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

E2K: initial support of LCC MCST compiler for the Elbrus 2000 CPU architecture

Browse Source
This commit is contained in:
Ramil Sattarov
2022-02-23 17:07:34 +00:00
committed by Antonio Sánchez
parent ae86a146b1
commit f9b7564faa
3 changed files with 14 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
Eigen/src/Core/arch/SSE/PacketMath.h
View File

@@ -32,7 +32,7 @@ namespace internal {
#endif
#endif
#if ((defined EIGEN_VECTORIZE_AVX) && (EIGEN_COMP_GNUC_STRICT || EIGEN_COMP_MINGW) && (__GXX_ABI_VERSION < 1004)) || EIGEN_OS_QNX
#if ((defined EIGEN_VECTORIZE_AVX) && (EIGEN_COMP_GNUC_STRICT || EIGEN_COMP_MINGW || EIGEN_COMP_LCC) && (__GXX_ABI_VERSION < 1004)) || EIGEN_OS_QNX
// With GCC's default ABI version, a __m128 or __m256 are the same types and therefore we cannot
// have overloads for both types without linking error.
// One solution is to increase ABI version using -fabi-version=4 (or greater).
@@ -443,7 +443,7 @@ template<> EIGEN_STRONG_INLINE Packet16b pcmp_eq(const Packet16b& a, const Packe
template<> EIGEN_STRONG_INLINE Packet4i pcmp_le(const Packet4i& a, const Packet4i& b) { return por(pcmp_lt(a,b), pcmp_eq(a,b)); }
template<> EIGEN_STRONG_INLINE Packet4f pmin<Packet4f>(const Packet4f& a, const Packet4f& b) {
#if EIGEN_COMP_GNUC && EIGEN_COMP_GNUC < 63
#if EIGEN_COMP_GNUC_STRICT && EIGEN_COMP_GNUC < 63
// There appears to be a bug in GCC, by which the optimizer may
// flip the argument order in calls to _mm_min_ps, so we have to
// resort to inline ASM here. This is supposed to be fixed in gcc6.3,
@@ -462,7 +462,7 @@ template<> EIGEN_STRONG_INLINE Packet4f pmin<Packet4f>(const Packet4f& a, const
#endif
}
template<> EIGEN_STRONG_INLINE Packet2d pmin<Packet2d>(const Packet2d& a, const Packet2d& b) {
#if EIGEN_COMP_GNUC && EIGEN_COMP_GNUC < 63
#if EIGEN_COMP_GNUC_STRICT && EIGEN_COMP_GNUC < 63
// There appears to be a bug in GCC, by which the optimizer may
// flip the argument order in calls to _mm_min_pd, so we have to
// resort to inline ASM here. This is supposed to be fixed in gcc6.3,
@@ -493,7 +493,7 @@ template<> EIGEN_STRONG_INLINE Packet4i pmin<Packet4i>(const Packet4i& a, const
template<> EIGEN_STRONG_INLINE Packet4f pmax<Packet4f>(const Packet4f& a, const Packet4f& b) {
#if EIGEN_COMP_GNUC && EIGEN_COMP_GNUC < 63
#if EIGEN_COMP_GNUC_STRICT && EIGEN_COMP_GNUC < 63
// There appears to be a bug in GCC, by which the optimizer may
// flip the argument order in calls to _mm_max_ps, so we have to
// resort to inline ASM here. This is supposed to be fixed in gcc6.3,