MirrorHub/construct
0
0
Fork 0
mirror of https://github.com/matrix-construct/construct synced 2024-09-29 04:08:54 +02:00
Code Issues Releases Wiki Activity

ircd::simd: Use _Float16 / __fp16 for half-precision when available.

Browse source
This commit is contained in:
Jason Volk 2021-02-26 14:17:27 -08:00
parent 3a5bcbb0dd
commit 9d1316b412
1 changed files with 7 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
include/ircd/simd/type.h
View file

@ -69,7 +69,13 @@ namespace ircd
typedef uint64_t u64;
typedef uint128_t u128;
typedef int8_t f8; // ???
typedef int16_t f16; // ???
#if defined(HAVE__FLOAT16)
typedef _Float16 f16;
#elif defined(HAVE___FP16)
typedef __fp16 f16;
#else
#error "Missing half-precision floating point support."
#endif
typedef float f32;
typedef double f64;
typedef long double f128;