Round half-precision number to nearest integer value using round-to-nearest-away-from-zero. \param value binary representation of half-precision value \return half-precision bits for nearest integral value
| 1347 | /// \param value binary representation of half-precision value |
| 1348 | /// \return half-precision bits for nearest integral value |
| 1349 | inline uint16 round_half_up(uint16 value) |
| 1350 | { |
| 1351 | return round_half_impl<std::round_to_nearest, 0>(value); |
| 1352 | } |
| 1353 | /// \} |
| 1354 | |
| 1355 | struct functions; |