GitHub-Proxy/eigen
1
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2025-08-13 12:19:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix #76, MayLinearVectorize depends on MaxSizeAtCompileTime and not on MaxInnerSize

Browse Source
This commit is contained in:
Gael Guennebaud 2009-12-23 13:19:28 +01:00
parent 0b5853d917
commit 24302ad5cd
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
Eigen/src/Core/Assign.h
View File

@ -49,6 +49,7 @@ private:
InnerMaxSize = int(Derived::Flags)&RowMajorBit InnerMaxSize = int(Derived::Flags)&RowMajorBit
? Derived::MaxColsAtCompileTime ? Derived::MaxColsAtCompileTime
: Derived::MaxRowsAtCompileTime, : Derived::MaxRowsAtCompileTime,
MaxSizeAtCompileTime = ei_size_at_compile_time<Derived::MaxColsAtCompileTime,Derived::MaxRowsAtCompileTime>::ret,
PacketSize = ei_packet_traits<typename Derived::Scalar>::size PacketSize = ei_packet_traits<typename Derived::Scalar>::size
}; };
@ -60,9 +61,9 @@ private:
&& int(DstIsAligned) && int(SrcIsAligned), && int(DstIsAligned) && int(SrcIsAligned),
MayLinearize = StorageOrdersAgree && (int(Derived::Flags) & int(OtherDerived::Flags) & LinearAccessBit), MayLinearize = StorageOrdersAgree && (int(Derived::Flags) & int(OtherDerived::Flags) & LinearAccessBit),
MayLinearVectorize = MightVectorize && MayLinearize MayLinearVectorize = MightVectorize && MayLinearize
&& (DstIsAligned || InnerMaxSize == Dynamic), && (DstIsAligned || MaxSizeAtCompileTime == Dynamic),
/* If the destination isn't aligned, we have to do runtime checks and we don't unroll, /* If the destination isn't aligned, we have to do runtime checks and we don't unroll,
so it's only good for large enough sizes. See remark below about InnerMaxSize. */ so it's only good for large enough sizes. */
MaySliceVectorize = MightVectorize && int(InnerMaxSize)>=3*PacketSize MaySliceVectorize = MightVectorize && int(InnerMaxSize)>=3*PacketSize
/* slice vectorization can be slow, so we only want it if the slices are big, which is /* slice vectorization can be slow, so we only want it if the slices are big, which is
indicated by InnerMaxSize rather than InnerSize, think of the case of a dynamic block indicated by InnerMaxSize rather than InnerSize, think of the case of a dynamic block