Reference documentation#
- Compilation API
- CUDA Kernel API
- Kernel declaration
- Dispatcher objects
- Kernel Arguments
- Intrinsic Attributes and Functions
- Cache Hints for Memory Operations
- CUDA-Specific Types
- Vector Types
- Narrow Data Types
- Bfloat16
is_bfloat16_supported()- Data Movement and Casts
- Arithmetic
- Math Intrinsics
- Arithmetic Intrinsics
- Comparison Intrinsics
- Precision Conversion and Data Movement
- Bfloat16
- Libdevice functions
- Environment Variables
- CUDA Host API (Deprecated)
- Memory Management (Deprecated)