VERTICES, 436, 642 MAX GEOMETRY SHADER INVO-

CATIONS, 642 MAX GEOMETRY SHADER STORAGE BLOCKS, 152, 404, 642 MAX GEOMETRY TEXTURE IMAGE UNITS, 400, 642 MAX GEOMETRY TOTAL OUTPUT COMPONENTS, 436, 642 MAX GEOMETRY UNIFORM BLOCKS, 144, 642, 700 MAX GEOMETRY UNIFORM COMPONENTS, 134, 642, 727 MAX HEIGHT, 577 MAX IMAGE SAMPLES, 294, 647 MAX IMAGE UNITS, 142, 158, 290, 291, 647 MAX INTEGER SAMPLES, 316, 578, 651 MAX LABEL LENGTH, 558, 650 MAX LAYERS, 577 MAX NAME LENGTH, 107, 108, 110, 622 MAX NUM ACTIVE VARIABLES, 108, 622 MAX NUM COMPATIBLE SUBROUTINES, 108, 622 MAX PATCH VERTICES, 346, 640 MAX PROGRAM TEXEL OFFSET, 257, 645 MAX PROGRAM TEXTURE GATHER OFFSET, 257, 263, 643 MAX RECTANGLE TEXTURE SIZE, 214, 637 MAX RENDERBUFFER SIZE, 316, 635 MAX SAMPLE MASK WORDS, 487, 651 MAX SAMPLES, 316, 579, 651

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

807

MAX SERVER WAIT TIMEOUT, 39, 651 MAX SHADER STORAGE BLOCK SIZE, 152, 646 MAX SHADER STORAGE BUFFER BINDINGS, 85, 153, 646 MAX SUBROUTINE UNIFORM LOCATIONS, 154, 645 MAX SUBROUTINES, 154, 645 MAX TESS CONTROL ATOMIC COUNTER BUFFERS, 150, 640 MAX TESS CONTROL ATOMIC COUNTERS, 402, 640 MAX TESS CONTROL IMAGE UNIFORMS, 403, 647 MAX TESS CONTROL INPUT COMPONENTS, 413, 640 MAX TESS CONTROL OUTPUT COMPONENTS, 414, 640 MAX TESS CONTROL SHADER STORAGE BLOCKS, 152, 404, 640 MAX TESS CONTROL TEXTURE IMAGE UNITS, 400, 640 MAX TESS CONTROL TOTAL OUTPUT COMPONENTS, 414, 640, 725 MAX TESS CONTROL UNIFORM BLOCKS, 144, 640, 700 MAX TESS CONTROL UNIFORM COMPONENTS, 134, 640 MAX TESS EVALUATION ATOMIC COUNTER BUFFERS, 150, 641 MAX TESS EVALUATION ATOMIC COUNTERS, 403,

641 MAX TESS EVALUATION IMAGE UNIFORMS, 403, 647 MAX TESS EVALUATION INPUT COMPONENTS, 428, 641 MAX TESS EVALUATION OUTPUT COMPONENTS, 429, 641 MAX TESS EVALUATION SHADER STORAGE BLOCKS, 152, 404, 641 MAX TESS EVALUATION TEXTURE IMAGE UNITS, 400, 641 MAX TESS EVALUATION UNIFORM BLOCKS, 144, 641, 700 MAX TESS EVALUATION UNIFORM COMPONENTS, 134, 641 MAX TESS GEN LEVEL, 416, 640 MAX TESS PATCH COMPONENTS, 414, 428, 640 MAX TEXTURE BUFFER SIZE, 236, 637 MAX TEXTURE COORDS, 682 MAX TEXTURE IMAGE UNITS, 401, 643 MAX TEXTURE LOD BIAS, 257, 635 MAX TEXTURE MAX ANISOTROPY, 255, 635 MAX TEXTURE SIZE, 214, 234, 244, 321, 323, 635 MAX TEXTURE UNITS, 682 MAX TRANSFORM FEEDBACK BUFFERS, 85, 393, 654 MAX TRANSFORM FEEDBACK INTERLEAVED COMPO-

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

808

NENTS, 394, 395, 654 MAX TRANSFORM FEEDBACK SEPARATE ATTRIBS, 393, 654 MAX TRANSFORM FEEDBACK SEPARATE COMPONENTS, 394, 654 MAX UNIFORM BLOCK SIZE, 144, 645, 726 MAX UNIFORM BUFFER BINDINGS, 85, 148, 149, 645, 696 MAX UNIFORM LOCATIONS, 135, 645 MAX VARYING COMPONENTS, 391, 645, 679 MAX VARYING FLOATS, 679 MAX VARYING VECTORS, 391, 645 MAX VERTEX ATOMIC COUNTER BUFFERS, 150, 639 MAX VERTEX ATOMIC COUNTERS, 402, 639 MAX VERTEX ATTRIB BINDINGS, 355, 357, 360, 378, 380, 637 MAX VERTEX ATTRIB RELATIVE OFFSET, 354, 637 MAX VERTEX ATTRIB STRIDE, 355, 357, 358, 637, 705, 714 MAX VERTEX ATTRIBS, 347–349, 353, 357, 359, 360, 378–380, 387–389, 639 MAX VERTEX IMAGE UNIFORMS, 403, 647 MAX VERTEX OUTPUT COMPONENTS, 391, 413, 415, 429, 436, 437, 493, 639 MAX VERTEX SHADER STORAGE BLOCKS, 152, 404, 639 MAX VERTEX STREAMS, 42, 45,

46, 48, 451, 642 MAX VERTEX TEXTURE IMAGE UNITS, 400, 639 MAX VERTEX UNIFORM BLOCKS, 144, 639, 700 MAX VERTEX UNIFORM COMPONENTS, 133, 134, 639 MAX VERTEX UNIFORM VECTORS, 133, 639 MAX VIEWPORT DIMS, 461, 528, 636 MAX VIEWPORTS, 458–460, 485, 486, 636 MAX WIDTH, 577 MEDIUM FLOAT, 172 MEDIUM INT, 172 Memory Semantics, 668 MemoryBarrier, 67, 161, 164–166, 739 memoryBarrier, 160, 161, 165 MemoryBarrierByRegion, 165, 166 MIN, 509 MIN/MAG, 571 MIN FRAGMENT INTERPOLATION OFFSET, 489, 652 MIN MAP BUFFER ALIGNMENT, 74, 77, 637, 688 MIN PROGRAM TEXEL OFFSET, 257, 645 MIN PROGRAM TEXTURE GATHER OFFSET, 257, 263, 643 MIN SAMPLE SHADING VALUE, 468, 593 MINOR VERSION, 567, 638 MinSampleShading, 468, 741 MIPMAP, 578 MIRROR CLAMP TO EDGE, 185, 240, 242, 260, 713, 729 MIRRORED REPEAT, 185, 240, 242,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

809

260 MultiDrawArrays, 367, 699 MultiDrawArraysIndirect, 364, 365, 368, 369, 699, 730 MultiDrawArraysIndirectCount, 369 MultiDrawElements, 364, 372, 699, 730 MultiDrawElementsBaseVertex, 364, 376, 699 MultiDrawElementsIndirect, 364, 365, 374, 375, 376, 699, 730 MultiDrawElementsIndirectCount, 375 MULTISAMPLE, 467, 468, 471, 476, 482, 486, 489, 501, 515, 516, 593 MultiViewport, 672 MultMatrix, 680 MultTransposeMatrix, 680

NAND, 515 NEAREST, 240, 242, 254, 259, 263– 266, 268, 269, 287, 398, 539, 541, 571 NEAREST MIPMAP LINEAR, 240, 264, 266, 285, 600 NEAREST MIPMAP NEAREST, 240, 264, 265, 269, 287, 571, 726 NEGATIVE ONE TO ONE, 454, 456, 458, 589 NEVER, 239, 288, 504, 505 NewList, 683 NICEST, 563 NO ERROR, 16, 18, 19 NO RESET NOTIFICATION, 19 NONE, 106, 225, 239, 245, 285–287, 309–311, 324, 328, 396, 401, Name, 94 402, 501, 512, 515, 517–521, NAME LENGTH, 112, 114, 138, 139, 524, 526, 530–533, 568, 570, 624 573–576, 578–580, 596–598, NamedBufferData, 68, 68–70 600, 607, 626, 695, 707, 715, NamedBufferStorage, 65, 65, 67, 68 720, 738 NamedBufferSubData, 70, 70 NOOP, 515 NamedFramebufferDrawBuffer, 517, NoPerspective, 668 517 noperspective, 456 NamedFramebufferDrawBuffers, 519, NOR, 515 519, 520 NORMALIZE, 680 NamedFramebufferParameteri, 305, NormalPointer, 679 305, 306 NOTEQUAL, 239, 288, 504, 505 NamedFramebufferReadBuffer, 529, NULL, 552, 557, 560, 585, 588, 649 529, 530 NUM ACTIVE VARIABLES, 111, NamedFramebufferRenderbuffer, 318, 113, 114, 139, 140, 624 318, 319 NUM COMPATIBLE SUBROUNamedFramebufferTexture, 320, 320, TINES, 111, 120, 156, 620, 322, 323 624, 739 NamedFramebufferTextureLayer, 322 NUM COMPRESSED TEXNamedRenderbufferStorage, 316 TURE FORMATS, 206, 637, NamedRenderbufferStorageMultisample, 679, 685 315, 315 NUM EXTENSIONS, 568, 638 OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

810

NUM PROGRAM BINARY FORMATS, 130, 132, 637 NUM SAMPLE COUNTS, 578, 653 NUM SHADER BINARY FORMATS, 87, 92, 637 NUM SHADING LANGUAGE VERSIONS, 568, 638, 696 NUM SPIR V EXTENSIONS, 569, 638 NumWorkgroups, 671

OpFUnordEqual, 673 OpFUnordGreaterThan, 673 OpFUnordGreaterThanEqual, 673 OpFUnordLessThan, 673 OpFUnordLessThanEqual, 673 OpImageQueryLevels, 667, 741 OpImageQuerySizeLod, 667, 741 OpImageWrite, 668, 737 OpMemberName, 106 OpName, 106 OpSampledImage, 667 OBJECT TYPE, 37, 41, 632 OpSMod, 670 ObjectLabel, 557, 558, 560 OpSRem, 670 ObjectPtrLabel, 558 OpTypeBool, 669, 737 OFFSET, 112, 114, 138, 624 OpTypeFloat, 498 Offset, 147, 148, 391, 395, 668, 669 OpTypeImage, 667 ONE, 240, 490, 509, 511, 512, 603 OpTypeInt, 498 ONE MINUS CONSTANT ALPHA, OpTypeRuntimeArray, 668, 742 511 OpTypeStruct, 668 ONE MINUS CONSTANT COLOR, OR, 515 511 OR INVERTED, 515 ONE MINUS DST ALPHA, 511 OR REVERSE, 515 ONE MINUS DST COLOR, 511 Ortho, 680 ONE MINUS SRC1 ALPHA, 510, 511 out, 414 510, OUT OF MEMORY, 16–19, 706, 709, ONE MINUS SRC1 COLOR, 511 742 ONE MINUS SRC ALPHA, 511 Output, 128, 497, 667 ONE MINUS SRC COLOR, 511 PACK ALIGNMENT, 532, 611 Op*Name, 106 OpCapability, 666, 668, 672 PACK COMPRESSED BLOCK OpEntryPoint, 94, 128, 390, 497 DEPTH, 251, 252, 532, 611 OpFAdd, 673 PACK COMPRESSED BLOCK OpFDiv, 673 HEIGHT, 251, 252, 532, 611 OpFMul, 673 PACK COMPRESSED BLOCK SIZE, OpFOrdEqual, 673 251, 532, 611 OpFOrdGreaterThan, 673 PACK COMPRESSED BLOCK OpFOrdGreaterThanEqual, 673 WIDTH, 251, 252, 532, 611 PACK IMAGE OpFOrdLessThan, 673 HEIGHT, 247, 250, 251, 532, OpFOrdLessThanEqual, 673 OpFSub, 673 611 OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

811

PACK LSB FIRST, 532, 611, 679, 693, 701 PACK ROW LENGTH, 250, 251, 532, 611 PACK SKIP IMAGES, 247, 250, 251, 532, 611 PACK SKIP PIXELS, 250, 251, 532, 611 PACK SKIP ROWS, 250, 251, 532, 611 PACK SWAP BYTES, 532, 611 PARAMETER BUFFER, 61, 162, 369, 375, 376, 739 PARAMETER BUFFER BINDING, 587 PassThrough, 682 Patch, 668 patch, 113, 410 patch in, 428 patch out, 414 PATCH DEFAULT INNER LEVEL, 416, 418, 584 PATCH DEFAULT OUTER LEVEL, 416, 418, 584 PATCH VERTICES, 345, 346, 584 PATCHES, 345, 361, 410, 635 PatchParameterfv, 416 PatchParameteri, 345, 426 PatchVertices, 671 PauseTransformFeedback, 445, 446 PERSPECTIVE CORRECTION HINT, 683 PIXEL BUFFER BARRIER BIT, 162 PIXEL PACK BUFFER, 61, 162, 190, 531 PIXEL PACK BUFFER BINDING, 247, 536, 611 PIXEL UNPACK BUFFER, 61, 162, 190

PIXEL UNPACK BUFFER BINDING, 192, 226, 610 PixelStore, 189, 532, 544, 713 PixelStore*, 189 PixelZoom, 681 POINT, 480, 482, 592 POINT FADE THRESHOLD SIZE, 469, 591 point mode, 418 POINT SIZE, 591 POINT SIZE GRANULARITY, 636 POINT SIZE RANGE, 636 POINT SMOOTH, 680 POINT SMOOTH HINT, 683 POINT SPRITE, 680 POINT SPRITE COORD ORIGIN, 469, 470, 591 PointCoord, 671 Pointer, 669 PointParameter, 469 POINTS, 168, 339, 431, 432, 438, 445, 446, 480 points, 431, 434 PointSize, 469, 671 POLYGON, 681 POLYGON MODE, 592 POLYGON OFFSET CLAMP, 482, 592 POLYGON OFFSET FACTOR, 592 POLYGON OFFSET FILL, 482, 592 POLYGON OFFSET LINE, 482, 592 POLYGON OFFSET POINT, 482, 592 POLYGON OFFSET UNITS, 592 POLYGON SMOOTH, 477, 482, 592 POLYGON SMOOTH HINT, 563, 633 POLYGON STIPPLE, 681 PolygonMode, 480, 480, 482, 483, 681 PolygonOffset, 481, 481 PolygonOffsetClamp, 480, 481 PolygonStipple, 681

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

812

PopAttrib, 683 PopClientAttrib, 683 PopDebugGroup, 556, 556 PopMatrix, 680 PopName, 682 Position, 671 pow(x,y), 673 PRIMITIVE RESTART, 361, 587 PRIMITIVE RESTART FIXED INDEX, 361, 587, 715 PRIMITIVE RESTART FOR PATCHES SUPPORTED, 361, 635, 711 PRIMITIVE RESTART INDEX, 587 PrimitiveId, 671 PrimitiveRestartIndex, 361, 361 PRIMITIVES GENERATED, 42, 45, 46, 48, 451 PRIMITIVES SUBMITTED, 42, 48, 383 PrioritizeTextures, 682 Private, 667 PROGRAM, 557 PROGRAM BINARY FORMATS, 132, 637 PROGRAM BINARY LENGTH, 130, 131, 614 PROGRAM BINARY RETRIEVABLE HINT, 99, 100, 132, 169, 614 PROGRAM INPUT, 103, 106, 107, 111, 112, 120, 388, 389, 700 PROGRAM OUTPUT, 103, 106, 107, 111, 112, 120, 496, 497, 700 PROGRAM PIPELINE, 557 PROGRAM PIPELINE BINDING, 614 PROGRAM POINT SIZE, 435, 469, 625 PROGRAM SEPARABLE, 99–101,

124, 131, 169, 406, 614, 687, 736 ProgramBinary, 98–100, 131, 131, 132, 167, 449, 700, 705, 736 ProgramParameteri, 99, 132 ProgramUniform, 143 ProgramUniform{1234}ui, 143 ProgramUniform{1234}uiv, 143 ProgramUniformMatrix{234}, 143 ProgramUniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}, 143 PROVOKING VERTEX, 438, 590 ProvokingVertex, 438, 452 PROXY TEXTURE 1D, 205, 216, 244, 276, 286 PROXY TEXTURE 1D ARRAY, 205, 216, 244, 277, 278, 286 PROXY TEXTURE 2D, 205, 216, 244, 277, 286 PROXY TEXTURE 2D ARRAY, 203, 205, 244, 279, 286 PROXY TEXTURE 2D MULTISAMPLE, 205, 233, 244, 280, 286 PROXY TEXTURE 2D MULTISAMPLE ARRAY, 205, 233, 244, 280, 281, 286 PROXY TEXTURE 3D, 203, 244, 278, 279, 286 PROXY TEXTURE CUBE MAP, 205, 216, 244, 277, 286 PROXY TEXTURE CUBE MAP ARRAY, 203, 205, 214, 244, 279, 286 PROXY TEXTURE RECTANGLE, 205, 216, 227, 232, 244, 277, 286 PushAttrib, 683 PushClientAttrib, 683 PushDebugGroup, 555, 556

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

813

PushMatrix, 680 PushName, 682 QUAD STRIP, 681 QUADS, 168, 619, 681 quads, 415, 416, 418, 421, 427 QUERY, 557 QUERY BUFFER, 61, 163 QUERY BUFFER BARRIER BIT, 163 QUERY BUFFER BINDING, 656 QUERY BY REGION NO WAIT, 382, 383 QUERY BY REGION NO WAIT INVERTED, 382, 383 QUERY BY REGION WAIT, 382, 383 QUERY BY REGION WAIT INVERTED, 382, 383 QUERY COUNTER BITS, 48, 49, 651, 743 QUERY NO WAIT, 382, 383 QUERY NO WAIT INVERTED, 382, 383 QUERY RESULT, 44, 45, 49, 50, 626 QUERY RESULT AVAILABLE, 20, 44, 45, 49, 50, 626, 712, 720 QUERY RESULT NO WAIT, 50 QUERY TARGET, 50, 626 QUERY WAIT, 382, 383 QUERY WAIT INVERTED, 382, 383 QueryCounter, 43, 51, 51 R11F G11F B10F, 207, 209, 273, 295, 297 r11f g11f b10f, 295 R16, 208, 237, 273, 296, 298 r16, 296 R16 SNORM, 208, 273, 296, 298 r16 snorm, 296

R16F, 209, 237, 273, 295, 297 r16f, 295 R16I, 209, 237, 273, 295, 298 r16i, 295 R16UI, 209, 237, 273, 295, 297 r16ui, 295 R32F, 209, 237, 273, 295, 297 r32f, 295 R32I, 209, 237, 273, 295, 298 r32i, 295 R32UI, 209, 237, 273, 295, 297 r32ui, 295 R3 G3 B2, 208 R8, 208, 237, 273, 285, 291, 292, 296, 298, 598, 627 r8, 296 R8 SNORM, 208, 273, 296, 298 r8 snorm, 296 R8I, 209, 237, 273, 295, 298 r8i, 295 R8UI, 209, 237, 273, 295, 297 r8ui, 295 RASTERIZER DISCARD, 332, 457, 464, 591 READ BUFFER, 530, 606 READ FRAMEBUFFER, 301, 302, 304–308, 311, 318–321, 331, 526, 527, 529, 538, 605 READ FRAMEBUFFER BINDING, 219, 225, 305, 333, 531, 533, 534, 605 READ ONLY, 61, 75–77, 291, 292, 627 READ PIXELS, 571 READ PIXELS FORMAT, 578 READ PIXELS TYPE, 578 READ WRITE, 61, 68, 75, 77, 291, 292, 588 ReadBuffer, 518, 529, 529, 715 ReadnPixels, 531, 531 ReadPixels, 162, 188, 189, 196, 218,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

814

247, 332, 335, 450, 529, 530, 531, 531–534, 536, 538, 578, 681 RECTANGLE TEXTURE, 742 RED, 194, 205, 208, 209, 212, 240, 249, 252, 253, 285, 297, 298, 327, 490, 491, 534, 536, 596, 598, 607 RED BITS, 682 RED INTEGER, 194, 297, 298 REFERENCED BY COMPUTE SHADER, 112, 115, 139, 140, 624 REFERENCED BY FRAGMENT SHADER, 112, 115, 139, 140, 624 REFERENCED BY GEOMETRY SHADER, 112, 115, 139, 140, 624 REFERENCED BY TESS CONTROL SHADER, 112, 115, 139, 140, 624 REFERENCED BY TESS EVALUATION SHADER, 112, 115, 139, 140, 624 REFERENCED BY VERTEX SHADER, 112, 115, 139, 140, 624 Relaxed, 668 ReleaseShaderCompiler, 90, 90 RENDERBUFFER, 309, 310, 312–315, 317–319, 333, 542, 543, 557, 570, 578, 579, 608 RENDERBUFFER ALPHA SIZE, 317, 609 313, RENDERBUFFER BINDING, 608 RENDERBUFFER BLUE SIZE, 317, 609 RENDERBUFFER DEPTH SIZE,

317, 609 RENDERBUFFER GREEN SIZE, 317, 609 RENDERBUFFER HEIGHT, 315, 317, 609 RENDERBUFFER INTERNAL FORMAT, 315, 317, 609 RENDERBUFFER RED SIZE, 317, 609 RENDERBUFFER SAMPLES, 312, 315, 317, 329, 609 RENDERBUFFER STENCIL SIZE, 317, 609 RENDERBUFFER WIDTH, 315, 317, 609 RenderbufferStorage, 316, 576 RenderbufferStorage*, 330 RenderbufferStorageMultisample, 305, 314, 315, 576, 706 RENDERER, 566, 567, 638 RenderMode, 682 REPEAT, 185, 240, 242, 260, 285, 600 REPLACE, 504 RESCALE NORMAL, 680 RESET NOTIFICATION STRATEGY, 19, 656 ResumeTransformFeedback, 445, 445, 446, 449, 722 RG, 194, 205, 208–210, 212, 249, 297, 298, 327, 491, 534, 536 RG16, 208, 237, 273, 296, 298 rg16, 296 RG16 SNORM, 208, 273, 296, 298 rg16 snorm, 296 RG16F, 209, 237, 273, 295, 297 rg16f, 295 RG16I, 210, 237, 273, 295, 298 rg16i, 295 RG16UI, 210, 237, 273, 295, 297 rg16ui, 295

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

815

RG32F, 209, 237, 273, 295, 297, 544 rg32f, 295 RG32I, 210, 237, 273, 295, 297, 544 rg32i, 295 RG32UI, 210, 237, 273, 295, 297, 544 rg32ui, 295 RG8, 208, 237, 273, 296, 298 rg8, 296 RG8 SNORM, 208, 273, 296, 298 rg8 snorm, 296 RG8I, 210, 237, 273, 295, 298 rg8i, 295 RG8UI, 210, 237, 273, 295, 297 rg8ui, 295 RG INTEGER, 194, 297, 298 RGB, 194, 197, 202, 205, 208–210, 212, 249, 297, 327, 491, 511, 534–536 RGB10, 209 RGB10 A2, 209, 273, 296, 298 rgb10 a2, 296 RGB10 A2UI, 209, 273, 295, 297 rgb10 a2ui, 295 RGB12, 209 RGB16, 209, 273 RGB16 SNORM, 209, 273 RGB16F, 209, 273 RGB16I, 210, 273 RGB16UI, 210, 273 RGB32F, 209, 237, 273 RGB32I, 210, 237, 273 RGB32UI, 210, 237, 273 RGB4, 208, 706 RGB5, 208 RGB565, 208, 706 RGB5 A1, 209 RGB8, 208, 273 RGB8 SNORM, 209, 273 RGB8I, 210, 273 RGB8UI, 210, 273

RGB9 E5, 207, 209, 273, 289, 535, 695 RGB INTEGER, 194, 197 RGBA, 194, 197, 202, 205, 209, 210, 212, 249, 283–285, 294, 297, 298, 327, 491, 534, 598, 609, 681 RGBA12, 209 RGBA16, 209, 237, 273, 296, 298, 544 rgba16, 296 RGBA16 SNORM, 209, 273, 296, 298, 544 rgba16 snorm, 296 RGBA16F, 209, 237, 273, 295, 297, 544 rgba16f, 295 RGBA16I, 210, 237, 273, 295, 297, 544 rgba16i, 295 RGBA16UI, 210, 238, 273, 295, 297, 544 rgba16ui, 295 RGBA2, 209 RGBA32F, 209, 237, 273, 295, 297, 544 rgba32f, 295 RGBA32I, 210, 238, 273, 295, 297, 544 rgba32i, 295 RGBA32UI, 210, 238, 273, 295, 297, 544 rgba32ui, 295 RGBA4, 209 RGBA8, 209, 237, 273, 296, 298 rgba8, 296 RGBA8 SNORM, 209, 273, 296, 298 rgba8 snorm, 296 RGBA8I, 210, 237, 273, 295, 297 rgba8i, 295 RGBA8UI, 210, 238, 273, 295, 297 rgba8ui, 295 RGBA INTEGER, 194, 197, 294, 295, 297 RIGHT, 507, 517, 518, 520, 521, 530 Rotate, 680

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

816

Sample, 668 sample, 489, 493 sample in, 467, 489 SAMPLE ALPHA TO COVERAGE, 501, 593 SAMPLE ALPHA TO ONE, 501, 503, 593 SAMPLE BUFFERS, 160, 219, 225, 307, 308, 312, 466, 471, 476, 482, 486, 501, 505, 515, 516, 523, 531, 533, 540, 655, 712, 715, 729 SAMPLE COVERAGE, 486, 493, 593 SAMPLE COVERAGE INVERT, 486, 487, 593 SAMPLE COVERAGE VALUE, 486, 487, 593 SAMPLE MASK, 486, 493, 593 SAMPLE MASK VALUE, 15, 486, 487, 593, 687 SAMPLE POSITION, 308, 466, 467, 655, 725 SAMPLE SHADING, 468, 593 SampleCoverage, 487 Sampled, 667, 741 Sampled Type, 667 Sampled1D, 672 SampledBuffer, 672 SampledCubeArray, 672 SampledImageArrayDynamicIndexing, 672 SampledRect, 672 SampleId, 671 SampleMask, 671 SampleMaski, 487 SamplePosition, 671 SAMPLER, 557 sampler*, 157 sampler*Shadow, 401 sampler1D, 117

sampler1DArray, 117 sampler1DArrayShadow, 117 sampler1DShadow, 117 sampler2D, 117, 157 sampler2DArray, 117 sampler2DArrayShadow, 117 sampler2DMS, 117 sampler2DMSArray, 117 sampler2DRect, 117 sampler2DRectShadow, 117 sampler2DShadow, 117 sampler3D, 117 samplerCubeArray, 117 SAMPLER 1D, 117 SAMPLER 1D ARRAY, 117 SAMPLER 1D ARRAY SHADOW, 117 SAMPLER 1D SHADOW, 117 SAMPLER 2D, 117 SAMPLER 2D ARRAY, 117 SAMPLER 2D ARRAY SHADOW, 117 SAMPLER 2D MULTISAMPLE, 117 SAMPLER 2D MULTISAMPLE ARRAY, 117 SAMPLER 2D RECT, 117 SAMPLER 2D RECT SHADOW, 117 SAMPLER 2D SHADOW, 117 SAMPLER 3D, 117 SAMPLER BINDING, 185, 581, 595 SAMPLER BUFFER, 117 SAMPLER CUBE, 117 SAMPLER CUBE MAP ARRAY, 117 SAMPLER CUBE MAP ARRAY SHADOW, 117 SAMPLER CUBE SHADOW, 117 SampleRateShading, 672 samplerBuffer, 117 samplerCube, 117 samplerCubeArrayShadow, 117

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

817

samplerCubeShadow, 117 SamplerParameter, 186 SamplerParameterI{i ui}v, 186 SamplerParameterIiv, 186 SamplerParameterIuiv, 186 SamplerParameteriv, 187, 706 SAMPLES, 234, 307, 308, 312, 466– 468, 506, 540, 541, 578, 653, 655, 712 SAMPLES PASSED, 42, 45, 48, 381, 382, 505 Scale, 680 Scissor, 484, 485 SCISSOR BOX, 602 SCISSOR TEST, 485, 486, 602 ScissorArrayv, 484, 484, 485 ScissorIndexed, 484, 485 ScissorIndexedv, 484, 485 Scope, 667 SecondaryColorPointer, 679 SelectBuffer, 682 SEPARATE ATTRIBS, 168, 392–394 SET, 515 ShadeModel, 680 SHADER, 557 Shader, 672 SHADER BINARY FORMAT SPIR V, 92, 93 SHADER BINARY FORMATS, 92, 637 SHADER COMPILER, 87, 637 SHADER IMAGE ACCESS BARRIER BIT, 162, 165 SHADER IMAGE ATOMIC, 572 SHADER IMAGE LOAD, 572 SHADER IMAGE STORE, 572 SHADER SOURCE LENGTH, 166, 167, 172, 612 SHADER STORAGE BARRIER BIT,

163, 165 SHADER STORAGE BLOCK, 104, 108, 111, 112 SHADER STORAGE BUFFER, 61, 62, 152, 153, 710 SHADER STORAGE BUFFER BINDING, 85, 629 SHADER STORAGE BUFFER OFFSET ALIGNMENT, 85, 646, 727, 730 SHADER STORAGE BUFFER SIZE, 85, 629 SHADER STORAGE BUFFER START, 85, 629 SHADER TYPE, 166, 167, 175, 612 ShaderBinary, 89, 91, 92, 93, 166, 666, 698 ShaderSource, 89, 89, 90, 92, 172 ShaderStorageBlockBinding, 152, 152, 710 SHADING LANGUAGE VERSION, 566–569, 638, 696 shared, 133, 548, 723 SHORT, 193, 297, 298, 352, 537, 538 SIGNALED, 20, 37, 41 SIGNED NORMALIZED, 245, 310, 576 SIMULTANEOUS TEXTURE AND DEPTH TEST, 572 SIMULTANEOUS TEXTURE AND DEPTH WRITE, 572 SIMULTANEOUS TEXTURE AND STENCIL TEST, 572 SIMULTANEOUS TEXTURE AND STENCIL WRITE, 572 SMOOTH LINE WIDTH GRANULARITY, 636 SMOOTH LINE WIDTH RANGE, 636

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

818

SpecializeShader, 92, 93 SPIR V BINARY, 89, 90, 92, 94, 97, 106, 166, 167, 612 SPIR V EXTENSIONS, 569, 638 sqrt(), 673 SRC1 ALPHA, 510–512 SRC1 COLOR, 510–512 SRC ALPHA, 511, 512 SRC ALPHA SATURATE, 511 SRC COLOR, 511, 512 SRGB, 219, 289, 310, 508, 513, 514, 540, 573 SRGB8, 209, 273, 289 SRGB8 ALPHA8, 209, 273, 289 SRGB ALPHA, 289 SRGB READ, 572 SRGB WRITE, 573 STACK OVERFLOW, 18, 556, 693, 697 STACK UNDERFLOW, 18, 556, 693, 697 STATIC COPY, 61, 69 STATIC DRAW, 61, 69, 588 STATIC READ, 61, 69 std140, 133, 146, 147, 151, 723 std430, 147, 151, 548, 738 STENCIL, 309, 525–527, 598, 607 STENCIL ATTACHMENT, 303, 319, 329 STENCIL BACK FAIL, 602 STENCIL BACK FUNC, 602 STENCIL BACK PASS DEPTH FAIL, 602 STENCIL BACK PASS DEPTH PASS, 602 STENCIL BACK REF, 602 STENCIL BACK VALUE MASK, 602 STENCIL BACK WRITEMASK, 604 STENCIL BITS, 682 STENCIL -

BUFFER BIT, 523, 526, 539, 541 STENCIL CLEAR VALUE, 604 STENCIL COMPONENTS, 579 STENCIL FAIL, 602 STENCIL FUNC, 602 STENCIL INDEX, 194, 205, 211, 215, 218, 239, 247, 248, 268, 269, 283, 284, 287, 327, 402, 533, 534, 726, 735 STENCIL INDEX1, 211 STENCIL INDEX16, 211 STENCIL INDEX4, 211 STENCIL INDEX8, 211, 695 STENCIL PASS DEPTH FAIL, 602 STENCIL PASS DEPTH PASS, 602 STENCIL REF, 602 STENCIL RENDERABLE, 579 STENCIL TEST, 503, 602 STENCIL VALUE MASK, 602 STENCIL WRITEMASK, 15, 604 StencilFunc, 503, 503, 504, 664 StencilFuncSeparate, 503, 503, 504 StencilMask, 522, 522, 523, 664 StencilMaskSeparate, 522, 522, 523 StencilOp, 503, 503, 504 StencilOpSeparate, 503, 503, 504 STEREO, 307, 655 Storage Class, 667 StorageBufferArrayDynamicIndexing, 672 StorageImageArrayDynamicIndexing, 672 StorageImageExtendedFormats, 672 StorageImageMultisample, 672 StorageImageWriteWithoutFormat, 668, 672, 737 STREAM COPY, 61, 69 STREAM DRAW, 61, 69 STREAM READ, 61, 69

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

819

Subgroup, 667 SubgroupVoteKHR, 672 SUBPIXEL BITS, 635 switch, 415 SYNC CONDITION, 37, 41, 632 SYNC FENCE, 37, 41, 632 SYNC FLAGS, 37, 41, 632, 723 SYNC FLUSH COMMANDS BIT, 20, 38–40, 723 SYNC GPU COMMANDS COMPLETE, 37, 41, 67, 632 SYNC STATUS, 20, 37, 41, 632 TESS CONTROL OUTPUT VERTICES, 168, 169, 411, 619 TESS CONTROL SHADER, 89, 155, 411, 613 TESS CONTROL SHADER BIT, 124 TESS CONTROL SHADER PATCHES, 43, 48, 429 TESS CONTROL SUBROUTINE, 103, 155 TESS CONTROL SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 TESS CONTROL TEXTURE, 573 TESS EVALUATION SHADER, 89, 155, 426, 613 TESS EVALUATION SHADER BIT, 124 TESS EVALUATION SHADER INVOCATIONS, 43, 48, 429 TESS EVALUATION SUBROUTINE, 103, 155 TESS EVALUATION SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 TESS EVALUATION TEXTURE, 573 TESS GEN MODE, 168, 169, 619 TESS GEN POINT MODE, 169, 619

TESS GEN SPACING, 168, 169, 619 TESS GEN VERTEX ORDER, 168, 169, 619 TessCoord, 671 Tessellation, 672 TessellationPointSize, 672 TessLevelInner, 671 TessLevelOuter, 671 TexBuffer, 236, 576 TexBufferRange, 235, 235 TexCoordPointer, 679 texelFetch, 397, 398, 725 TexEnv, 682 TexImage, 179, 222, 283 TexImage*D, 188, 190, 226 TexImage1D, 211, 216, 216, 217, 219, 220, 226, 230, 265, 286, 576 TexImage2D, 211, 214, 215, 215–218, 222, 226, 230, 265, 286, 294, 576 TexImage2DMultisample, 233, 233, 280, 286, 305, 576, 710 TexImage3D, 203, 204, 211, 213–217, 226, 230, 247, 265, 283, 286, 576, 706 TexImage3DMultisample, 233, 233, 234, 280, 286, 576, 706, 710 TexParameter, 56, 179, 187, 238, 682 TexParameter*, 706 TexParameterI, 238 TexParameterIiv, 239 TexParameterIuiv, 239 TexParameteriv, 239 TexStorage, 283 TexStorage1D, 276, 276, 576 TexStorage2D, 277, 277, 278, 576 TexStorage2DMultisample, 279, 280, 576 TexStorage3D, 278, 278, 279, 576, 735 TexStorage3DMultisample, 280, 280,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

820

576 TexSubImage, 162 TexSubImage1D, 220, 221 TexSubImage2D, 220, 221, 222, 274 TexSubImage3D, 220, 221, 223, 249, 281, 282, 284, 297 TEXTURE, 309, 310, 324, 333, 557 TEXTURE0, 179, 181, 601 TEXTURE x SIZE, 598 TEXTURE x TYPE, 598 TEXTURE xD, 595 TEXTURE 1D, 180, 205, 216, 219, 221, 239, 242, 244, 246, 247, 250, 266, 272, 274, 276, 277, 293, 322, 570, 682 TEXTURE 1D ARRAY, 180, 205, 215, 218, 221, 239, 242, 244, 246, 247, 250, 266, 272, 275, 277, 278, 293, 570, 594, 595, 682 TEXTURE 2D, 157, 180, 205, 215, 218, 221, 239, 242, 244, 246, 247, 250, 266, 272, 274, 277, 293, 322, 570, 682 TEXTURE 2D ARRAY, 180, 203, 205, 215, 221, 230, 232, 239, 242, 244, 246, 247, 266, 272, 274, 275, 279, 293, 570, 594, 595, 682 TEXTURE 2D MULTISAMPLE, 180, 205, 233, 239, 241, 242, 244, 272, 274, 280, 282, 293, 321, 322, 570, 578, 579, 594 TEXTURE 2D MULTISAMPLE ARRAY, 180, 205, 233, 239, 241, 242, 244, 272, 275, 280–282, 293, 570, 578, 579, 594 TEXTURE 3D, 180, 203, 221, 239, 242, 244, 246, 247, 266, 272, 274, 278, 279, 286, 293, 321, 322, 570, 682

TEXTURE ALPHA SIZE, 245 TEXTURE ALPHA TYPE, 245 TEXTURE BASE LEVEL, 239, 241, 242, 255, 270, 285, 596 TEXTURE BINDING xD, 594 TEXTURE BINDING 1D, 581 TEXTURE BINDING 1D ARRAY, 581, 594 TEXTURE BINDING 2D, 581 TEXTURE BINDING 2D ARRAY, 581, 594 TEXTURE BINDING 2D MULTISAMPLE, 581, 594 TEXTURE BINDING 2D MULTISAMPLE ARRAY, 581, 594 TEXTURE BINDING 3D, 581 TEXTURE BINDING BUFFER, 581, 594 TEXTURE BINDING CUBE MAP, 581, 594 TEXTURE BINDING CUBE MAP ARRAY, 581, 594 TEXTURE BINDING RECTANGLE, 581, 594 TEXTURE BLUE SIZE, 245 TEXTURE BLUE TYPE, 244 TEXTURE BORDER COLOR, 186–188, 239, 241, 243, 260, 285, 596, 600 TEXTURE BUFFER, 61, 180, 235, 238, 244, 272, 282, 293, 542, 543, 570, 594 TEXTURE BUFFER BINDING, 656, 712 TEXTURE BUFFER DATA STORE BINDING, 599 TEXTURE BUFFER OFFSET, 599 TEXTURE BUFFER OFFSET ALIGNMENT, 236, 637, 727, 730

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

821

TEXTURE BUFFER SIZE, 599 TEXTURE COMPARE FAIL VALUE ARB, 750 TEXTURE COMPARE FUNC, 239, 285, 287, 597, 600 TEXTURE COMPARE MODE, 239, 285, 287, 401, 402, 597, 600 TEXTURE COMPONENTS, 681 TEXTURE COMPRESSED, 579, 599 TEXTURE COMPRESSED BLOCK HEIGHT, 579 TEXTURE COMPRESSED BLOCK SIZE, 579 TEXTURE COMPRESSED BLOCK WIDTH, 579 TEXTURE COMPRESSED IMAGE SIZE, 229, 231, 245, 246, 250, 599 TEXTURE COMPRESSION HINT, 563, 633 TEXTURE CUBE MAP, 180, 205, 216, 217, 221, 224, 230, 232, 239, 242, 244, 246, 247, 250, 251, 266, 267, 271, 272, 274, 275, 277, 293, 570, 594, 682, 721, 724, 738 TEXTURE CUBE MAP ARRAY, 180, 203, 205, 213, 214, 217, 221, 224, 230, 232, 239, 242, 244, 246, 247, 250, 251, 266, 267, 271, 272, 274, 275, 279, 293, 570, 594, 595, 721, 724, 735, 738–740 TEXTURE CUBE MAP NEGATIVE X, 253, 336, 595 TEXTURE CUBE MAP NEGATIVE Y, 253, 336, 595 TEXTURE CUBE MAP NEGATIVE Z, 253, 336, 595

TEXTURE CUBE MAP POSITIVE X, 244, 253, 291, 336, 595 TEXTURE CUBE MAP POSITIVE Y, 253, 336, 595 TEXTURE CUBE MAP POSITIVE Z, 253, 336, 595 TEXTURE CUBE MAP SEAMLESS, 253, 656 TEXTURE DEPTH, 229, 231, 245, 598 TEXTURE DEPTH SIZE, 245 TEXTURE DEPTH TYPE, 245 TEXTURE ENV, 682 TEXTURE FETCH BARRIER BIT, 162, 165 TEXTURE FILTER CONTROL, 682 TEXTURE FIXED SAMPLE LOCATIONS, 234, 245, 330, 598 TEXTURE GATHER, 573 TEXTURE GATHER SHADOW, 573 TEXTURE GREEN SIZE, 245 TEXTURE GREEN TYPE, 244 TEXTURE HEIGHT, 224, 229, 231, 233, 234, 245, 598 TEXTURE IMAGE FORMAT, 579 TEXTURE IMAGE TYPE, 580 TEXTURE IMMUTABLE FORMAT, 234, 243, 271, 274, 275, 285, 597 TEXTURE IMMUTABLE LEVELS, 243, 265, 271, 275, 286, 597, 743 TEXTURE INTERNAL FORMAT, 229, 231, 234, 245, 598, 681 TEXTURE LOD BIAS, 240, 256, 596, 600, 682 TEXTURE MAG FILTER, 240, 267,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

822

268, 285, 287, 596, 600 TEXTURE MAX ANISOTROPHY, 256 TEXTURE MAX ANISOTROPY, 240, 241, 254, 255, 597, 600, 743 TEXTURE MAX LEVEL, 240, 241, 255, 270, 285, 596 TEXTURE MAX LOD, 240, 241, 255, 285, 596, 600 TEXTURE MIN FILTER, 184, 240, 242, 259, 260, 264, 268, 270, 285, 287, 596, 600 TEXTURE MIN LOD, 240, 241, 255, 285, 596, 600 TEXTURE PRIORITY, 682 TEXTURE RECTANGLE, 180, 205, 215, 218, 221, 225, 227, 232, 239, 242, 244, 246–248, 250, 272, 274, 277, 278, 282, 293, 321, 322, 570, 594, 595 TEXTURE RECTANGLE ARB, 752 TEXTURE RED SIZE, 245 TEXTURE RED TYPE, 244 TEXTURE SAMPLES, 233, 234, 245, 312, 329, 598 TEXTURE SHADOW, 573 TEXTURE SHARED SIZE, 245, 598 TEXTURE STENCIL SIZE, 245 TEXTURE SWIZZLE A, 239, 240, 285, 490, 596 TEXTURE SWIZZLE B, 239, 240, 285, 490, 596 TEXTURE SWIZZLE G, 239, 240, 285, 490, 596 TEXTURE SWIZZLE R, 239, 240, 285, 490, 596 187, TEXTURE SWIZZLE RGBA, 239–241 TEXTURE TARGET, 221, 243, 248, 250, 251, 596

TEXTURE UPDATE BARRIER BIT, 163 TEXTURE VIEW, 573 TEXTURE VIEW MIN LAYER, 243, 271, 286, 597 TEXTURE VIEW MIN LEVEL, 243, 271, 274, 286, 597 TEXTURE VIEW NUM LAYERS, 243, 271, 275, 286, 597 TEXTURE VIEW NUM LEVELS, 243, 271, 275, 286, 321, 323, 597, 743 TEXTURE WIDTH, 224, 229, 231, 233, 234, 245, 598 TEXTURE WRAP R, 185, 240, 260, 596, 600, 681 TEXTURE WRAP S, 185, 240, 242, 260, 596, 600, 681 TEXTURE WRAP T, 185, 240, 242, 260, 596, 600, 681 TextureBarrier, 326, 326 TextureBuffer, 236 TextureBufferRange, 235, 235, 742 textureGather, 257, 262, 263, 643, 758 textureGather*, 724 textureGatherOffset, 263 textureGatherOffsets, 263, 724 textureLOD, 758 TextureParameter, 238 TextureParameter*, 241 TextureParameterI, 239 textureQueryLevels, 400 textureSize, 399, 738 TextureStorage*, 276 TextureStorage1D, 276, 276, 742 TextureStorage2D, 277, 278 TextureStorage2DMultisample, 279, 280 TextureStorage3D, 278, 279 TextureStorage3DMultisample, 280,

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

823

281 TextureSubImage1D, 220, 221 TextureSubImage2D, 220, 221, 222 TextureSubImage3D, 220, 221, 222 TextureView, 206, 241, 270, 272, 273, 573, 576, 699, 724, 729 TIME ELAPSED, 42, 48, 51, 697 TIMEOUT EXPIRED, 38 TIMEOUT IGNORED, 39 TIMESTAMP, 42, 44–48, 51, 52, 697 TOP LEVEL ARRAY SIZE, 112, 115, 624, 710, 727 TOP LEVEL ARRAY STRIDE, 112, 115, 624, 727 TRANSFORM FEEDBACK, 443, 444, 557 TRANSFORM FEEDBACK ACTIVE, 581, 630, 686 TRANSFORM FEEDBACK BARRIER BIT, 163 TRANSFORM FEEDBACK BINDING, 589 TRANSFORM FEEDBACK BUFFER, 61, 62, 103, 106, 108–112, 395, 446, 447, 449 TRANSFORM FEEDBACK BUFFER ACTIVE, 686 TRANSFORM FEEDBACK BUFFER BINDING, 85, 443, 581, 589, 630, 740 TRANSFORM FEEDBACK BUFFER INDEX, 112, 115 TRANSFORM FEEDBACK BUFFER MODE, 168, 616 TRANSFORM FEEDBACK BUFFER PAUSED, 686 TRANSFORM FEEDBACK BUFFER SIZE, 85, 581, 630 TRANSFORM FEED-

BACK BUFFER START, 85, 581, 630 TRANSFORM FEEDBACK BUFFER STRIDE, 112, 115 TRANSFORM FEEDBACK OVERFLOW, 42, 48, 381, 382, 452 TRANSFORM FEEDBACK PAUSED, 581, 630, 686 TRANSFORM FEEDBACK PRIMITIVES WRITTEN, 42, 45, 46, 48, 448, 452 TRANSFORM FEEDBACK STREAM OVERFLOW, 42, 45, 46, 48, 381–383, 452 TRANSFORM FEEDBACK VARYING, 103, 106, 109, 111, 112, 114, 395, 396 TRANSFORM FEEDBACK VARYING MAX LENGTH, 168, 616 TRANSFORM FEEDBACK VARYINGS, 168, 396, 616 TransformFeedback, 672 TransformFeedbackBufferBase, 446, 447 TransformFeedbackBufferRange, 446, 447 TransformFeedbackVarying, 396 TransformFeedbackVaryings, 103, 106, 109, 115, 392, 393–396 Translate, 680 TRIANGLE FAN, 341, 346, 432, 438, 446 TRIANGLE STRIP, 168, 340, 346, 432, 438, 446, 459, 616 TRIANGLE STRIP ADJACENCY, 344, 346, 432, 438

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

824

Uniform2f, 11 Uniform2i, 11 Uniform3f, 11 Uniform3i, 11 Uniform4f, 11, 12 Uniform4f{v}, 142 Uniform4i, 12 UNIFORM ARRAY STRIDE, 138, 145, 150, 618, 698, 739 UNIFORM ATOMIC COUNTER BUFFER INDEX, 138, 621 UNIFORM BARRIER BIT, 162, 166 UNIFORM BLOCK, 102, 108, 111, 112, 137, 138 UNIFORM BLOCK ACTIVE UNIFORM INDICES, 139, 618 UNIFORM BLOCK ACTIVE UNIFORMS, 139, 618 UNIFORM BLOCK BINDING, 139, 618 UNIFORM BLOCK DATA SIZE, 139, uimage1D, 119 149, 618 uimage1DArray, 119 UNIFORM BLOCK INDEX, 138, 617 uimage2D, 119 UNIFORM BLOCK NAME uimage2DArray, 119 LENGTH, 139, 618, 716 uimage2DMS, 119 UNIFORM BLOCK REFERENCED uimage2DMSArray, 119 BY COMPUTE SHADER, uimage2DRect, 119 139, 619 uimage3D, 119 UNIFORM BLOCK REFERENCED uimageBuffer, 119 BY FRAGMENT SHADER, uimageCube, 119 139, 619 uimageCubeArray, 119 UNIFORM BLOCK REFERENCED uint, 116, 127, 145, 150, 386 BY GEOMETRY SHADER, UNDEFINED VERTEX, 438 139, 619 UNIFORM, 102, 106, 111, 112, 114, UNIFORM BLOCK REFERENCED 120, 136, 137 BY TESS CONTROL Uniform, 11, 147, 148, 667, 668 SHADER, 139, 618 Uniform1f, 11 UNIFORM BLOCK REFERENCED Uniform1i, 11 BY TESS EVALUATION TRIANGLES, 168, 341, 346, 432, 438, 445, 446, 616 triangles, 415, 416, 418, 419, 432 TRIANGLES ADJACENCY, 168, 343, 346, 432, 438 triangles adjacency, 432 TRUE, 14, 15, 20, 41, 44, 47, 49, 50, 60, 61, 68, 70, 71, 75, 78, 87, 90– 92, 94, 96, 99–101, 106, 122, 131, 132, 142, 166, 167, 169, 183, 188, 189, 233, 234, 271, 274, 275, 290–294, 305, 311, 314, 324, 330, 348, 351, 353, 361, 379, 382, 406, 443, 486, 487, 506, 521, 532, 535, 549, 554, 573, 574, 576, 578, 579, 588, 593, 598, 603, 604, 626, 637, 649, 680, 720 true, 492 TYPE, 106, 112, 113, 115, 137, 138, 388, 396, 624

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

825

SHADER, 139 UnmapNamedBuffer, 78, 78, 79 UNIFORM BLOCK REFERENCED - UNPACK ALIGNMENT, 189, 195, BY TESS EVALUTION 204, 228, 610 UNPACK COMPRESSED BLOCK SHADER, 618 UNIFORM BLOCK REFERENCED DEPTH, 189, 226, 610 BY VERTEX SHADER, 139, UNPACK COMPRESSED BLOCK 618 HEIGHT, 189, 226, 610 UNIFORM BUFFER, 61, 62, 148 UNPACK COMPRESSED BLOCK UNIFORM BUFFER BINDING, 85, SIZE, 189, 226, 610 UNPACK COMPRESSED BLOCK 631 UNIFORM BUFFER OFFSET WIDTH, 189, 226, 610 ALIGNMENT, 85, 645, 727, UNPACK IMAGE HEIGHT, 189, 204, 730 226–228, 610 UNIFORM BUFFER SIZE, 85, 631 UNPACK LSB FIRST, 189, 610, 679, UNIFORM BUFFER START, 85, 631 693, 701 UNIFORM IS ROW MAJOR, 138, UNPACK ROW LENGTH, 189, 195, 618 204, 226, 228, 610 UNIFORM MATRIX STRIDE, 138, UNPACK SKIP IMAGES, 189, 204, 618 216, 226–228, 610 UNIFORM NAME LENGTH, 138, UNPACK SKIP PIXELS, 189, 195, 617, 620 226–228, 610 UNIFORM OFFSET, 138, 617 UNPACK SKIP ROWS, 189, 195, UNIFORM SIZE, 138, 617, 620 226–228, 610 UNIFORM TYPE, 138, 617 UNPACK SWAP BYTES, 189, 192, Uniform{1234}{ifd ui}, 140 194, 610 Uniform{1234}{ifd ui}v, 140 UNSIGNALED, 37, 41, 632 UniformBlockBinding, 149, 149 UNSIGNED BYTE, 193, 297, 298, UniformBufferArrayDynamicIndexing, 352, 354, 361, 370, 537, 538 672 UNSIGNED BYTE 2 3 3 REV, 193, UniformConstant, 667 197, 198, 537 UniformMatrix2x4fv, 141 UNSIGNED BYTE 3 3 2, 193, 197, UniformMatrix3dv, 142 198, 537 UniformMatrix{234}, 141 UNSIGNED INT, 116, 193, 245, 295, UniformMatrix{2x3,3x2,2x4,4x2,3x4,4x3}, 297, 310, 352, 361, 370, 537, 141 538, 576 UniformSubroutinesuiv, 156, 706 193, UNSIGNED INT 10 10 10 2, Unknown, 668 197, 200, 537 UNKNOWN CONTEXT RESET, 19 UNSIGNED INT 10F 11F 11F REV, UnmapBuffer, 56, 60, 67, 69, 75, 78, 78, 193, 197, 200, 202, 297, 348, 79 349, 353, 354, 362, 363, 535, OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

826

537 UNSIGNED INT 24 8, 193, 197, 200, 533, 537, 538 UNSIGNED INT 2 10 10 10 REV, 193, 197, 200, 297, 298, 348, 353, 354, 362, 537 UNSIGNED INT 5 9 9 9 REV, 193, 197, 200, 202, 208, 535, 537 UNSIGNED INT 8 8 8 8, 193, 197, 200, 537 UNSIGNED INT 8 8 8 8 REV, 193, 197, 200, 537 UNSIGNED INT ATOMIC COUNTER, 119 UNSIGNED INT IMAGE 1D, 119 UNSIGNED INT IMAGE 1D ARRAY, 119 UNSIGNED INT IMAGE 2D, 119 UNSIGNED INT IMAGE 2D ARRAY, 119 UNSIGNED INT IMAGE 2D MULTISAMPLE, 119 UNSIGNED INT IMAGE 2D MULTISAMPLE ARRAY, 119 UNSIGNED INT IMAGE 2D RECT, 119 UNSIGNED INT IMAGE 3D, 119 UNSIGNED INT IMAGE BUFFER, 119 UNSIGNED INT IMAGE CUBE, 119 UNSIGNED INT IMAGE CUBE MAP ARRAY, 119 UNSIGNED INT SAMPLER 1D, 118 UNSIGNED INT SAMPLER 1D ARRAY, 118 UNSIGNED INT SAMPLER 2D, 118 UNSIGNED INT SAMPLER 2D ARRAY, 118 UNSIGNED INT SAMPLER 2D -

MULTISAMPLE, 118 UNSIGNED INT SAMPLER 2D MULTISAMPLE ARRAY, 118 UNSIGNED INT SAMPLER 2D RECT, 118 UNSIGNED INT SAMPLER 3D, 118 UNSIGNED INT SAMPLER BUFFER, 118 UNSIGNED INT SAMPLER CUBE, 118 UNSIGNED INT SAMPLER CUBE MAP ARRAY, 118 UNSIGNED INT VEC2, 116 UNSIGNED INT VEC3, 116 UNSIGNED INT VEC4, 116 UNSIGNED NORMALIZED, 245, 310, 576 UNSIGNED SHORT, 193, 297, 298, 352, 361, 370, 537, 538 UNSIGNED SHORT 1 5 5 5 REV, 193, 197, 199, 537 UNSIGNED SHORT 4 4 4 4, 193, 197, 199, 537 UNSIGNED SHORT 4 4 4 4 REV, 193, 197, 199, 537 UNSIGNED SHORT 5 5 5 1, 193, 197, 199, 537 UNSIGNED SHORT 5 6 5, 193, 197, 199, 537 UNSIGNED SHORT 5 6 5 REV, 193, 197, 199, 537 UPPER LEFT, 454, 458, 469, 470, 477, 591 usampler1D, 118 usampler1DArray, 118 usampler2D, 118 usampler2DArray, 118 usampler2DMS, 118 usampler2DMSArray, 118

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

827

usampler2DRect, 118 usampler3D, 118 usamplerBuffer, 118 usamplerCube, 118 usamplerCubeArray, 118 UseProgram, 98, 98, 99, 123, 141, 157, 406, 407, 449, 698, 721, 728 UseProgramStages, 99, 123, 124, 157, 169, 406, 449, 698, 710, 721 uvec2, 116, 386 uvec3, 116, 386 uvec4, 116, 294, 295, 386 VALIDATE STATUS, 167, 170, 406, 407, 613, 614 ValidateProgram, 167, 406, 407, 711 ValidateProgramPipeline, 170, 407, 711 vec2, 116, 386 vec3, 116, 386 vec4, 116, 142, 146, 147, 294, 386 VENDOR, 566, 567, 638 VERSION, 566, 567, 638 VERTEX ARRAY, 557 VERTEX ARRAY BINDING, 352, 378, 379, 587 VERTEX ATTRIB ARRAY BARRIER BIT, 162 VERTEX ATTRIB ARRAY BUFFER, 162 VERTEX ATTRIB ARRAY 363, BUFFER BINDING, 378, 379, 586, 708, 736 VERTEX ATTRIB ARRAY DIVISOR, 377–379, 585, 708 VERTEX ATTRIB ARRAY ENABLED, 377–379, 585 VERTEX ATTRIB ARRAY INTEGER, 377–379, 585

VERTEX ATTRIB ARRAY LONG, 377–379, 585, 693, 698 VERTEX ATTRIB ARRAY NORMALIZED, 377–379, 585 VERTEX ATTRIB ARRAY POINTER, 358, 380, 585 VERTEX ATTRIB ARRAY SIZE, 377–379, 585 VERTEX ATTRIB ARRAY STRIDE, 358, 377–379, 585 VERTEX ATTRIB ARRAY TYPE, 377–379, 585 VERTEX ATTRIB BINDING, 379, 586 VERTEX ATTRIB RELATIVE OFFSET, 378, 379, 586 VERTEX BINDING BUFFER, 378, 379, 586, 709, 736, 737 VERTEX BINDING DIVISOR, 378, 379, 586, 709, 736, 737 VERTEX BINDING OFFSET, 378, 586 VERTEX BINDING STRIDE, 358, 378, 586, 736 VERTEX PROGRAM TWO SIDE, 680 VERTEX SHADER, 89, 155, 172, 173, 613 VERTEX SHADER BIT, 124 VERTEX SHADER INVOCATIONS, 42, 48, 409 VERTEX SUBROUTINE, 103, 155 VERTEX SUBROUTINE UNIFORM, 103, 108, 111, 112, 120, 155 VERTEX TEXTURE, 573 VertexArrayAttribBinding, 357, 357 VertexArrayAttribFormat, 352

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

828

VertexArrayAttribIFormat, 352 VertexArrayAttribLFormat, 352 VertexArrayBindingDivisor, 360, 360 VertexArrayElementBuffer, 351 VertexArrayVertexBuffer, 354, 355 VertexArrayVertexBuffers, 356, 356, 357 VertexAttrib, 347, 381 VertexAttrib4, 347 VertexAttrib4N, 347 VertexAttrib4Nub, 347, 713 VertexAttribBinding, 357, 357, 358, 363 VertexAttribDivisor, 360, 365, 366, 370 VertexAttribFormat, 352, 354 VertexAttribI, 347 VertexAttribI1i, 386 VertexAttribI1ui, 386 VertexAttribI2i, 386 VertexAttribI2ui, 386 VertexAttribI3i, 386 VertexAttribI3ui, 386 VertexAttribI4, 347 VertexAttribI4i, 386 VertexAttribI4ui, 386 VertexAttribIFormat, 352, 354 VertexAttribIPointer, 358, 379 VertexAttribL1d, 386 VertexAttribL2d, 386 VertexAttribL3d, 386 VertexAttribL3dv, 385 VertexAttribL4d, 386 VertexAttribL{1234}d, 347 VertexAttribL{1234}dv, 347 VertexAttribLFormat, 352, 354 VertexAttribLPointer, 358, 385 VertexAttribP1ui, 348 VertexAttribP2ui, 348 VertexAttribP3ui, 348 VertexAttribP4ui, 348, 349

VertexAttribP4uiv, 349 VertexAttribPointer, 351, 358, 363, 379, 680 VertexBindingDivisor, 359, 360 VertexId, 671 VertexPointer, 679 vertices, 410 VERTICES SUBMITTED, 42, 48, 383 VIEW CLASS 128 BITS, 273 VIEW CLASS 16 BITS, 273 VIEW CLASS 24 BITS, 273 VIEW CLASS 32 BITS, 273 VIEW CLASS 48 BITS, 273 VIEW CLASS 64 BITS, 273 VIEW CLASS 8 BITS, 273 VIEW CLASS 96 BITS, 273 VIEW CLASS BPTC FLOAT, 273 VIEW CLASS BPTC UNORM, 273 VIEW CLASS RGTC1 RED, 273 VIEW CLASS RGTC2 RG, 273 VIEW COMPATIBILITY CLASS, 580, 713 VIEWPORT, 589 Viewport, 460, 460 VIEWPORT BOUNDS RANGE, 461, 636 VIEWPORT INDEX PROVOKING VERTEX, 438, 636 VIEWPORT SUBPIXEL BITS, 461, 636 ViewportArrayv, 460, 460 ViewportIndex, 671 ViewportIndexedf, 460, 460 ViewportIndexedfv, 460, 460 WAIT FAILED, 38 WaitSync, 36–38, 39, 39, 40, 55, 56, 651 WGL ARB create context, 549, 755

OpenGL 4.6 (Core Profile) - May 5, 2022

INDEX

829

WGL ARB create context no error, 778 WGL ARB create context profile, 678, 759 WGL ARB create context robustness, 763 WGL ARB framebuffer sRGB, 754 WGL ARB pixel format float, 752 WGL ARB robustness application isolation, 769 WGL ARB robustness share group isolation, 769 while, 415 Width, 498 Workgroup, 667 WorkgroupId, 671 WorkgroupSize, 669, 671 WRITE ONLY, 61, 75, 77, 291 Xfb, 391, 395, 668 xfb buffer, 391, 392, 395 xfb offset, 115, 391, 392, 395 xfb stride, 391, 392, 394, 395 XfbBuffer, 391, 395, 668 XfbStride, 391, 668, 669 XOR, 515 ZERO, 240, 490, 504, 509, 511, 513, 603 ZERO TO ONE, 454, 458

OpenGL 4.6 (Core Profile) - May 5, 2022

OpenGL 4.6 Specification — OpenGL 4.6 Specification

Related documents