Vulkan® Memory Allocator
VMA is our single-header, MIT-licensed, C++ library for easily and efficiently managing memory allocation for your Vulkan® games and applications.
You love our Performance Guide but sometimes it can be hard to work with. All those optimization suggestions are great, but how can you tell if they apply to your engine? Enter the AMD checks for the Vulkan® Best Practice validation layer.
We took many of our performance suggestions and incorporated them into the validation layer. This layer will analyze Vulkan API calls and alert you on sub-optimal usage. Now you don’t have to remember which flags you should or shouldn’t use or if you are clearing with a fast color. The Best Practice layer will check all of that, and more, for you!
There are several ways to enable the best practice layer:
- Set an environment variable.
Simply set:
VK_LAYER_ENABLES=VK_VALIDATION_FEATURE_ENABLE_BEST_PRACTICES_EXT;VALIDATION_CHECK_ENABLE_VENDOR_SPECIFIC_AMD- Use the Vulkan Configurator.
This easy-to-use GUI app can check which layers are present and configure the layers and their log output.
- Enable the layer in your engine during the Vulkan Instance creation.
This method allows you to more finely control the log output of the layer. Maybe you want to ignore certain suggestions, or elevate the severity of others. While this method requires changes in code, it allows for more control.
That’s it! You are ready to tune your engine with AMD’s Best Practices. We plan on adding even more checks in the future, as well as tune the existing checks based on your feedback.
The layer was recently merged into the validation layer Github Repo, and will be in version 1.2.189 of the SDK.
