Package org.lwjgl.vulkan

Class EXTDeviceGeneratedCommands

java.lang.Object
org.lwjgl.vulkan.EXTDeviceGeneratedCommands
public class EXTDeviceGeneratedCommands extends Object
This extension allows the device to generate a number of commands for command buffers. It provides a subset of functionality from both VK_NV_device_generated_commands and VK_NV_device_generated_commands_compute as well as some new features.

When rendering a large number of objects, the device can be leveraged to implement a number of critical functions, like updating matrices, or implementing occlusion culling, frustum culling, front to back sorting, etc. Implementing those on the device does not require any special extension, since an application is free to define its own data structures, and just process them using shaders.

To render objects which have been processed on the device, Vulkan has several ways to perform indirect rendering, from the most basic vkCmdDrawIndirect with one indirect draw to vkCmdDrawIndirectCount which supports multiple indirect draws batched together, with a way to determine number of draws at device execution time.

However, if rendering state needs to change between the indirect draws, then unextended Vulkan forces the application to speculatively record a prohibitive number of redundant indirect commands covering all possible state combinations - which could end up processing nothing after culling - or read back the processed stream and issue graphics command from the host. For very large scenes, the synchronization overhead and cost to generate the command buffer can become the bottleneck. This extension allows an application to generate a device side stream of state changes and commands, and convert it efficiently into a command buffer without having to read it back to the host.

Furthermore, it allows incremental changes to such command buffers by manipulating only partial sections of a command stream — for example pipeline and shader object bindings. Unextended Vulkan requires re-creation of entire command buffers in such a scenario, or updates synchronized on the host.

The intended usage for this extension is for the application to:

  • create VkBuffer objects and retrieve physical addresses from them via GetBufferDeviceAddress
  • create a VkIndirectExecutionSetEXT for the ability to change shaders on the device.
  • create a VkIndirectCommandsLayoutEXT, which lists the VkIndirectCommandsTokenTypeEXT it wants to dynamically execute as an atomic command sequence. This step likely involves some internal device code compilation, since the intent is for the GPU to generate the command buffer based on the layout.
  • fill the input stream buffers with the data for each of the inputs it needs. Each input is an array that will be filled with token-dependent data.
  • set up a preprocess VkBuffer that uses memory according to the information retrieved via GetGeneratedCommandsMemoryRequirementsEXT.
  • optionally preprocess the generated content using CmdPreprocessGeneratedCommandsEXT, for example on an asynchronous compute queue, or for the purpose of re-using the data in multiple executions.
  • call CmdExecuteGeneratedCommandsEXT to create and execute the actual device commands for all sequences based on the inputs provided.

For each draw in a sequence, the following can be specified:

  • a number of vertex buffer bindings
  • a different index buffer, with an optional dynamic offset and index type
  • a number of different push constants
  • updates to bound shader stages

For each dispatch in a sequence, the following can be specified:

  • a number of different push constants
  • updates to bound shader stages

For each trace rays in a sequence, the following can be specified:

  • a number of different push constants
  • updates to bound shader stages

While the GPU can be faster than a CPU to generate the commands, it will not happen asynchronously to the device, therefore the primary use case is generating “less” total work (occlusion culling, classification to use specialized shaders, etc.).

Example Code

TODO

Name String
VK_EXT_device_generated_commands
Extension Type
Device extension
Registered Extension Number
573
Revision
1
Extension and Version Dependencies
VK_KHR_buffer_device_address and VK_KHR_maintenance5
API Interactions
  • Interacts with VK_EXT_shader_object
Contact
  • Mike Blumenkrantz zmike
Extension Proposal
VK_EXT_device_generated_commands
Other Extension Metadata
Last Modified Date
2024-02-23
Interactions and External Dependencies
  • This extension requires Vulkan 1.1
  • This extension requires VK_EXT_buffer_device_address or VK_KHR_buffer_device_address or Vulkan 1.2 for the ability to bind vertex and index buffers on the device.
  • This extension requires VK_KHR_maintenance5 for the ability to use VkPipelineCreateFlags2KHR.
  • This extension interacts with VK_NV_mesh_shader. If the latter extension is not supported, remove the command tokens to initiate NV mesh tasks drawing in this extension.
  • This extension interacts with VK_EXT_mesh_shader. If the latter extension is not supported, remove the command tokens to initiate EXT mesh tasks drawing in this extension.
  • This extension interacts with VK_KHR_ray_tracing_pipeline. If the latter extension is not supported, remove the command tokens to initiate ray tracing in this extension.
  • This extension interacts with VK_EXT_shader_object. If the latter extension is not supported, remove references to shader objects in this extension.
Contributors
  • Mike Blumenkrantz, VALVE
  • Hans-Kristian Arntzen, VALVE
  • Jan-Harald Fredriksen, ARM
  • Spencer Fricke, LunarG
  • Ricardo Garcia, Igalia
  • Tobias Hector, AMD
  • Baldur Karlsson, VALVE
  • Christoph Kubisch, NVIDIA
  • Lionel Landwerlin, INTEL
  • Jon Leech, Khronos
  • Ting Wei, ARM
  • Ken Shanyi Zhang, AMD
  • Faith Ekstrand, Collabora
  • Vikram Kushwaha, NVIDIA
  • Connor Abbott, VALVE
  • Samuel Pitoiset, VALVE