NVIDIACUDA Toolkit Documentation
Search In:
CUDA Toolkit v12.3.1
CUDA Math API
  • 1. Modules
    • 1.1. FP8 Intrinsics
      • 1.1.1. FP8 Conversion and Data Movement
      • 1.1.2. C++ struct for handling fp8 data type of e5m2 kind.
      • 1.1.3. C++ struct for handling vector type of two fp8 values of e5m2 kind.
      • 1.1.4. C++ struct for handling vector type of four fp8 values of e5m2 kind.
      • 1.1.5. C++ struct for handling fp8 data type of e4m3 kind.
      • 1.1.6. C++ struct for handling vector type of two fp8 values of e4m3 kind.
      • 1.1.7. C++ struct for handling vector type of four fp8 values of e4m3 kind.
    • 1.2. Half Precision Intrinsics
      • 1.2.1. Half Arithmetic Constants
      • 1.2.2. Half Arithmetic Functions
      • 1.2.3. Half2 Arithmetic Functions
      • 1.2.4. Half Comparison Functions
      • 1.2.5. Half2 Comparison Functions
      • 1.2.6. Half Precision Conversion and Data Movement
      • 1.2.7. Half Math Functions
      • 1.2.8. Half2 Math Functions
    • 1.3. Bfloat16 Precision Intrinsics
      • 1.3.1. Bfloat16 Arithmetic Constants
      • 1.3.2. Bfloat16 Arithmetic Functions
      • 1.3.3. Bfloat162 Arithmetic Functions
      • 1.3.4. Bfloat16 Comparison Functions
      • 1.3.5. Bfloat162 Comparison Functions
      • 1.3.6. Bfloat16 Precision Conversion and Data Movement
      • 1.3.7. Bfloat16 Math Functions
      • 1.3.8. Bfloat162 Math Functions
    • 1.4. Mathematical Functions
    • 1.5. Single Precision Mathematical Functions
    • 1.6. Double Precision Mathematical Functions
    • 1.7. Integer Mathematical Functions
    • 1.8. Single Precision Intrinsics
    • 1.9. Double Precision Intrinsics
    • 1.10. Integer Intrinsics
    • 1.11. Type Casting Intrinsics
    • 1.12. SIMD Intrinsics
  • 2. Data Structures
    • 2.1. __half
    • 2.2. __half2
    • 2.3. __half2_raw
    • 2.4. __half_raw
    • 2.5. __nv_bfloat16
    • 2.6. __nv_bfloat162
    • 2.7. __nv_bfloat162_raw
    • 2.8. __nv_bfloat16_raw
    • 2.9. __nv_fp8_e4m3
    • 2.10. __nv_fp8_e5m2
    • 2.11. __nv_fp8x2_e4m3
    • 2.12. __nv_fp8x2_e5m2
    • 2.13. __nv_fp8x4_e4m3
    • 2.14. __nv_fp8x4_e5m2
  • 3. Data Fields

Search Results

    < Previous | Next >
    CUDA Math API (PDF) - v12.3.1 (older) - Last updated November 15, 2023 - Send Feedback

    1. Modules

    Here is a list of all modules:

    • FP8 Intrinsics
      • FP8 Conversion and Data Movement
      • C++ struct for handling fp8 data type of e5m2 kind.
      • C++ struct for handling vector type of two fp8 values of e5m2 kind.
      • C++ struct for handling vector type of four fp8 values of e5m2 kind.
      • C++ struct for handling fp8 data type of e4m3 kind.
      • C++ struct for handling vector type of two fp8 values of e4m3 kind.
      • C++ struct for handling vector type of four fp8 values of e4m3 kind.
    • Half Precision Intrinsics
      • Half Arithmetic Constants
      • Half Arithmetic Functions
      • Half2 Arithmetic Functions
      • Half Comparison Functions
      • Half2 Comparison Functions
      • Half Precision Conversion and Data Movement
      • Half Math Functions
      • Half2 Math Functions
    • Bfloat16 Precision Intrinsics
      • Bfloat16 Arithmetic Constants
      • Bfloat16 Arithmetic Functions
      • Bfloat162 Arithmetic Functions
      • Bfloat16 Comparison Functions
      • Bfloat162 Comparison Functions
      • Bfloat16 Precision Conversion and Data Movement
      • Bfloat16 Math Functions
      • Bfloat162 Math Functions
    • Mathematical Functions
    • Single Precision Mathematical Functions
    • Double Precision Mathematical Functions
    • Integer Mathematical Functions
    • Single Precision Intrinsics
    • Double Precision Intrinsics
    • Integer Intrinsics
    • Type Casting Intrinsics
    • SIMD Intrinsics