NVIDIACUDA Toolkit Documentation
Search In:
CUDA Toolkit v7.5
CUDA Math API
  • 1. Modules
    • 1.1. Mathematical Functions
    • 1.2. Single Precision Mathematical Functions
    • 1.3. Double Precision Mathematical Functions
    • 1.4. Single Precision Intrinsics
    • 1.5. Double Precision Intrinsics
    • 1.6. Integer Intrinsics
    • 1.7. Type Casting Intrinsics
    • 1.8. SIMD Intrinsics
    • 1.9. Half Precision Intrinsics
      • 1.9.1. Half Arithmetic Functions
      • 1.9.2. Half2 Arithmetic Functions
      • 1.9.3. Half Comparison Functions
      • 1.9.4. Half2 Comparison Functions
      • 1.9.5. Half Precision Conversion And Data Movement
  • Notices

    Search Results

      < Previous | Next >
      CUDA Math API (PDF) - v7.5 (older) - Last updated September 1, 2015 - Send Feedback -

      1.9. Half Precision Intrinsics

      This section describes half precision intrinsic functions that are only supported in device code.

      Modules

       
      Half Arithmetic Functions
       
      Half Comparison Functions
       
      Half Precision Conversion And Data Movement
       
      Half2 Arithmetic Functions
       
      Half2 Comparison Functions

      Half Arithmetic Functions

      Half2 Arithmetic Functions

      Half Comparison Functions

      Half2 Comparison Functions

      Half Precision Conversion And Data Movement