|
HIP: Heterogenous-computing Interface for Portability
|
Topics | |
| Bfloat16 Arithmetic Functions | |
| Bfloat16 Comparision Functions | |
| Bfloat162 Comparision Functions | |
| Bfloat162 Arithmetic Functions | |
| Bfloat16 Conversion Functions | |
| Bfloat162 Conversion Functions | |
| Bfloat16 Math Functions | |
| Bfloat162 Math Functions | |
| Bfloat16 Raw Struct | |
| Bfloat162 Raw Struct | |
| HIP_INTRINSIC_BFLOAT16_STRUCT | |
| Struct to represent a 16 bit brain floating point number. | |
| HIP_INTRINSIC_BFLOAT162_STRUCT | |
| Struct to represent a two 16 bit brain floating point number. | |
This section describes hip_bfloat16 precision intrinsic functions. To use these functions, include the header file hip_bf16.h in your program.