Improve slice-vectorization logic for redux (significant speed-up for reduxion of blocks)

This commit is contained in:
Gael Guennebaud
2018-09-21 13:45:56 +02:00
parent 3ec2985914
commit b00e48a867
2 changed files with 25 additions and 6 deletions

View File

@@ -32,14 +32,20 @@ public:
PacketSize = unpacket_traits<PacketType>::size,
InnerMaxSize = int(Evaluator::IsRowMajor)
? Evaluator::MaxColsAtCompileTime
: Evaluator::MaxRowsAtCompileTime
: Evaluator::MaxRowsAtCompileTime,
OuterMaxSize = int(Evaluator::IsRowMajor)
? Evaluator::MaxRowsAtCompileTime
: Evaluator::MaxColsAtCompileTime,
SliceVectorizedWork = int(InnerMaxSize)==Dynamic ? Dynamic
: int(OuterMaxSize)==Dynamic ? (int(InnerMaxSize)>=int(PacketSize) ? Dynamic : 0)
: (int(InnerMaxSize)/int(PacketSize)) * int(OuterMaxSize)
};
enum {
MightVectorize = (int(Evaluator::Flags)&ActualPacketAccessBit)
&& (functor_traits<Func>::PacketAccess),
MayLinearVectorize = bool(MightVectorize) && (int(Evaluator::Flags)&LinearAccessBit),
MaySliceVectorize = bool(MightVectorize) && int(InnerMaxSize)>=3*PacketSize
MaySliceVectorize = bool(MightVectorize) && (int(SliceVectorizedWork)==Dynamic || int(SliceVectorizedWork)>=3)
};
public:
@@ -69,12 +75,15 @@ public:
EIGEN_DEBUG_VAR(Evaluator::Flags)
std::cerr.unsetf(std::ios::hex);
EIGEN_DEBUG_VAR(InnerMaxSize)
EIGEN_DEBUG_VAR(OuterMaxSize)
EIGEN_DEBUG_VAR(SliceVectorizedWork)
EIGEN_DEBUG_VAR(PacketSize)
EIGEN_DEBUG_VAR(MightVectorize)
EIGEN_DEBUG_VAR(MayLinearVectorize)
EIGEN_DEBUG_VAR(MaySliceVectorize)
EIGEN_DEBUG_VAR(Traversal)
EIGEN_DEBUG_VAR(UnrollingLimit)
EIGEN_DEBUG_VAR(Unrolling)
std::cerr << std::endl;
}