[PATCH 1/2] Misc. typos
From 68d431b4c14ad60a778ee93c1f59ecc4b931950e Mon Sep 17 00:00:00 2001 Found via `codespell -q 3 -I ../eigen-word-whitelist.txt` where the whitelists consists of: ``` als ans cas dum lastr lowd nd overfl pres preverse substraction te uint whch ``` --- CMakeLists.txt | 26 +++++++++---------- Eigen/src/Core/GenericPacketMath.h | 2 +- Eigen/src/SparseLU/SparseLU.h | 2 +- bench/bench_norm.cpp | 2 +- doc/HiPerformance.dox | 2 +- doc/QuickStartGuide.dox | 2 +- .../Eigen/CXX11/src/Tensor/TensorChipping.h | 6 ++--- .../Eigen/CXX11/src/Tensor/TensorDeviceGpu.h | 2 +- .../src/Tensor/TensorForwardDeclarations.h | 4 +-- .../src/Tensor/TensorGpuHipCudaDefines.h | 2 +- .../Eigen/CXX11/src/Tensor/TensorReduction.h | 2 +- .../CXX11/src/Tensor/TensorReductionGpu.h | 2 +- .../test/cxx11_tensor_concatenation.cpp | 2 +- unsupported/test/cxx11_tensor_executor.cpp | 2 +- 14 files changed, 29 insertions(+), 29 deletions(-)
This commit is contained in:
luz.paz"
parent
77b447c24e
commit
f67b19a884
@ -159,7 +159,7 @@ if(NOT MSVC)
|
||||
ei_add_cxx_compiler_flag("-Wall")
|
||||
ei_add_cxx_compiler_flag("-Wextra")
|
||||
#ei_add_cxx_compiler_flag("-Weverything") # clang
|
||||
|
||||
|
||||
ei_add_cxx_compiler_flag("-Wundef")
|
||||
ei_add_cxx_compiler_flag("-Wcast-align")
|
||||
ei_add_cxx_compiler_flag("-Wchar-subscripts")
|
||||
@ -174,25 +174,25 @@ if(NOT MSVC)
|
||||
ei_add_cxx_compiler_flag("-Wc++11-extensions")
|
||||
ei_add_cxx_compiler_flag("-Wdouble-promotion")
|
||||
# ei_add_cxx_compiler_flag("-Wconversion")
|
||||
|
||||
|
||||
ei_add_cxx_compiler_flag("-Wshadow")
|
||||
|
||||
|
||||
ei_add_cxx_compiler_flag("-Wno-psabi")
|
||||
ei_add_cxx_compiler_flag("-Wno-variadic-macros")
|
||||
ei_add_cxx_compiler_flag("-Wno-long-long")
|
||||
|
||||
|
||||
ei_add_cxx_compiler_flag("-fno-check-new")
|
||||
ei_add_cxx_compiler_flag("-fno-common")
|
||||
ei_add_cxx_compiler_flag("-fstrict-aliasing")
|
||||
ei_add_cxx_compiler_flag("-wd981") # disable ICC's "operands are evaluated in unspecified order" remark
|
||||
ei_add_cxx_compiler_flag("-wd2304") # disable ICC's "warning #2304: non-explicit constructor with single argument may cause implicit type conversion" produced by -Wnon-virtual-dtor
|
||||
|
||||
|
||||
|
||||
|
||||
# The -ansi flag must be added last, otherwise it is also used as a linker flag by check_cxx_compiler_flag making it fails
|
||||
# Moreover we should not set both -strict-ansi and -ansi
|
||||
check_cxx_compiler_flag("-strict-ansi" COMPILER_SUPPORT_STRICTANSI)
|
||||
ei_add_cxx_compiler_flag("-Qunused-arguments") # disable clang warning: argument unused during compilation: '-ansi'
|
||||
|
||||
|
||||
if(COMPILER_SUPPORT_STRICTANSI)
|
||||
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -strict-ansi")
|
||||
else()
|
||||
@ -203,7 +203,7 @@ if(NOT MSVC)
|
||||
ei_add_cxx_compiler_flag("-pie")
|
||||
ei_add_cxx_compiler_flag("-fPIE")
|
||||
endif()
|
||||
|
||||
|
||||
set(CMAKE_REQUIRED_FLAGS "")
|
||||
|
||||
option(EIGEN_TEST_SSE2 "Enable/Disable SSE2 in tests/examples" OFF)
|
||||
@ -327,7 +327,7 @@ else(NOT MSVC)
|
||||
# because we are oftentimes returning objects that have a destructor or may
|
||||
# throw exceptions - in particular in the unit tests we are throwing extra many
|
||||
# exceptions to cover indexing errors.
|
||||
# C4505 - unreferenced local function has been removed (impossible to deactive selectively)
|
||||
# C4505 - unreferenced local function has been removed (impossible to deactivate selectively)
|
||||
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /EHsc /wd4127 /wd4505 /wd4714")
|
||||
|
||||
# replace all /Wx by /W4
|
||||
@ -596,11 +596,11 @@ if (NOT CMAKE_VERSION VERSION_LESS 3.0)
|
||||
|
||||
else (NOT CMAKE_VERSION VERSION_LESS 3.0)
|
||||
# Fallback to legacy Eigen3Config.cmake without the imported target
|
||||
|
||||
|
||||
# If CMakePackageConfigHelpers module is available (CMake >= 2.8.8)
|
||||
# create a relocatable Config file, otherwise leave the hardcoded paths
|
||||
# create a relocatable Config file, otherwise leave the hardcoded paths
|
||||
include(CMakePackageConfigHelpers OPTIONAL RESULT_VARIABLE CPCH_PATH)
|
||||
|
||||
|
||||
if(CPCH_PATH)
|
||||
configure_package_config_file (
|
||||
${CMAKE_CURRENT_SOURCE_DIR}/cmake/Eigen3ConfigLegacy.cmake.in
|
||||
@ -609,7 +609,7 @@ else (NOT CMAKE_VERSION VERSION_LESS 3.0)
|
||||
INSTALL_DESTINATION ${CMAKEPACKAGE_INSTALL_DIR}
|
||||
NO_CHECK_REQUIRED_COMPONENTS_MACRO # Eigen does not provide components
|
||||
)
|
||||
else()
|
||||
else()
|
||||
# The PACKAGE_* variables are defined by the configure_package_config_file
|
||||
# but without it we define them manually to the hardcoded paths
|
||||
set(PACKAGE_INIT "")
|
||||
|
||||
@ -450,7 +450,7 @@ Packet pceil(const Packet& a) { using numext::ceil; return ceil(a); }
|
||||
* The following functions might not have to be overwritten for vectorized types
|
||||
***************************************************************************/
|
||||
|
||||
/** \internal copy a packet with constant coeficient \a a (e.g., [a,a,a,a]) to \a *to. \a to must be 16 bytes aligned */
|
||||
/** \internal copy a packet with constant coefficient \a a (e.g., [a,a,a,a]) to \a *to. \a to must be 16 bytes aligned */
|
||||
// NOTE: this function must really be templated on the packet type (think about different packet types for the same scalar type)
|
||||
template<typename Packet>
|
||||
inline void pstore1(typename unpacket_traits<Packet>::type* to, const typename unpacket_traits<Packet>::type& a)
|
||||
|
||||
@ -104,7 +104,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
|
||||
typedef typename internal::traits<Derived>::StorageKind StorageKind;
|
||||
typedef typename internal::traits<Derived>::Scalar Scalar;
|
||||
|
||||
|
||||
typedef typename internal::packet_traits<Scalar>::type PacketScalar;
|
||||
typedef typename NumTraits<Scalar>::Real RealScalar;
|
||||
typedef Derived DenseType;
|
||||
@ -358,7 +358,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* remain row-vectors and vectors remain vectors.
|
||||
*/
|
||||
template<typename OtherDerived>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE void resizeLike(const EigenBase<OtherDerived>& _other)
|
||||
{
|
||||
const OtherDerived& other = _other.derived();
|
||||
@ -383,7 +383,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* of rows and/or of columns, you can use conservativeResize(NoChange_t, Index) or
|
||||
* conservativeResize(Index, NoChange_t).
|
||||
*
|
||||
* Matrices are resized relative to the top-left element. In case values need to be
|
||||
* Matrices are resized relative to the top-left element. In case values need to be
|
||||
* appended to the matrix they will be uninitialized.
|
||||
*/
|
||||
EIGEN_DEVICE_FUNC
|
||||
@ -440,7 +440,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* of rows and/or of columns, you can use conservativeResize(NoChange_t, Index) or
|
||||
* conservativeResize(Index, NoChange_t).
|
||||
*
|
||||
* Matrices are resized relative to the top-left element. In case values need to be
|
||||
* Matrices are resized relative to the top-left element. In case values need to be
|
||||
* appended to the matrix they will copied from \c other.
|
||||
*/
|
||||
template<typename OtherDerived>
|
||||
@ -564,7 +564,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* \copydetails DenseBase::operator=(const EigenBase<OtherDerived> &other)
|
||||
*/
|
||||
template<typename OtherDerived>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE Derived& operator=(const EigenBase<OtherDerived> &other)
|
||||
{
|
||||
_resize_to_match(other);
|
||||
@ -678,7 +678,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* remain row-vectors and vectors remain vectors.
|
||||
*/
|
||||
template<typename OtherDerived>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE void _resize_to_match(const EigenBase<OtherDerived>& other)
|
||||
{
|
||||
#ifdef EIGEN_NO_AUTOMATIC_RESIZING
|
||||
@ -705,10 +705,10 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
*
|
||||
* \internal
|
||||
*/
|
||||
// aliasing is dealt once in internall::call_assignment
|
||||
// aliasing is dealt once in internal::call_assignment
|
||||
// so at this stage we have to assume aliasing... and resising has to be done later.
|
||||
template<typename OtherDerived>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE Derived& _set(const DenseBase<OtherDerived>& other)
|
||||
{
|
||||
internal::call_assignment(this->derived(), other.derived());
|
||||
@ -721,7 +721,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
* \sa operator=(const MatrixBase<OtherDerived>&), _set()
|
||||
*/
|
||||
template<typename OtherDerived>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE Derived& _set_noalias(const DenseBase<OtherDerived>& other)
|
||||
{
|
||||
// I don't think we need this resize call since the lazyAssign will anyways resize
|
||||
@ -742,18 +742,18 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
FLOATING_POINT_ARGUMENT_PASSED__INTEGER_WAS_EXPECTED)
|
||||
resize(rows,cols);
|
||||
}
|
||||
|
||||
|
||||
template<typename T0, typename T1>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE void _init2(const T0& val0, const T1& val1, typename internal::enable_if<Base::SizeAtCompileTime==2,T0>::type* = 0)
|
||||
{
|
||||
EIGEN_STATIC_ASSERT_VECTOR_SPECIFIC_SIZE(PlainObjectBase, 2)
|
||||
m_storage.data()[0] = Scalar(val0);
|
||||
m_storage.data()[1] = Scalar(val1);
|
||||
}
|
||||
|
||||
|
||||
template<typename T0, typename T1>
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_DEVICE_FUNC
|
||||
EIGEN_STRONG_INLINE void _init2(const Index& val0, const Index& val1,
|
||||
typename internal::enable_if< (!internal::is_same<Index,Scalar>::value)
|
||||
&& (internal::is_same<T0,Index>::value)
|
||||
@ -779,7 +779,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
FLOATING_POINT_ARGUMENT_PASSED__INTEGER_WAS_EXPECTED)
|
||||
resize(size);
|
||||
}
|
||||
|
||||
|
||||
// We have a 1x1 matrix/array => the argument is interpreted as the value of the unique coefficient (case where scalar type can be implicitly converted)
|
||||
template<typename T>
|
||||
EIGEN_DEVICE_FUNC
|
||||
@ -788,7 +788,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
EIGEN_STATIC_ASSERT_VECTOR_SPECIFIC_SIZE(PlainObjectBase, 1)
|
||||
m_storage.data()[0] = val0;
|
||||
}
|
||||
|
||||
|
||||
// We have a 1x1 matrix/array => the argument is interpreted as the value of the unique coefficient (case where scalar type match the index type)
|
||||
template<typename T>
|
||||
EIGEN_DEVICE_FUNC
|
||||
@ -844,7 +844,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
{
|
||||
this->derived() = r;
|
||||
}
|
||||
|
||||
|
||||
// For fixed-size Array<Scalar,...>
|
||||
template<typename T>
|
||||
EIGEN_DEVICE_FUNC
|
||||
@ -856,7 +856,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
{
|
||||
Base::setConstant(val0);
|
||||
}
|
||||
|
||||
|
||||
// For fixed-size Array<Index,...>
|
||||
template<typename T>
|
||||
EIGEN_DEVICE_FUNC
|
||||
@ -870,12 +870,12 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
{
|
||||
Base::setConstant(val0);
|
||||
}
|
||||
|
||||
|
||||
template<typename MatrixTypeA, typename MatrixTypeB, bool SwapPointers>
|
||||
friend struct internal::matrix_swap_impl;
|
||||
|
||||
public:
|
||||
|
||||
|
||||
#ifndef EIGEN_PARSED_BY_DOXYGEN
|
||||
/** \internal
|
||||
* \brief Override DenseBase::swap() since for dynamic-sized matrices
|
||||
@ -888,7 +888,7 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
enum { SwapPointers = internal::is_same<Derived, OtherDerived>::value && Base::SizeAtCompileTime==Dynamic };
|
||||
internal::matrix_swap_impl<Derived, OtherDerived, bool(SwapPointers)>::run(this->derived(), other.derived());
|
||||
}
|
||||
|
||||
|
||||
/** \internal
|
||||
* \brief const version forwarded to DenseBase::swap
|
||||
*/
|
||||
@ -896,8 +896,8 @@ class PlainObjectBase : public internal::dense_xpr_base<Derived>::type
|
||||
EIGEN_DEVICE_FUNC
|
||||
void swap(DenseBase<OtherDerived> const & other)
|
||||
{ Base::swap(other.derived()); }
|
||||
|
||||
EIGEN_DEVICE_FUNC
|
||||
|
||||
EIGEN_DEVICE_FUNC
|
||||
static EIGEN_STRONG_INLINE void _check_template_params()
|
||||
{
|
||||
EIGEN_STATIC_ASSERT((EIGEN_IMPLIES(MaxRowsAtCompileTime==1 && MaxColsAtCompileTime!=1, (Options&RowMajor)==RowMajor)
|
||||
@ -994,7 +994,7 @@ struct conservative_resize_like_impl<Derived,OtherDerived,true>
|
||||
typedef conservative_resize_like_impl<Derived,OtherDerived,false> Base;
|
||||
using Base::run;
|
||||
using Base::IsRelocatable;
|
||||
|
||||
|
||||
static void run(DenseBase<Derived>& _this, Index size)
|
||||
{
|
||||
const Index new_rows = Derived::RowsAtCompileTime==1 ? 1 : size;
|
||||
|
||||
@ -11,7 +11,7 @@
|
||||
#ifndef EIGEN_JACOBI_H
|
||||
#define EIGEN_JACOBI_H
|
||||
|
||||
namespace Eigen {
|
||||
namespace Eigen {
|
||||
|
||||
/** \ingroup Jacobi_Module
|
||||
* \jacobi_module
|
||||
@ -173,7 +173,7 @@ void JacobiRotation<Scalar>::makeGivens(const Scalar& p, const Scalar& q, Scalar
|
||||
using std::sqrt;
|
||||
using std::abs;
|
||||
using numext::conj;
|
||||
|
||||
|
||||
if(q==Scalar(0))
|
||||
{
|
||||
m_c = numext::real(p)<0 ? Scalar(-1) : Scalar(1);
|
||||
@ -273,7 +273,7 @@ void JacobiRotation<Scalar>::makeGivens(const Scalar& p, const Scalar& q, Scalar
|
||||
|
||||
namespace internal {
|
||||
/** \jacobi_module
|
||||
* Applies the clock wise 2D rotation \a j to the set of 2D vectors of cordinates \a x and \a y:
|
||||
* Applies the clock wise 2D rotation \a j to the set of 2D vectors of coordinates \a x and \a y:
|
||||
* \f$ \left ( \begin{array}{cc} x \\ y \end{array} \right ) = J \left ( \begin{array}{cc} x \\ y \end{array} \right ) \f$
|
||||
*
|
||||
* \sa MatrixBase::applyOnTheLeft(), MatrixBase::applyOnTheRight()
|
||||
@ -463,7 +463,7 @@ void /*EIGEN_DONT_INLINE*/ apply_rotation_in_the_plane(DenseBase<VectorX>& xpr_x
|
||||
|
||||
Scalar* EIGEN_RESTRICT x = &xpr_x.derived().coeffRef(0);
|
||||
Scalar* EIGEN_RESTRICT y = &xpr_y.derived().coeffRef(0);
|
||||
|
||||
|
||||
OtherScalar c = j.c();
|
||||
OtherScalar s = j.s();
|
||||
if (c==OtherScalar(1) && s==OtherScalar(0))
|
||||
|
||||
@ -26,7 +26,7 @@ template <typename MatrixLType, typename MatrixUType> struct SparseLUMatrixURetu
|
||||
* This class implements the supernodal LU factorization for general matrices.
|
||||
* It uses the main techniques from the sequential SuperLU package
|
||||
* (http://crd-legacy.lbl.gov/~xiaoye/SuperLU/). It handles transparently real
|
||||
* and complex arithmetics with single and double precision, depending on the
|
||||
* and complex arithmetic with single and double precision, depending on the
|
||||
* scalar type of your input matrix.
|
||||
* The code has been optimized to provide BLAS-3 operations during supernode-panel updates.
|
||||
* It benefits directly from the built-in high-performant Eigen BLAS routines.
|
||||
|
||||
@ -134,7 +134,7 @@ EIGEN_DONT_INLINE typename T::Scalar pblueNorm(const T& v)
|
||||
iexp = - ((iemax+it)/2);
|
||||
s2m = std::pow(ibeta,iexp); // scaling factor for upper range
|
||||
|
||||
overfl = rbig*s2m; // overfow boundary for abig
|
||||
overfl = rbig*s2m; // overflow boundary for abig
|
||||
eps = std::pow(ibeta, 1-it);
|
||||
relerr = std::sqrt(eps); // tolerance for neglecting asml
|
||||
abig = 1.0/eps - 1.0;
|
||||
|
||||
@ -105,7 +105,7 @@ m1.noalias() += m2 * m3; \endcode</td>
|
||||
<td>First of all, here the .noalias() in the first expression is useless because
|
||||
m2*m3 will be evaluated anyway. However, note how this expression can be rewritten
|
||||
so that no temporary is required. (tip: for very small fixed size matrix
|
||||
it is slighlty better to rewrite it like this: m1.noalias() = m2 * m3; m1 += m4;</td>
|
||||
it is slightly better to rewrite it like this: m1.noalias() = m2 * m3; m1 += m4;</td>
|
||||
</tr>
|
||||
<tr class="alt">
|
||||
<td>\code
|
||||
|
||||
@ -66,7 +66,7 @@ The output is as follows:
|
||||
|
||||
\section GettingStartedExplanation2 Explanation of the second example
|
||||
|
||||
The second example starts by declaring a 3-by-3 matrix \c m which is initialized using the \link DenseBase::Random(Index,Index) Random() \endlink method with random values between -1 and 1. The next line applies a linear mapping such that the values are between 10 and 110. The function call \link DenseBase::Constant(Index,Index,const Scalar&) MatrixXd::Constant\endlink(3,3,1.2) returns a 3-by-3 matrix expression having all coefficients equal to 1.2. The rest is standard arithmetics.
|
||||
The second example starts by declaring a 3-by-3 matrix \c m which is initialized using the \link DenseBase::Random(Index,Index) Random() \endlink method with random values between -1 and 1. The next line applies a linear mapping such that the values are between 10 and 110. The function call \link DenseBase::Constant(Index,Index,const Scalar&) MatrixXd::Constant\endlink(3,3,1.2) returns a 3-by-3 matrix expression having all coefficients equal to 1.2. The rest is standard arithmetic.
|
||||
|
||||
The next line of the \c main function introduces a new type: \c VectorXd. This represents a (column) vector of arbitrary size. Here, the vector \c v is created to contain \c 3 coefficients which are left uninitialized. The one but last line uses the so-called comma-initializer, explained in \ref TutorialAdvancedInitialization, to set all coefficients of the vector \c v to be as follows:
|
||||
|
||||
|
||||
@ -244,7 +244,7 @@ struct TensorEvaluator<const TensorChippingOp<DimId, ArgType>, Device>
|
||||
return rslt;
|
||||
} else if ((static_cast<int>(Layout) == static_cast<int>(ColMajor) && m_dim.actualDim() == NumInputDims - 1) ||
|
||||
(static_cast<int>(Layout) == static_cast<int>(RowMajor) && m_dim.actualDim() == 0)) {
|
||||
// m_stride is aways greater than index, so let's avoid the integer division.
|
||||
// m_stride is always greater than index, so let's avoid the integer division.
|
||||
eigen_assert(m_stride > index);
|
||||
return m_impl.template packet<LoadMode>(index + m_inputOffset);
|
||||
} else {
|
||||
@ -377,7 +377,7 @@ struct TensorEvaluator<const TensorChippingOp<DimId, ArgType>, Device>
|
||||
inputIndex = index * m_inputStride + m_inputOffset;
|
||||
} else if ((static_cast<int>(Layout) == static_cast<int>(ColMajor) && m_dim.actualDim() == NumInputDims - 1) ||
|
||||
(static_cast<int>(Layout) == static_cast<int>(RowMajor) && m_dim.actualDim() == 0)) {
|
||||
// m_stride is aways greater than index, so let's avoid the integer
|
||||
// m_stride is always greater than index, so let's avoid the integer
|
||||
// division.
|
||||
eigen_assert(m_stride > index);
|
||||
inputIndex = index + m_inputOffset;
|
||||
@ -462,7 +462,7 @@ struct TensorEvaluator<TensorChippingOp<DimId, ArgType>, Device>
|
||||
}
|
||||
} else if ((static_cast<int>(this->Layout) == static_cast<int>(ColMajor) && this->m_dim.actualDim() == NumInputDims-1) ||
|
||||
(static_cast<int>(this->Layout) == static_cast<int>(RowMajor) && this->m_dim.actualDim() == 0)) {
|
||||
// m_stride is aways greater than index, so let's avoid the integer division.
|
||||
// m_stride is always greater than index, so let's avoid the integer division.
|
||||
eigen_assert(this->m_stride > index);
|
||||
this->m_impl.template writePacket<StoreMode>(index + this->m_inputOffset, x);
|
||||
} else {
|
||||
|
||||
@ -12,7 +12,7 @@
|
||||
|
||||
// This header file container defines fo gpu* macros which will resolve to
|
||||
// their equivalent hip* or cuda* versions depending on the compiler in use
|
||||
// A separte header (included at the end of this file) will undefine all
|
||||
// A separate header (included at the end of this file) will undefine all
|
||||
#include "TensorGpuHipCudaDefines.h"
|
||||
|
||||
namespace Eigen {
|
||||
|
||||
@ -25,9 +25,9 @@ template<typename T> struct MakePointer {
|
||||
};
|
||||
|
||||
// The PointerType class is a container of the device specefic pointer
|
||||
// used for refering to a Pointer on TensorEvaluator class. While the TensorExpression
|
||||
// used for referring to a Pointer on TensorEvaluator class. While the TensorExpression
|
||||
// is a device-agnostic type and need MakePointer class for type conversion,
|
||||
// the TensorEvaluator calss can be specialized for a device, hence it is possible
|
||||
// the TensorEvaluator calls can be specialized for a device, hence it is possible
|
||||
// to construct different types of temproray storage memory in TensorEvaluator
|
||||
// for different devices by specializing the following PointerType class.
|
||||
template<typename T, typename Device> struct PointerType : MakePointer<T>{};
|
||||
|
||||
@ -16,7 +16,7 @@
|
||||
// for some reason gets sent to the gcc/host compiler instead of the gpu/nvcc/hipcc compiler
|
||||
// When compiling such files, gcc will end up trying to pick up the CUDA headers by
|
||||
// default (see the code within "unsupported/Eigen/CXX11/Tensor" that is guarded by EIGEN_USE_GPU)
|
||||
// This will obsviously not work when trying to compile tensorflow on a sytem with no CUDA
|
||||
// This will obsviously not work when trying to compile tensorflow on a system with no CUDA
|
||||
// To work around this issue for HIP systems (and leave the default behaviour intact), the
|
||||
// HIP tensorflow build defines EIGEN_USE_HIP when compiling all source files, and
|
||||
// "unsupported/Eigen/CXX11/Tensor" has been updated to use HIP header when EIGEN_USE_HIP is
|
||||
|
||||
@ -965,7 +965,7 @@ struct TensorEvaluator<const TensorReductionOp<Op, Dims, ArgType, MakePointer_>,
|
||||
}
|
||||
}
|
||||
|
||||
// Intialize output coefficient reducers.
|
||||
// Initialize output coefficient reducers.
|
||||
for (int i = 0; i < num_reducers; ++i) {
|
||||
new (&reducers[i]) BlockReducer(m_reducer);
|
||||
}
|
||||
|
||||
@ -771,7 +771,7 @@ struct OuterReducer<Self, Op, GpuDevice> {
|
||||
// terminate called after throwing an instance of 'std::runtime_error'
|
||||
// what(): No device code available for function: _ZN5Eigen8internal20OuterReductionKernelIL...
|
||||
//
|
||||
// dont know why this happens (and why is it a runtime error instead of a compile time errror)
|
||||
// don't know why this happens (and why is it a runtime error instead of a compile time error)
|
||||
//
|
||||
// this will be fixed by HIP PR#457
|
||||
EIGEN_DEVICE_FUNC
|
||||
|
||||
@ -50,7 +50,7 @@ static void test_static_dimension_failure()
|
||||
.reshape(Tensor<int, 3>::Dimensions(2, 3, 1))
|
||||
.concatenate(right, 0);
|
||||
Tensor<int, 2, DataLayout> alternative = left
|
||||
// Clang compiler break with {{{}}} with an ambigous error on copy constructor
|
||||
// Clang compiler break with {{{}}} with an ambiguous error on copy constructor
|
||||
// the variadic DSize constructor added for #ifndef EIGEN_EMULATE_CXX11_META_H.
|
||||
// Solution:
|
||||
// either the code should change to
|
||||
|
||||
@ -433,7 +433,7 @@ static void test_execute_slice_lvalue(Device d)
|
||||
Tensor<T, NumDims, Options, Index> slice(slice_size);
|
||||
slice.setRandom();
|
||||
|
||||
// Asign a slice using default executor.
|
||||
// Assign a slice using default executor.
|
||||
Tensor<T, NumDims, Options, Index> golden = src;
|
||||
golden.slice(slice_start, slice_size) = slice;
|
||||
|
||||
|
||||
Loading…
Reference in New Issue
Block a user