CUTLASS: half.h File Reference

| | CUTLASS

CUDA Templates for Linear Algebra Subroutines and Solvers |

• Main Page

• Modules

• Namespaces

• Classes

• Files

• File List

• File Members

• include

• cutlass

Classes | Namespaces | Macros | Functions

half.h File Reference

Defines a class for using IEEE half-precision floating-point types in host or device code. More...

#include <cmath>
#include <limits>
#include <cstdint>
#include <cuda_fp16.h>
#include "cutlass/cutlass.h"

Include dependency graph for half.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

|

Classes

| | struct | cutlass::half_t | | | IEEE half-precision floating-point type. More...
| | | | struct | [std::numeric_limits< cutlass::half_t >](structstd_1_1numeric limits_3_01cutlass_1_1half t_01_4.html) | | | Numeric limits. [More...](structstd_1_1numeric limits_3_01cutlass_1_1half t_01_4.html#details)
| | |

|

Namespaces

| | | cutlass | | |

|

Macros

| | #define | CUTLASS_ENABLE_F16C 0 | | |

|

Functions

| | CUTLASS_HOST_DEVICE bool | cutlass::signbit (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE cutlass::half_t | cutlass::abs (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE bool | cutlass::isnan (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE bool | cutlass::isfinite (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE cutlass::half_t | cutlass::nanh (const char *) | | | | CUTLASS_HOST_DEVICE bool | cutlass::isinf (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE bool | cutlass::isnormal (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE int | cutlass::fpclassify (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE cutlass::half_t | cutlass::sqrt (cutlass::half_t const &h) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::copysign (half_t const &a, half_t const &b) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator== (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator!= (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator< (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator<= (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator> (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE bool | cutlass::operator>= (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator+ (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator- (half_t const &lhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator- (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator* (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator/ (half_t const &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator+= (half_t &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator-= (half_t &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator*= (half_t &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator/= (half_t &lhs, half_t const &rhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator++ (half_t &lhs) | | | | CUTLASS_HOST_DEVICE half_t & | cutlass::operator-- (half_t &lhs) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator++ (half_t &lhs, int) | | | | CUTLASS_HOST_DEVICE half_t | cutlass::operator-- (half_t &lhs, int) | | | | CUTLASS_HOST_DEVICE cutlass::half_t | operator""_hf (long double x) | | | | CUTLASS_HOST_DEVICE cutlass::half_t | operator""_hf (unsigned long long int x) | | |

Macro Definition Documentation

| #define CUTLASS_ENABLE_F16C 0 |

Function Documentation

| CUTLASS_HOST_DEVICE cutlass::half_t operator""_hf | ( | long double | x | ) | |

| CUTLASS_HOST_DEVICE cutlass::half_t operator""_hf | ( | unsigned long long int | x | ) | |

---

Generated by 1.8.11