From e763fcd09e620300226ca22d152b94867123b603 Mon Sep 17 00:00:00 2001 From: Gustavo Lima Chaves Date: Wed, 19 Dec 2018 14:24:44 -0800 Subject: Introducing "vectorized" byte on unpacket_traits structs This is a preparation to a change on gebp_traits, where a new template argument will be introduced to dictate the packet size, so it won't be bound to the current/max packet size only anymore. By having packet types defined early on gebp_traits, one has now to act on packet types, not scalars anymore, for the enum values defined on that class. One approach for reaching the vectorizable/size properties one needs there could be getting the packet's scalar again with unpacket_traits<>, then the size/Vectorizable enum entries from packet_traits<>. It turns out guards like "#ifndef EIGEN_VECTORIZE_AVX512" at AVX/PacketMath.h will hide smaller packet variations of packet_traits<> for some types (and it makes sense to keep that). In other words, one can't go back to the scalar and create a new PacketType, as this will always lead to the maximum packet type for the architecture. The less costly/invasive solution for that, thus, is to add the vectorizable info on every unpacket_traits struct as well. --- Eigen/src/Core/arch/NEON/PacketMath.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'Eigen/src/Core/arch/NEON/PacketMath.h') diff --git a/Eigen/src/Core/arch/NEON/PacketMath.h b/Eigen/src/Core/arch/NEON/PacketMath.h index ed3cec88a..ca4f2bf94 100644 --- a/Eigen/src/Core/arch/NEON/PacketMath.h +++ b/Eigen/src/Core/arch/NEON/PacketMath.h @@ -145,13 +145,13 @@ template<> struct unpacket_traits typedef float type; typedef Packet4f half; typedef Packet4i integer_packet; - enum {size=4, alignment=Aligned16}; + enum {size=4, alignment=Aligned16, vectorizable=true}; }; template<> struct unpacket_traits { typedef int32_t type; typedef Packet4i half; - enum {size=4, alignment=Aligned16}; + enum {size=4, alignment=Aligned16, vectorizable=true}; }; template<> EIGEN_STRONG_INLINE Packet4f pset1(const float& from) { return vdupq_n_f32(from); } @@ -650,7 +650,7 @@ template<> struct packet_traits : default_packet_traits }; }; -template<> struct unpacket_traits { typedef double type; enum {size=2, alignment=Aligned16}; typedef Packet2d half; }; +template<> struct unpacket_traits { typedef double type; enum {size=2, alignment=Aligned16, vectorizable=true}; typedef Packet2d half; }; template<> EIGEN_STRONG_INLINE Packet2d pset1(const double& from) { return vdupq_n_f64(from); } -- cgit v1.2.3