devtools/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2026-04-10 11:34:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix Half NaN definition and test.

Browse Source 
The `half_float` test was failing with `-mcpu=cortex-a55` (native `__fp16`) due
to a bad NaN bit-pattern comparison (in the case of casting a float to `__fp16`,
the signaling `NaN` is quieted). There was also an inconsistency between
`numeric_limits<half>::quiet_NaN()` and `NumTraits::quiet_NaN()`.  Here we
correct the inconsistency and compare NaNs according to the IEEE 754
definition.

Also modified the `bfloat16_float` test to match.

Tested with `cortex-a53` and `cortex-a55`.
This commit is contained in:
Antonio Sanchez
2020-11-23 14:13:59 -08:00
parent 4cf01d2cf5
commit 38abf2be42
3 changed files with 35 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
Eigen/src/Core/arch/Default/Half.h
View File

@@ -200,7 +200,7 @@ struct numeric_limits<Eigen::half> {
static Eigen::half round_error() { return Eigen::half(0.5); }
static Eigen::half infinity() { return Eigen::half_impl::raw_uint16_to_half(0x7c00); }
static Eigen::half quiet_NaN() { return Eigen::half_impl::raw_uint16_to_half(0x7e00); }
static Eigen::half signaling_NaN() { return Eigen::half_impl::raw_uint16_to_half(0x7e00); }
static Eigen::half signaling_NaN() { return Eigen::half_impl::raw_uint16_to_half(0x7d00); }
static Eigen::half denorm_min() { return Eigen::half_impl::raw_uint16_to_half(0x1); }
};
@@ -744,7 +744,7 @@ template<> struct NumTraits<Eigen::half>
return half_impl::raw_uint16_to_half(0x7c00);
}
EIGEN_DEVICE_FUNC EIGEN_CONSTEXPR static EIGEN_STRONG_INLINE Eigen::half quiet_NaN() {
return half_impl::raw_uint16_to_half(0x7c01);
return half_impl::raw_uint16_to_half(0x7e00);
}
};