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

* Block: row and column expressions in the inner direction

Browse Source 
now have the Like1D flag.

* Big renaming:
  packetCoeff ---> packet
  VectorizableBit ---> PacketAccessBit
  Like1DArrayBit ---> LinearAccessBit
This commit is contained in:
Benoit Jacob
2008-06-16 14:54:31 +00:00
parent 9857764ae7
commit bb1f4e44f1
22 changed files with 192 additions and 188 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
Eigen/src/Core/Assign.h
View File

@@ -34,8 +34,8 @@
enum {
NoVectorization,
InnerVectorization,
Like1DVectorization,
SlicedVectorization
LinearVectorization,
SliceVectorization
};
enum {
@@ -56,18 +56,18 @@ private:
};
enum {
MightVectorize = (int(Derived::Flags) & int(OtherDerived::Flags) & VectorizableBit)
MightVectorize = (int(Derived::Flags) & int(OtherDerived::Flags) & PacketAccessBit)
&& ((int(Derived::Flags)&RowMajorBit)==(int(OtherDerived::Flags)&RowMajorBit)),
MayInnerVectorize = MightVectorize && InnerSize!=Dynamic && int(InnerSize)%int(PacketSize)==0,
MayLike1DVectorize = MightVectorize && (int(Derived::Flags) & int(OtherDerived::Flags) & Like1DArrayBit),
MaySlicedVectorize = MightVectorize && InnerSize==Dynamic
MayLinearVectorize = MightVectorize && (int(Derived::Flags) & int(OtherDerived::Flags) & LinearAccessBit),
MaySliceVectorize = MightVectorize && InnerSize==Dynamic
};
public:
enum {
Vectorization = MayInnerVectorize ? InnerVectorization
: MayLike1DVectorize ? Like1DVectorization
: MaySlicedVectorize ? SlicedVectorization
: MayLinearVectorize ? LinearVectorization
: MaySliceVectorize ? SliceVectorization
: NoVectorization
};
@@ -86,7 +86,7 @@ public:
: MayUnrollInner ? InnerUnrolling
: NoUnrolling
)
: int(Vectorization) == int(Like1DVectorization)
: int(Vectorization) == int(LinearVectorization)
? ( MayUnrollCompletely ? CompleteUnrolling : NoUnrolling )
: NoUnrolling
};
@@ -162,7 +162,7 @@ struct ei_assign_innervec_CompleteUnrolling
inline static void run(Derived1 &dst, const Derived2 &src)
{
dst.template writePacketCoeff<Aligned>(row, col, src.template packetCoeff<Aligned>(row, col));
dst.template writePacket<Aligned>(row, col, src.template packet<Aligned>(row, col));
ei_assign_innervec_CompleteUnrolling<Derived1, Derived2,
Index+ei_packet_traits<typename Derived1::Scalar>::size, Stop>::run(dst, src);
}
@@ -181,7 +181,7 @@ struct ei_assign_innervec_InnerUnrolling
{
const int row = int(Derived1::Flags)&RowMajorBit ? row_or_col : Index;
const int col = int(Derived1::Flags)&RowMajorBit ? Index : row_or_col;
dst.template writePacketCoeff<Aligned>(row, col, src.template packetCoeff<Aligned>(row, col));
dst.template writePacket<Aligned>(row, col, src.template packet<Aligned>(row, col));
ei_assign_innervec_InnerUnrolling<Derived1, Derived2,
Index+ei_packet_traits<typename Derived1::Scalar>::size, Stop>::run(dst, src, row_or_col);
}
@@ -267,7 +267,7 @@ struct ei_assign_impl<Derived1, Derived2, InnerVectorization, NoUnrolling>
{
const int row = rowMajor ? j : i;
const int col = rowMajor ? i : j;
dst.template writePacketCoeff<Aligned>(row, col, src.template packetCoeff<Aligned>(row, col));
dst.template writePacket<Aligned>(row, col, src.template packet<Aligned>(row, col));
}
}
}
@@ -298,11 +298,11 @@ struct ei_assign_impl<Derived1, Derived2, InnerVectorization, InnerUnrolling>
};
/***************************
*** Like1D vectorization ***
*** Linear vectorization ***
***************************/
template<typename Derived1, typename Derived2>
struct ei_assign_impl<Derived1, Derived2, Like1DVectorization, NoUnrolling>
struct ei_assign_impl<Derived1, Derived2, LinearVectorization, NoUnrolling>
{
static void run(Derived1 &dst, const Derived2 &src)
{
@@ -320,7 +320,7 @@ struct ei_assign_impl<Derived1, Derived2, Like1DVectorization, NoUnrolling>
// FIXME the following is not really efficient
const int row = rowMajor ? index/innerSize : index%innerSize;
const int col = rowMajor ? index%innerSize : index/innerSize;
dst.template writePacketCoeff<Aligned>(row, col, src.template packetCoeff<Aligned>(row, col));
dst.template writePacket<Aligned>(row, col, src.template packet<Aligned>(row, col));
}
// now we must do the rest without vectorization.
@@ -347,7 +347,7 @@ struct ei_assign_impl<Derived1, Derived2, Like1DVectorization, NoUnrolling>
};
template<typename Derived1, typename Derived2>
struct ei_assign_impl<Derived1, Derived2, Like1DVectorization, CompleteUnrolling>
struct ei_assign_impl<Derived1, Derived2, LinearVectorization, CompleteUnrolling>
{
inline static void run(Derived1 &dst, const Derived2 &src)
{
@@ -375,16 +375,16 @@ struct ei_assign_impl<Derived1, Derived2, Like1DVectorization, CompleteUnrolling
}
};
/***************************
*** Sliced vectorization ***
/**************************
*** Slice vectorization ***
***************************/
template<typename Derived1, typename Derived2>
struct ei_assign_impl<Derived1, Derived2, SlicedVectorization, NoUnrolling>
struct ei_assign_impl<Derived1, Derived2, SliceVectorization, NoUnrolling>
{
static void run(Derived1 &dst, const Derived2 &src)
{
//FIXME unimplemented
//FIXME unimplemented, so for now we fall back to non-vectorized path
ei_assign_impl<Derived1, Derived2, NoVectorization, NoUnrolling>::run(dst, src);
}
};