diff options
author | Sergey Sharybin <sergey.vfx@gmail.com> | 2012-05-09 12:33:05 +0400 |
---|---|---|
committer | Sergey Sharybin <sergey.vfx@gmail.com> | 2012-05-09 12:33:05 +0400 |
commit | 91227ff6d0d49b67f4b332deeb29052502280981 (patch) | |
tree | e7dc3d4cc7baeb759be0f3d8907f6a93107b2fd7 /extern/Eigen3/Eigen/src/Core | |
parent | e132098599a8562bec7bff9aeaa69a7891d50708 (diff) |
Update Eigen3 library, would be needed for some further integraiton.
Diffstat (limited to 'extern/Eigen3/Eigen/src/Core')
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/Array.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/Block.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/DenseBase.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/Map.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/MapBase.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/Matrix.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/MatrixBase.h | 3 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/PlainObjectBase.h | 23 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/ProductBase.h | 14 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/Replicate.h | 9 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/SolveTriangular.h | 2 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/arch/NEON/Complex.h | 4 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/arch/NEON/PacketMath.h | 58 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/products/GeneralBlockPanelKernel.h | 21 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/util/Macros.h | 36 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/util/Memory.h | 54 | ||||
-rw-r--r-- | extern/Eigen3/Eigen/src/Core/util/XprHelper.h | 3 |
17 files changed, 153 insertions, 98 deletions
diff --git a/extern/Eigen3/Eigen/src/Core/Array.h b/extern/Eigen3/Eigen/src/Core/Array.h index a3a2167ad3e..a11fb1b53d5 100644 --- a/extern/Eigen3/Eigen/src/Core/Array.h +++ b/extern/Eigen3/Eigen/src/Core/Array.h @@ -68,10 +68,8 @@ class Array friend struct internal::conservative_resize_like_impl; using Base::m_storage; + public: - enum { NeedsToAlign = (!(Options&DontAlign)) - && SizeAtCompileTime!=Dynamic && ((static_cast<int>(sizeof(Scalar))*SizeAtCompileTime)%16)==0 }; - EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF(NeedsToAlign) using Base::base; using Base::coeff; diff --git a/extern/Eigen3/Eigen/src/Core/Block.h b/extern/Eigen3/Eigen/src/Core/Block.h index 2b251bc2ca9..d470bc13400 100644 --- a/extern/Eigen3/Eigen/src/Core/Block.h +++ b/extern/Eigen3/Eigen/src/Core/Block.h @@ -94,7 +94,7 @@ struct traits<Block<XprType, BlockRows, BlockCols, InnerPanel, HasDirectAccess> MaskPacketAccessBit = (InnerSize == Dynamic || (InnerSize % packet_traits<Scalar>::size) == 0) && (InnerStrideAtCompileTime == 1) ? PacketAccessBit : 0, - MaskAlignedBit = (InnerPanel && (OuterStrideAtCompileTime!=Dynamic) && ((OuterStrideAtCompileTime % packet_traits<Scalar>::size) == 0)) ? AlignedBit : 0, + MaskAlignedBit = (InnerPanel && (OuterStrideAtCompileTime!=Dynamic) && (((OuterStrideAtCompileTime * int(sizeof(Scalar))) % 16) == 0)) ? AlignedBit : 0, FlagsLinearAccessBit = (RowsAtCompileTime == 1 || ColsAtCompileTime == 1) ? LinearAccessBit : 0, FlagsLvalueBit = is_lvalue<XprType>::value ? LvalueBit : 0, FlagsRowMajorBit = IsRowMajor ? RowMajorBit : 0, @@ -342,7 +342,7 @@ class Block<XprType,BlockRows,BlockCols, InnerPanel,true> } const typename XprType::Nested m_xpr; - int m_outerStride; + Index m_outerStride; }; diff --git a/extern/Eigen3/Eigen/src/Core/DenseBase.h b/extern/Eigen3/Eigen/src/Core/DenseBase.h index 838fa40307a..920904f243a 100644 --- a/extern/Eigen3/Eigen/src/Core/DenseBase.h +++ b/extern/Eigen3/Eigen/src/Core/DenseBase.h @@ -169,8 +169,8 @@ template<typename Derived> class DenseBase IsRowMajor = int(Flags) & RowMajorBit, /**< True if this expression has row-major storage order. */ - InnerSizeAtCompileTime = int(IsVectorAtCompileTime) ? SizeAtCompileTime - : int(IsRowMajor) ? ColsAtCompileTime : RowsAtCompileTime, + InnerSizeAtCompileTime = int(IsVectorAtCompileTime) ? int(SizeAtCompileTime) + : int(IsRowMajor) ? int(ColsAtCompileTime) : int(RowsAtCompileTime), CoeffReadCost = internal::traits<Derived>::CoeffReadCost, /**< This is a rough measure of how expensive it is to read one coefficient from diff --git a/extern/Eigen3/Eigen/src/Core/Map.h b/extern/Eigen3/Eigen/src/Core/Map.h index dd0673609c5..2bf80b3af3d 100644 --- a/extern/Eigen3/Eigen/src/Core/Map.h +++ b/extern/Eigen3/Eigen/src/Core/Map.h @@ -102,7 +102,7 @@ struct traits<Map<PlainObjectType, MapOptions, StrideType> > || HasNoOuterStride || ( OuterStrideAtCompileTime!=Dynamic && ((static_cast<int>(sizeof(Scalar))*OuterStrideAtCompileTime)%16)==0 ) ), - Flags0 = TraitsBase::Flags, + Flags0 = TraitsBase::Flags & (~NestByRefBit), Flags1 = IsAligned ? (int(Flags0) | AlignedBit) : (int(Flags0) & ~AlignedBit), Flags2 = (bool(HasNoStride) || bool(PlainObjectType::IsVectorAtCompileTime)) ? int(Flags1) : int(Flags1 & ~LinearAccessBit), @@ -120,7 +120,6 @@ template<typename PlainObjectType, int MapOptions, typename StrideType> class Ma public: typedef MapBase<Map> Base; - EIGEN_DENSE_PUBLIC_INTERFACE(Map) typedef typename Base::PointerType PointerType; @@ -181,7 +180,6 @@ template<typename PlainObjectType, int MapOptions, typename StrideType> class Ma PlainObjectType::Base::_check_template_params(); } - EIGEN_INHERIT_ASSIGNMENT_OPERATORS(Map) protected: diff --git a/extern/Eigen3/Eigen/src/Core/MapBase.h b/extern/Eigen3/Eigen/src/Core/MapBase.h index c23bcbfdcca..9426e2d24dd 100644 --- a/extern/Eigen3/Eigen/src/Core/MapBase.h +++ b/extern/Eigen3/Eigen/src/Core/MapBase.h @@ -170,8 +170,8 @@ template<typename Derived> class MapBase<Derived, ReadOnlyAccessors> EIGEN_STATIC_ASSERT(EIGEN_IMPLIES(internal::traits<Derived>::Flags&PacketAccessBit, internal::inner_stride_at_compile_time<Derived>::ret==1), PACKET_ACCESS_REQUIRES_TO_HAVE_INNER_STRIDE_FIXED_TO_1); - eigen_assert(EIGEN_IMPLIES(internal::traits<Derived>::Flags&AlignedBit, (size_t(m_data) % (sizeof(Scalar)*internal::packet_traits<Scalar>::size)) == 0) - && "data is not aligned"); + eigen_assert(EIGEN_IMPLIES(internal::traits<Derived>::Flags&AlignedBit, (size_t(m_data) % 16) == 0) + && "data is not aligned"); } PointerType m_data; diff --git a/extern/Eigen3/Eigen/src/Core/Matrix.h b/extern/Eigen3/Eigen/src/Core/Matrix.h index 44de22cb4d5..982c9256af0 100644 --- a/extern/Eigen3/Eigen/src/Core/Matrix.h +++ b/extern/Eigen3/Eigen/src/Core/Matrix.h @@ -153,10 +153,6 @@ class Matrix typedef typename Base::PlainObject PlainObject; - enum { NeedsToAlign = (!(Options&DontAlign)) - && SizeAtCompileTime!=Dynamic && ((static_cast<int>(sizeof(Scalar))*SizeAtCompileTime)%16)==0 }; - EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF(NeedsToAlign) - using Base::base; using Base::coeffRef; diff --git a/extern/Eigen3/Eigen/src/Core/MatrixBase.h b/extern/Eigen3/Eigen/src/Core/MatrixBase.h index db156f6e9d0..62877bce09e 100644 --- a/extern/Eigen3/Eigen/src/Core/MatrixBase.h +++ b/extern/Eigen3/Eigen/src/Core/MatrixBase.h @@ -250,7 +250,8 @@ template<typename Derived> class MatrixBase // huuuge hack. make Eigen2's matrix.part<Diagonal>() work in eigen3. Problem: Diagonal is now a class template instead // of an integer constant. Solution: overload the part() method template wrt template parameters list. - template<template<typename T, int n> class U> + // Note: replacing next line by "template<template<typename T, int n> class U>" produces a mysterious error C2082 in MSVC. + template<template<typename, int> class U> const DiagonalWrapper<ConstDiagonalReturnType> part() const { return diagonal().asDiagonal(); } #endif // EIGEN2_SUPPORT diff --git a/extern/Eigen3/Eigen/src/Core/PlainObjectBase.h b/extern/Eigen3/Eigen/src/Core/PlainObjectBase.h index c70db92479a..612254e9da9 100644 --- a/extern/Eigen3/Eigen/src/Core/PlainObjectBase.h +++ b/extern/Eigen3/Eigen/src/Core/PlainObjectBase.h @@ -34,6 +34,19 @@ namespace internal { +template<typename Index> +EIGEN_ALWAYS_INLINE void check_rows_cols_for_overflow(Index rows, Index cols) +{ + // http://hg.mozilla.org/mozilla-central/file/6c8a909977d3/xpcom/ds/CheckedInt.h#l242 + // we assume Index is signed + Index max_index = (size_t(1) << (8 * sizeof(Index) - 1)) - 1; // assume Index is signed + bool error = (rows < 0 || cols < 0) ? true + : (rows == 0 || cols == 0) ? false + : (rows > max_index / cols); + if (error) + throw_std_bad_alloc(); +} + template <typename Derived, typename OtherDerived = Derived, bool IsVector = static_cast<bool>(Derived::IsVectorAtCompileTime)> struct conservative_resize_like_impl; template<typename MatrixTypeA, typename MatrixTypeB, bool SwapPointers> struct matrix_swap_impl; @@ -84,14 +97,12 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type template<typename StrideType> struct StridedConstMapType { typedef Eigen::Map<const Derived, Unaligned, StrideType> type; }; template<typename StrideType> struct StridedAlignedMapType { typedef Eigen::Map<Derived, Aligned, StrideType> type; }; template<typename StrideType> struct StridedConstAlignedMapType { typedef Eigen::Map<const Derived, Aligned, StrideType> type; }; - protected: DenseStorage<Scalar, Base::MaxSizeAtCompileTime, Base::RowsAtCompileTime, Base::ColsAtCompileTime, Options> m_storage; public: - enum { NeedsToAlign = (!(Options&DontAlign)) - && SizeAtCompileTime!=Dynamic && ((static_cast<int>(sizeof(Scalar))*SizeAtCompileTime)%16)==0 }; + enum { NeedsToAlign = SizeAtCompileTime != Dynamic && (internal::traits<Derived>::Flags & AlignedBit) != 0 }; EIGEN_MAKE_ALIGNED_OPERATOR_NEW_IF(NeedsToAlign) Base& base() { return *static_cast<Base*>(this); } @@ -200,11 +211,13 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type EIGEN_STRONG_INLINE void resize(Index rows, Index cols) { #ifdef EIGEN_INITIALIZE_MATRICES_BY_ZERO + internal::check_rows_cols_for_overflow(rows, cols); Index size = rows*cols; bool size_changed = size != this->size(); m_storage.resize(size, rows, cols); if(size_changed) EIGEN_INITIALIZE_BY_ZERO_IF_THAT_OPTION_IS_ENABLED #else + internal::check_rows_cols_for_overflow(rows, cols); m_storage.resize(rows*cols, rows, cols); #endif } @@ -273,6 +286,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type EIGEN_STRONG_INLINE void resizeLike(const EigenBase<OtherDerived>& _other) { const OtherDerived& other = _other.derived(); + internal::check_rows_cols_for_overflow(other.rows(), other.cols()); const Index othersize = other.rows()*other.cols(); if(RowsAtCompileTime == 1) { @@ -417,6 +431,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type : m_storage(other.derived().rows() * other.derived().cols(), other.derived().rows(), other.derived().cols()) { _check_template_params(); + internal::check_rows_cols_for_overflow(other.derived().rows(), other.derived().cols()); Base::operator=(other.derived()); } @@ -581,6 +596,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type { eigen_assert(rows >= 0 && (RowsAtCompileTime == Dynamic || RowsAtCompileTime == rows) && cols >= 0 && (ColsAtCompileTime == Dynamic || ColsAtCompileTime == cols)); + internal::check_rows_cols_for_overflow(rows, cols); m_storage.resize(rows*cols,rows,cols); EIGEN_INITIALIZE_BY_ZERO_IF_THAT_OPTION_IS_ENABLED } @@ -638,6 +654,7 @@ struct internal::conservative_resize_like_impl if ( ( Derived::IsRowMajor && _this.cols() == cols) || // row-major and we change only the number of rows (!Derived::IsRowMajor && _this.rows() == rows) ) // column-major and we change only the number of columns { + internal::check_rows_cols_for_overflow(rows, cols); _this.derived().m_storage.conservativeResize(rows*cols,rows,cols); } else diff --git a/extern/Eigen3/Eigen/src/Core/ProductBase.h b/extern/Eigen3/Eigen/src/Core/ProductBase.h index 3bd3487d6a2..91975880fdc 100644 --- a/extern/Eigen3/Eigen/src/Core/ProductBase.h +++ b/extern/Eigen3/Eigen/src/Core/ProductBase.h @@ -152,7 +152,8 @@ class ProductBase : public MatrixBase<Derived> #else EIGEN_STATIC_ASSERT_SIZE_1x1(Derived) eigen_assert(this->rows() == 1 && this->cols() == 1); - return derived().coeff(row,col); + Matrix<Scalar,1,1> result = *this; + return result.coeff(row,col); #endif } @@ -160,7 +161,8 @@ class ProductBase : public MatrixBase<Derived> { EIGEN_STATIC_ASSERT_SIZE_1x1(Derived) eigen_assert(this->rows() == 1 && this->cols() == 1); - return derived().coeff(i); + Matrix<Scalar,1,1> result = *this; + return result.coeff(i); } const Scalar& coeffRef(Index row, Index col) const @@ -256,16 +258,16 @@ class ScaledProduct : Base(prod.lhs(),prod.rhs()), m_prod(prod), m_alpha(x) {} template<typename Dest> - inline void evalTo(Dest& dst) const { dst.setZero(); scaleAndAddTo(dst,m_alpha); } + inline void evalTo(Dest& dst) const { dst.setZero(); scaleAndAddTo(dst, Scalar(1)); } template<typename Dest> - inline void addTo(Dest& dst) const { scaleAndAddTo(dst,m_alpha); } + inline void addTo(Dest& dst) const { scaleAndAddTo(dst, Scalar(1)); } template<typename Dest> - inline void subTo(Dest& dst) const { scaleAndAddTo(dst,-m_alpha); } + inline void subTo(Dest& dst) const { scaleAndAddTo(dst, Scalar(-1)); } template<typename Dest> - inline void scaleAndAddTo(Dest& dst,Scalar alpha) const { m_prod.derived().scaleAndAddTo(dst,alpha); } + inline void scaleAndAddTo(Dest& dst,Scalar alpha) const { m_prod.derived().scaleAndAddTo(dst,alpha * m_alpha); } const Scalar& alpha() const { return m_alpha; } diff --git a/extern/Eigen3/Eigen/src/Core/Replicate.h b/extern/Eigen3/Eigen/src/Core/Replicate.h index d2f9712db0f..4c171f8d580 100644 --- a/extern/Eigen3/Eigen/src/Core/Replicate.h +++ b/extern/Eigen3/Eigen/src/Core/Replicate.h @@ -48,7 +48,10 @@ struct traits<Replicate<MatrixType,RowFactor,ColFactor> > typedef typename MatrixType::Scalar Scalar; typedef typename traits<MatrixType>::StorageKind StorageKind; typedef typename traits<MatrixType>::XprKind XprKind; - typedef typename nested<MatrixType>::type MatrixTypeNested; + enum { + Factor = (RowFactor==Dynamic || ColFactor==Dynamic) ? Dynamic : RowFactor*ColFactor + }; + typedef typename nested<MatrixType,Factor>::type MatrixTypeNested; typedef typename remove_reference<MatrixTypeNested>::type _MatrixTypeNested; enum { RowsAtCompileTime = RowFactor==Dynamic || int(MatrixType::RowsAtCompileTime)==Dynamic @@ -72,6 +75,8 @@ struct traits<Replicate<MatrixType,RowFactor,ColFactor> > template<typename MatrixType,int RowFactor,int ColFactor> class Replicate : public internal::dense_xpr_base< Replicate<MatrixType,RowFactor,ColFactor> >::type { + typedef typename internal::traits<Replicate>::MatrixTypeNested MatrixTypeNested; + typedef typename internal::traits<Replicate>::_MatrixTypeNested _MatrixTypeNested; public: typedef typename internal::dense_xpr_base<Replicate>::type Base; @@ -124,7 +129,7 @@ template<typename MatrixType,int RowFactor,int ColFactor> class Replicate protected: - const typename MatrixType::Nested m_matrix; + const MatrixTypeNested m_matrix; const internal::variable_if_dynamic<Index, RowFactor> m_rowFactor; const internal::variable_if_dynamic<Index, ColFactor> m_colFactor; }; diff --git a/extern/Eigen3/Eigen/src/Core/SolveTriangular.h b/extern/Eigen3/Eigen/src/Core/SolveTriangular.h index 71e129c7f12..a23014a343f 100644 --- a/extern/Eigen3/Eigen/src/Core/SolveTriangular.h +++ b/extern/Eigen3/Eigen/src/Core/SolveTriangular.h @@ -180,7 +180,7 @@ void TriangularView<MatrixType,Mode>::solveInPlace(const MatrixBase<OtherDerived eigen_assert(cols() == rows()); eigen_assert( (Side==OnTheLeft && cols() == other.rows()) || (Side==OnTheRight && cols() == other.cols()) ); eigen_assert(!(Mode & ZeroDiag)); - eigen_assert(Mode & (Upper|Lower)); + eigen_assert((Mode & (Upper|Lower)) != 0); enum { copy = internal::traits<OtherDerived>::Flags & RowMajorBit && OtherDerived::IsVectorAtCompileTime }; typedef typename internal::conditional<copy, diff --git a/extern/Eigen3/Eigen/src/Core/arch/NEON/Complex.h b/extern/Eigen3/Eigen/src/Core/arch/NEON/Complex.h index 8e55548c946..212887184c2 100644 --- a/extern/Eigen3/Eigen/src/Core/arch/NEON/Complex.h +++ b/extern/Eigen3/Eigen/src/Core/arch/NEON/Complex.h @@ -27,8 +27,8 @@ namespace internal { -static uint32x4_t p4ui_CONJ_XOR = { 0x00000000, 0x80000000, 0x00000000, 0x80000000 }; -static uint32x2_t p2ui_CONJ_XOR = { 0x00000000, 0x80000000 }; +static uint32x4_t p4ui_CONJ_XOR = EIGEN_INIT_NEON_PACKET4(0x00000000, 0x80000000, 0x00000000, 0x80000000); +static uint32x2_t p2ui_CONJ_XOR = EIGEN_INIT_NEON_PACKET2(0x00000000, 0x80000000); //---------- float ---------- struct Packet2cf diff --git a/extern/Eigen3/Eigen/src/Core/arch/NEON/PacketMath.h b/extern/Eigen3/Eigen/src/Core/arch/NEON/PacketMath.h index 478ef8038c0..6c7cd159097 100644 --- a/extern/Eigen3/Eigen/src/Core/arch/NEON/PacketMath.h +++ b/extern/Eigen3/Eigen/src/Core/arch/NEON/PacketMath.h @@ -52,6 +52,16 @@ typedef uint32x4_t Packet4ui; #define _EIGEN_DECLARE_CONST_Packet4i(NAME,X) \ const Packet4i p4i_##NAME = pset1<Packet4i>(X) +#if defined(__llvm__) && !defined(__clang__) + //Special treatment for Apple's llvm-gcc, its NEON packet types are unions + #define EIGEN_INIT_NEON_PACKET2(X, Y) {{X, Y}} + #define EIGEN_INIT_NEON_PACKET4(X, Y, Z, W) {{X, Y, Z, W}} +#else + //Default initializer for packets + #define EIGEN_INIT_NEON_PACKET2(X, Y) {X, Y} + #define EIGEN_INIT_NEON_PACKET4(X, Y, Z, W) {X, Y, Z, W} +#endif + #ifndef __pld #define __pld(x) asm volatile ( " pld [%[addr]]\n" :: [addr] "r" (x) : "cc" ); #endif @@ -84,7 +94,7 @@ template<> struct packet_traits<int> : default_packet_traits }; }; -#if EIGEN_GNUC_AT_MOST(4,4) +#if EIGEN_GNUC_AT_MOST(4,4) && !defined(__llvm__) // workaround gcc 4.2, 4.3 and 4.4 compilatin issue EIGEN_STRONG_INLINE float32x4_t vld1q_f32(const float* x) { return ::vld1q_f32((const float32_t*)x); } EIGEN_STRONG_INLINE float32x2_t vld1_f32 (const float* x) { return ::vld1_f32 ((const float32_t*)x); } @@ -100,12 +110,12 @@ template<> EIGEN_STRONG_INLINE Packet4i pset1<Packet4i>(const int& from) { template<> EIGEN_STRONG_INLINE Packet4f plset<float>(const float& a) { - Packet4f countdown = { 0, 1, 2, 3 }; + Packet4f countdown = EIGEN_INIT_NEON_PACKET4(0, 1, 2, 3); return vaddq_f32(pset1<Packet4f>(a), countdown); } template<> EIGEN_STRONG_INLINE Packet4i plset<int>(const int& a) { - Packet4i countdown = { 0, 1, 2, 3 }; + Packet4i countdown = EIGEN_INIT_NEON_PACKET4(0, 1, 2, 3); return vaddq_s32(pset1<Packet4i>(a), countdown); } @@ -395,25 +405,29 @@ template<> EIGEN_STRONG_INLINE int predux_max<Packet4i>(const Packet4i& a) return s[0]; } -template<int Offset> -struct palign_impl<Offset,Packet4f> -{ - EIGEN_STRONG_INLINE static void run(Packet4f& first, const Packet4f& second) - { - if (Offset!=0) - first = vextq_f32(first, second, Offset); - } -}; - -template<int Offset> -struct palign_impl<Offset,Packet4i> -{ - EIGEN_STRONG_INLINE static void run(Packet4i& first, const Packet4i& second) - { - if (Offset!=0) - first = vextq_s32(first, second, Offset); - } -}; +// this PALIGN_NEON business is to work around a bug in LLVM Clang 3.0 causing incorrect compilation errors, +// see bug 347 and this LLVM bug: http://llvm.org/bugs/show_bug.cgi?id=11074 +#define PALIGN_NEON(Offset,Type,Command) \ +template<>\ +struct palign_impl<Offset,Type>\ +{\ + EIGEN_STRONG_INLINE static void run(Type& first, const Type& second)\ + {\ + if (Offset!=0)\ + first = Command(first, second, Offset);\ + }\ +};\ + +PALIGN_NEON(0,Packet4f,vextq_f32) +PALIGN_NEON(1,Packet4f,vextq_f32) +PALIGN_NEON(2,Packet4f,vextq_f32) +PALIGN_NEON(3,Packet4f,vextq_f32) +PALIGN_NEON(0,Packet4i,vextq_s32) +PALIGN_NEON(1,Packet4i,vextq_s32) +PALIGN_NEON(2,Packet4i,vextq_s32) +PALIGN_NEON(3,Packet4i,vextq_s32) + +#undef PALIGN_NEON } // end namespace internal diff --git a/extern/Eigen3/Eigen/src/Core/products/GeneralBlockPanelKernel.h b/extern/Eigen3/Eigen/src/Core/products/GeneralBlockPanelKernel.h index 6f3f2717007..cd1c37c780e 100644 --- a/extern/Eigen3/Eigen/src/Core/products/GeneralBlockPanelKernel.h +++ b/extern/Eigen3/Eigen/src/Core/products/GeneralBlockPanelKernel.h @@ -30,19 +30,16 @@ namespace internal { template<typename _LhsScalar, typename _RhsScalar, bool _ConjLhs=false, bool _ConjRhs=false> class gebp_traits; +inline std::ptrdiff_t manage_caching_sizes_second_if_negative(std::ptrdiff_t a, std::ptrdiff_t b) +{ + return a<=0 ? b : a; +} + /** \internal */ inline void manage_caching_sizes(Action action, std::ptrdiff_t* l1=0, std::ptrdiff_t* l2=0) { - static std::ptrdiff_t m_l1CacheSize = 0; - static std::ptrdiff_t m_l2CacheSize = 0; - if(m_l1CacheSize==0) - { - m_l1CacheSize = queryL1CacheSize(); - m_l2CacheSize = queryTopLevelCacheSize(); - - if(m_l1CacheSize<=0) m_l1CacheSize = 8 * 1024; - if(m_l2CacheSize<=0) m_l2CacheSize = 1 * 1024 * 1024; - } + static std::ptrdiff_t m_l1CacheSize = manage_caching_sizes_second_if_negative(queryL1CacheSize(),8 * 1024); + static std::ptrdiff_t m_l2CacheSize = manage_caching_sizes_second_if_negative(queryTopLevelCacheSize(),1*1024*1024); if(action==SetAction) { @@ -118,14 +115,14 @@ inline void computeProductBlockingSizes(std::ptrdiff_t& k, std::ptrdiff_t& m, st // FIXME (a bit overkill maybe ?) template<typename CJ, typename A, typename B, typename C, typename T> struct gebp_madd_selector { - EIGEN_STRONG_INLINE EIGEN_ALWAYS_INLINE_ATTRIB static void run(const CJ& cj, A& a, B& b, C& c, T& /*t*/) + EIGEN_ALWAYS_INLINE static void run(const CJ& cj, A& a, B& b, C& c, T& /*t*/) { c = cj.pmadd(a,b,c); } }; template<typename CJ, typename T> struct gebp_madd_selector<CJ,T,T,T,T> { - EIGEN_STRONG_INLINE EIGEN_ALWAYS_INLINE_ATTRIB static void run(const CJ& cj, T& a, T& b, T& c, T& t) + EIGEN_ALWAYS_INLINE static void run(const CJ& cj, T& a, T& b, T& c, T& t) { t = b; t = cj.pmul(a,t); c = padd(c,t); } diff --git a/extern/Eigen3/Eigen/src/Core/util/Macros.h b/extern/Eigen3/Eigen/src/Core/util/Macros.h index 6c3f1e421f0..b7c2b79af92 100644 --- a/extern/Eigen3/Eigen/src/Core/util/Macros.h +++ b/extern/Eigen3/Eigen/src/Core/util/Macros.h @@ -1,3 +1,4 @@ + // This file is part of Eigen, a lightweight C++ template library // for linear algebra. // @@ -28,7 +29,7 @@ #define EIGEN_WORLD_VERSION 3 #define EIGEN_MAJOR_VERSION 0 -#define EIGEN_MINOR_VERSION 2 +#define EIGEN_MINOR_VERSION 5 #define EIGEN_VERSION_AT_LEAST(x,y,z) (EIGEN_WORLD_VERSION>x || (EIGEN_WORLD_VERSION>=x && \ (EIGEN_MAJOR_VERSION>y || (EIGEN_MAJOR_VERSION>=y && \ @@ -45,7 +46,7 @@ #define EIGEN_GNUC_AT_MOST(x,y) 0 #endif -#if EIGEN_GNUC_AT_MOST(4,3) +#if EIGEN_GNUC_AT_MOST(4,3) && !defined(__clang__) // see bug 89 #define EIGEN_SAFE_TO_USE_STANDARD_ASSERT_MACRO 0 #else @@ -130,31 +131,34 @@ #define EIGEN_MAKESTRING2(a) #a #define EIGEN_MAKESTRING(a) EIGEN_MAKESTRING2(a) -// EIGEN_ALWAYS_INLINE_ATTRIB should be use in the declaration of function -// which should be inlined even in debug mode. -// FIXME with the always_inline attribute, -// gcc 3.4.x reports the following compilation error: -// Eval.h:91: sorry, unimplemented: inlining failed in call to 'const Eigen::Eval<Derived> Eigen::MatrixBase<Scalar, Derived>::eval() const' -// : function body not available -#if EIGEN_GNUC_AT_LEAST(4,0) -#define EIGEN_ALWAYS_INLINE_ATTRIB __attribute__((always_inline)) -#else -#define EIGEN_ALWAYS_INLINE_ATTRIB -#endif - #if EIGEN_GNUC_AT_LEAST(4,1) && !defined(__clang__) && !defined(__INTEL_COMPILER) #define EIGEN_FLATTEN_ATTRIB __attribute__((flatten)) #else #define EIGEN_FLATTEN_ATTRIB #endif -// EIGEN_FORCE_INLINE means "inline as much as possible" +// EIGEN_STRONG_INLINE is a stronger version of the inline, using __forceinline on MSVC, +// but it still doesn't use GCC's always_inline. This is useful in (common) situations where MSVC needs forceinline +// but GCC is still doing fine with just inline. #if (defined _MSC_VER) || (defined __INTEL_COMPILER) #define EIGEN_STRONG_INLINE __forceinline #else #define EIGEN_STRONG_INLINE inline #endif +// EIGEN_ALWAYS_INLINE is the stronget, it has the effect of making the function inline and adding every possible +// attribute to maximize inlining. This should only be used when really necessary: in particular, +// it uses __attribute__((always_inline)) on GCC, which most of the time is useless and can severely harm compile times. +// FIXME with the always_inline attribute, +// gcc 3.4.x reports the following compilation error: +// Eval.h:91: sorry, unimplemented: inlining failed in call to 'const Eigen::Eval<Derived> Eigen::MatrixBase<Scalar, Derived>::eval() const' +// : function body not available +#if EIGEN_GNUC_AT_LEAST(4,0) +#define EIGEN_ALWAYS_INLINE __attribute__((always_inline)) inline +#else +#define EIGEN_ALWAYS_INLINE EIGEN_STRONG_INLINE +#endif + #if (defined __GNUC__) #define EIGEN_DONT_INLINE __attribute__((noinline)) #elif (defined _MSC_VER) @@ -249,7 +253,7 @@ #define EIGEN_UNUSED_VARIABLE(var) (void)var; #if (defined __GNUC__) -#define EIGEN_ASM_COMMENT(X) asm("#"X) +#define EIGEN_ASM_COMMENT(X) asm("#" X) #else #define EIGEN_ASM_COMMENT(X) #endif diff --git a/extern/Eigen3/Eigen/src/Core/util/Memory.h b/extern/Eigen3/Eigen/src/Core/util/Memory.h index a580b95ad0d..023716dc9e0 100644 --- a/extern/Eigen3/Eigen/src/Core/util/Memory.h +++ b/extern/Eigen3/Eigen/src/Core/util/Memory.h @@ -82,6 +82,16 @@ namespace internal { +inline void throw_std_bad_alloc() +{ + #ifdef EIGEN_EXCEPTIONS + throw std::bad_alloc(); + #else + std::size_t huge = -1; + new int[huge]; + #endif +} + /***************************************************************************** *** Implementation of handmade aligned functions *** *****************************************************************************/ @@ -192,7 +202,7 @@ inline void check_that_malloc_is_allowed() #endif /** \internal Allocates \a size bytes. The returned pointer is guaranteed to have 16 bytes alignment. - * On allocation error, the returned pointer is null, and if exceptions are enabled then a std::bad_alloc is thrown. + * On allocation error, the returned pointer is null, and std::bad_alloc is thrown. */ inline void* aligned_malloc(size_t size) { @@ -213,10 +223,9 @@ inline void* aligned_malloc(size_t size) result = handmade_aligned_malloc(size); #endif - #ifdef EIGEN_EXCEPTIONS - if(result == 0) - throw std::bad_alloc(); - #endif + if(!result && size) + throw_std_bad_alloc(); + return result; } @@ -241,7 +250,7 @@ inline void aligned_free(void *ptr) /** * \internal * \brief Reallocates an aligned block of memory. -* \throws std::bad_alloc if EIGEN_EXCEPTIONS are defined. +* \throws std::bad_alloc on allocation failure **/ inline void* aligned_realloc(void *ptr, size_t new_size, size_t old_size) { @@ -269,10 +278,9 @@ inline void* aligned_realloc(void *ptr, size_t new_size, size_t old_size) result = handmade_aligned_realloc(ptr,new_size,old_size); #endif -#ifdef EIGEN_EXCEPTIONS - if (result==0 && new_size!=0) - throw std::bad_alloc(); -#endif + if (!result && new_size) + throw_std_bad_alloc(); + return result; } @@ -281,7 +289,7 @@ inline void* aligned_realloc(void *ptr, size_t new_size, size_t old_size) *****************************************************************************/ /** \internal Allocates \a size bytes. If Align is true, then the returned ptr is 16-byte-aligned. - * On allocation error, the returned pointer is null, and if exceptions are enabled then a std::bad_alloc is thrown. + * On allocation error, the returned pointer is null, and a std::bad_alloc is thrown. */ template<bool Align> inline void* conditional_aligned_malloc(size_t size) { @@ -293,9 +301,8 @@ template<> inline void* conditional_aligned_malloc<false>(size_t size) check_that_malloc_is_allowed(); void *result = std::malloc(size); - #ifdef EIGEN_EXCEPTIONS - if(!result) throw std::bad_alloc(); - #endif + if(!result && size) + throw_std_bad_alloc(); return result; } @@ -347,18 +354,27 @@ template<typename T> inline void destruct_elements_of_array(T *ptr, size_t size) *** Implementation of aligned new/delete-like functions *** *****************************************************************************/ +template<typename T> +EIGEN_ALWAYS_INLINE void check_size_for_overflow(size_t size) +{ + if(size > size_t(-1) / sizeof(T)) + throw_std_bad_alloc(); +} + /** \internal Allocates \a size objects of type T. The returned pointer is guaranteed to have 16 bytes alignment. - * On allocation error, the returned pointer is undefined, but if exceptions are enabled then a std::bad_alloc is thrown. + * On allocation error, the returned pointer is undefined, but a std::bad_alloc is thrown. * The default constructor of T is called. */ template<typename T> inline T* aligned_new(size_t size) { + check_size_for_overflow<T>(size); T *result = reinterpret_cast<T*>(aligned_malloc(sizeof(T)*size)); return construct_elements_of_array(result, size); } template<typename T, bool Align> inline T* conditional_aligned_new(size_t size) { + check_size_for_overflow<T>(size); T *result = reinterpret_cast<T*>(conditional_aligned_malloc<Align>(sizeof(T)*size)); return construct_elements_of_array(result, size); } @@ -383,6 +399,8 @@ template<typename T, bool Align> inline void conditional_aligned_delete(T *ptr, template<typename T, bool Align> inline T* conditional_aligned_realloc_new(T* pts, size_t new_size, size_t old_size) { + check_size_for_overflow<T>(new_size); + check_size_for_overflow<T>(old_size); if(new_size < old_size) destruct_elements_of_array(pts+new_size, old_size-new_size); T *result = reinterpret_cast<T*>(conditional_aligned_realloc<Align>(reinterpret_cast<void*>(pts), sizeof(T)*new_size, sizeof(T)*old_size)); @@ -394,6 +412,7 @@ template<typename T, bool Align> inline T* conditional_aligned_realloc_new(T* pt template<typename T, bool Align> inline T* conditional_aligned_new_auto(size_t size) { + check_size_for_overflow<T>(size); T *result = reinterpret_cast<T*>(conditional_aligned_malloc<Align>(sizeof(T)*size)); if(NumTraits<T>::RequireInitialization) construct_elements_of_array(result, size); @@ -402,6 +421,8 @@ template<typename T, bool Align> inline T* conditional_aligned_new_auto(size_t s template<typename T, bool Align> inline T* conditional_aligned_realloc_new_auto(T* pts, size_t new_size, size_t old_size) { + check_size_for_overflow<T>(new_size); + check_size_for_overflow<T>(old_size); if(NumTraits<T>::RequireInitialization && (new_size < old_size)) destruct_elements_of_array(pts+new_size, old_size-new_size); T *result = reinterpret_cast<T*>(conditional_aligned_realloc<Align>(reinterpret_cast<void*>(pts), sizeof(T)*new_size, sizeof(T)*old_size)); @@ -536,6 +557,7 @@ template<typename T> class aligned_stack_memory_handler #endif #define ei_declare_aligned_stack_constructed_variable(TYPE,NAME,SIZE,BUFFER) \ + Eigen::internal::check_size_for_overflow<TYPE>(SIZE); \ TYPE* NAME = (BUFFER)!=0 ? (BUFFER) \ : reinterpret_cast<TYPE*>( \ (sizeof(TYPE)*SIZE<=EIGEN_STACK_ALLOCATION_LIMIT) ? EIGEN_ALIGNED_ALLOCA(sizeof(TYPE)*SIZE) \ @@ -545,6 +567,7 @@ template<typename T> class aligned_stack_memory_handler #else #define ei_declare_aligned_stack_constructed_variable(TYPE,NAME,SIZE,BUFFER) \ + Eigen::internal::check_size_for_overflow<TYPE>(SIZE); \ TYPE* NAME = (BUFFER)!=0 ? BUFFER : reinterpret_cast<TYPE*>(Eigen::internal::aligned_malloc(sizeof(TYPE)*SIZE)); \ Eigen::internal::aligned_stack_memory_handler<TYPE> EIGEN_CAT(NAME,_stack_memory_destructor)((BUFFER)==0 ? NAME : 0,SIZE,true) @@ -669,6 +692,7 @@ public: pointer allocate( size_type num, const void* hint = 0 ) { EIGEN_UNUSED_VARIABLE(hint); + internal::check_size_for_overflow<T>(num); return static_cast<pointer>( internal::aligned_malloc( num * sizeof(T) ) ); } diff --git a/extern/Eigen3/Eigen/src/Core/util/XprHelper.h b/extern/Eigen3/Eigen/src/Core/util/XprHelper.h index 9047c5f8350..c2078f13786 100644 --- a/extern/Eigen3/Eigen/src/Core/util/XprHelper.h +++ b/extern/Eigen3/Eigen/src/Core/util/XprHelper.h @@ -125,10 +125,9 @@ class compute_matrix_flags aligned_bit = ( ((Options&DontAlign)==0) - && packet_traits<Scalar>::Vectorizable && ( #if EIGEN_ALIGN_STATICALLY - ((!is_dynamic_size_storage) && (((MaxCols*MaxRows) % packet_traits<Scalar>::size) == 0)) + ((!is_dynamic_size_storage) && (((MaxCols*MaxRows*int(sizeof(Scalar))) % 16) == 0)) #else 0 #endif |