Custom compute capability - Specify a custom CUDA architecture flag value - MATLAB (original) (raw)
Main Content
Specify a custom CUDA architecture flag value
Since R2020b
Model Configuration Pane: Code Generation / GPU Code
Description
The Custom compute capability parameter specifies the name of the NVIDIA® virtual GPU architecture for code generation.
Dependencies
- This parameter requires a GPU Coder™ license.
- To enable this parameter, select Generate GPU code on the Code Generation pane.
Settings
'' (default) | _`argument-list`_
Specify the name of the NVIDIA virtual GPU architecture for which the CUDA® input files must be compiled.
For example, to specify a virtual architecture type -arch=compute_50. You can specify a real architecture using -arch=sm_50. For more information, see the Options for Steering GPU Code Generation topic in the CUDA Toolkit documentation.
Recommended Settings
| Application | Setting |
|---|---|
| Debugging | No impact |
| Traceability | No impact |
| Efficiency | No impact |
| Safety precaution | No impact |
Programmatic Use
| Parameter: GPUCustomComputeCapability |
|---|
| Type: character vector |
| Value: '' or a valid user-specified virtual architecture specification |
| Default: '' |
Version History
Introduced in R2020b