GitHub-Proxy/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-08-11 11:19:02 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix floor/ceil for NEON fp16.

Browse Source 
Forgot to test this.  Fixes bug introduced in !416.
This commit is contained in:
Antonio Sanchez 2021-02-25 20:39:56 -08:00
parent 5529db7524
commit e19829c3b0
2 changed files with 59 additions and 57 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

112
Eigen/src/Core/arch/NEON/PacketMath.h
View File

@ -4182,62 +4182,6 @@ EIGEN_STRONG_INLINE Packet4hf pcmp_lt_or_nan<Packet4hf>(const Packet4hf& a, cons
return vreinterpret_f16_u16(vmvn_u16(vcge_f16(a, b)));
}
template <>
EIGEN_STRONG_INLINE Packet8hf pfloor<Packet8hf>(const Packet8hf& a) {
const Packet8hf cst_1 = pset1<Packet8hf>(Eigen::half(1.0f));
// Round to nearest.
Packet8hf tmp = vcvtq_f16_s16(vcvtq_s16_f16(a));
// If greater, substract one.
uint16x8_t mask = vcgtq_f16(tmp, a);
mask = vandq_u16(mask, vreinterpretq_u16_f16(cst_1));
tmp = vsubq_f16(tmp, vreinterpretq_f16_u16(mask));
// Handle saturation cases.
EIGEN_CONSTEXPR Packet8hf cst_max = pset1<Packet8hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet4hf pfloor<Packet4hf>(const Packet4hf& a) {
const Packet4hf cst_1 = pset1<Packet4hf>(Eigen::half(1.0f));
// Round to nearest.
Packet4hf tmp = vcvt_f16_s16(vcvt_s16_f16(a));
// If greater, substract one.
uint16x4_t mask = vcgt_f16(tmp, a);
mask = vand_u16(mask, vreinterpret_u16_f16(cst_1));
tmp = vsub_f16(tmp, vreinterpret_f16_u16(mask));
// Handle saturation cases.
EIGEN_CONSTEXPR Packet4hf cst_max = pset1<Packet4hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet8hf pceil<Packet8hf>(const Packet8hf& a) {
const Packet8hf cst_1 = pset1<Packet8hf>(Eigen::half(1.0f));
// Round to nearest.
Packet8hf tmp = vcvtq_f16_s16(vcvtq_s16_f16(a));
// If smaller, add one.
uint16x8_t mask = vcltq_f16(tmp, a);
mask = vandq_u16(mask, vreinterpretq_u16_f16(cst_1));
tmp = vaddq_f16(tmp, vreinterpretq_f16_u16(mask));
// Handle saturation cases.
EIGEN_CONSTEXPR Packet8hf cst_max = pset1<Packet8hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet4hf pceil<Packet4hf>(const Packet4hf& a) {
const Packet4hf cst_1 = pset1<Packet4hf>(Eigen::half(1.0f));
// Round to nearest.
Packet4hf tmp = vcvt_f16_s16(vcvt_s16_f16(a));
// If smaller, add one.
uint16x4_t mask = vclt_f16(tmp, a);
mask = vand_u16(mask, vreinterpret_u16_f16(cst_1));
tmp = vadd_f16(tmp, vreinterpret_f16_u16(mask));
// Handle saturation cases.
EIGEN_CONSTEXPR Packet4hf cst_max = pset1<Packet4hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet8hf psqrt<Packet8hf>(const Packet8hf& a) {
return vsqrtq_f16(a);
@ -4472,6 +4416,62 @@ EIGEN_STRONG_INLINE Packet4hf pabs<Packet4hf>(const Packet4hf& a) {
return vabs_f16(a);
}
template <>
EIGEN_STRONG_INLINE Packet8hf pfloor<Packet8hf>(const Packet8hf& a) {
const Packet8hf cst_1 = pset1<Packet8hf>(Eigen::half(1.0f));
// Round to nearest.
Packet8hf tmp = vcvtq_f16_s16(vcvtq_s16_f16(a));
// If greater, substract one.
uint16x8_t mask = vcgtq_f16(tmp, a);
mask = vandq_u16(mask, vreinterpretq_u16_f16(cst_1));
tmp = vsubq_f16(tmp, vreinterpretq_f16_u16(mask));
// Handle saturation cases.
const Packet8hf cst_max = pset1<Packet8hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet4hf pfloor<Packet4hf>(const Packet4hf& a) {
const Packet4hf cst_1 = pset1<Packet4hf>(Eigen::half(1.0f));
// Round to nearest.
Packet4hf tmp = vcvt_f16_s16(vcvt_s16_f16(a));
// If greater, substract one.
uint16x4_t mask = vcgt_f16(tmp, a);
mask = vand_u16(mask, vreinterpret_u16_f16(cst_1));
tmp = vsub_f16(tmp, vreinterpret_f16_u16(mask));
// Handle saturation cases.
const Packet4hf cst_max = pset1<Packet4hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet8hf pceil<Packet8hf>(const Packet8hf& a) {
const Packet8hf cst_1 = pset1<Packet8hf>(Eigen::half(1.0f));
// Round to nearest.
Packet8hf tmp = vcvtq_f16_s16(vcvtq_s16_f16(a));
// If smaller, add one.
uint16x8_t mask = vcltq_f16(tmp, a);
mask = vandq_u16(mask, vreinterpretq_u16_f16(cst_1));
tmp = vaddq_f16(tmp, vreinterpretq_f16_u16(mask));
// Handle saturation cases.
const Packet8hf cst_max = pset1<Packet8hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Packet4hf pceil<Packet4hf>(const Packet4hf& a) {
const Packet4hf cst_1 = pset1<Packet4hf>(Eigen::half(1.0f));
// Round to nearest.
Packet4hf tmp = vcvt_f16_s16(vcvt_s16_f16(a));
// If smaller, add one.
uint16x4_t mask = vclt_f16(tmp, a);
mask = vand_u16(mask, vreinterpret_u16_f16(cst_1));
tmp = vadd_f16(tmp, vreinterpret_f16_u16(mask));
// Handle saturation cases.
const Packet4hf cst_max = pset1<Packet4hf>(static_cast<Eigen::half>(NumTraits<int16_t>::highest()));
return pselect(pcmp_lt(pabs(a), cst_max), tmp, a);
}
template <>
EIGEN_STRONG_INLINE Eigen::half predux<Packet8hf>(const Packet8hf& a) {
float16x4_t a_lo, a_hi, sum;

4
test/packetmath.cpp
View File

@ -551,7 +551,9 @@ void packetmath_real() {
if (PacketTraits::HasRound || PacketTraits::HasCeil || PacketTraits::HasFloor || PacketTraits::HasRint) {
typedef typename internal::make_integer<Scalar>::type IntType;
// Start with values that cannot fit inside an integer, work down to less than one.
Scalar val = Scalar(2) * static_cast<Scalar>(NumTraits<IntType>::highest());
Scalar val = numext::mini(
Scalar(2) * static_cast<Scalar>(NumTraits<IntType>::highest()),
NumTraits<Scalar>::highest());
std::vector<Scalar> values;
while (val > Scalar(0.25)) {
// Cover both even and odd, positive and negative cases.