use __hmax, __hmin (#1813)

use hmax, hmin (#1813)
d208adfc · Umang Yadav · GitHub · 41ba30d5 · d208adfc
Unverified Commit d208adfc authored Jun 15, 2023 by Umang Yadav Committed by GitHub Jun 15, 2023
Show whitespace changes
Inline Side-by-side

Showing with 2 additions and 4 deletions

src/targets/gpu/kernels/include/migraphx/kernels/math.hpp src/targets/gpu/kernels/include/migraphx/kernels/math.hpp +2 -4

No files found.
--- a/src/targets/gpu/kernels/include/migraphx/kernels/math.hpp
+++ b/src/targets/gpu/kernels/include/migraphx/kernels/math.hpp
@@ -188,10 +188,8 @@ MIGRAPHX_DEVICE_MATH_BINARY_FOR(float, max, ::max)
 MIGRAPHX_DEVICE_MATH_BINARY_FOR(float, min, ::min)
 MIGRAPHX_DEVICE_MATH_BINARY_FOR(double, max, ::max)
 MIGRAPHX_DEVICE_MATH_BINARY_FOR(double, min, ::min)
-// Add overloads for half that calls the float version, this should use "hmax" and "hmin" once
-// perf CI docker is upgraded to rocm-5.5
-MIGRAPHX_DEVICE_MATH_BINARY_FOR(migraphx::half, max, ::fmaxf)
-MIGRAPHX_DEVICE_MATH_BINARY_FOR(migraphx::half, min, ::fminf)
+MIGRAPHX_DEVICE_MATH_BINARY_FOR(migraphx::half, max, ::__hmax)
+MIGRAPHX_DEVICE_MATH_BINARY_FOR(migraphx::half, min, ::__hmin)

 template <class T, MIGRAPHX_REQUIRES(not is_any_vec<T>())>
 constexpr auto max(const T& a, const T& b)