![]() |
cuLite v0.3.1
A lite CUDA C++ Interface
|
The floating point device dense vector class. More...

Constructors | |
| XxVector () | |
| Default constructor. | |
| XxVector (int_t n) | |
| Dimension constructor. | |
| XxVector (int_t n, T_Scalar *vals, bool bind) | |
| Auxiliary constructor. | |
| XxVector (const XxVector< T_Scalar > &other)=default | |
| Copy constructor. | |
| XxVector (XxVector< T_Scalar > &&other)=default | |
| Move constructor. | |
| ~XxVector () | |
| Destructor. | |
Operators | |
| XxVector< T_Scalar > & | operator= (const XxVector< T_Scalar > &other)=default |
| Copy assignment operator. | |
| XxVector< T_Scalar > & | operator= (XxVector< T_Scalar > &&other)=default |
| Move assignment operator. | |
| void | operator= (T_Scalar val) |
| Fill operator. | |
| alias::VirtualScal_vec< T_Scalar > | operator- () const |
| Unary negation operator. | |
Public Member Functions | |
| void | iscale (const T_Scalar &val) |
| Scale the device vector in-place. | |
| VirtualRowvec< T_Scalar > | transpose () const |
| Transpose the vector. | |
| VirtualRowvec< T_Scalar > | ctranspose () const |
| Conjugate transpose the vector. | |
| alias::VirtualConj_vec< T_Scalar > | conjugate () const |
| Compute the complex conjugate. | |
| void | iconjugate () |
| Conjugate the device vector in-place. | |
| T_RScalar | normOne () const |
| Compute the 1-norm. | |
| T_RScalar | normInf () const |
| Compute the infinity norm. | |
| T_RScalar | normEuc () const |
| Compute the Euclidean norm (2-norm). | |
| XxVector< T_Scalar > | block (int_t ibgn, int_t ni) const |
| Extract a block as a new device vector. | |
| XxVector< T_Scalar > | rblock (int_t ibgn, int_t ni) |
| Extract a reference block. | |
| Guard< XxVector< T_Scalar > > | rblock (int_t ibgn, int_t ni) const |
| Extract a guarded reference block. | |
| void | setBlock (int_t ibgn, const XxVector< T_Scalar > &src) |
| Set a block of elements. | |
Additional Inherited Members | |
| void | clear () |
| Clear the device vector. | |
| void | fill (T_Scalar val) |
| Fill all elements with a value. | |
| XiVector< T_Scalar > | copy () const |
| Create a deep copy of the device vector. | |
| XiVector< T_Scalar > | rcopy () |
| Create a reference copy (shallow copy) of the device vector. | |
| Guard< XiVector< T_Scalar > > | rcopy () const |
| Create a guarded reference copy (shallow copy) of the device vector. | |
| XiVector< T_Scalar > | move () |
| Move the device vector's resources. | |
| std::string | info (const std::string &header="") const |
| Get information about the device vector. | |
| void | copyToHost (::cla3p::dns::XiVector< T_Cla3pScalar > &dest) const |
| Copies the device vector to a host vector. | |
| void | copyFromHost (const ::cla3p::dns::XiVector< T_Cla3pScalar > &src) |
| Copies a host vector to the device vector. | |
| static Guard< XiVector< T_Scalar > > | view (int_t n, const T_Scalar *vals) |
| Create a view of existing device memory. | |
The floating point device dense vector class.
Represents a floating-point dense vector stored on the GPU device.
| T_Scalar | The scalar type (e.g., float, double, complex). |
| culite::dns::XxVector< T_Scalar >::XxVector | ( | ) |
Default constructor.
Creates an empty device vector with no allocated device memory.
|
explicit |
Dimension constructor.
Creates a device vector of the specified size and allocates device memory.
| [in] | n | The number of elements in the vector. |
|
explicit |
Auxiliary constructor.
Creates a device vector using existing device memory.
| [in] | n | The number of elements in the vector. |
| [in] | vals | Pointer to existing device memory. |
| [in] | bind | If true, the vector takes ownership of the memory. |
|
default |
Copy constructor.
Creates a new device vector by copying another device vector.
| [in] | other | The device vector to copy. |
|
default |
Move constructor.
Creates a new device vector by moving resources from another device vector.
| [in] | other | The device vector to move from. |
| culite::dns::XxVector< T_Scalar >::~XxVector | ( | ) |
Destructor.
Destroys the device vector and releases allocated device memory.
|
default |
Copy assignment operator.
Copies the contents of another device vector to this device vector.
| [in] | other | The device vector to copy. |
|
default |
Move assignment operator.
Moves resources from another device vector to this device vector.
| [in] | other | The device vector to move from. |
| void culite::dns::XxVector< T_Scalar >::operator= | ( | T_Scalar | val | ) |
Fill operator.
Fills all elements with the specified value.
| [in] | val | The scalar value to fill with. |
| alias::VirtualScal_vec< T_Scalar > culite::dns::XxVector< T_Scalar >::operator- | ( | ) | const |
Unary negation operator.
Returns a negated copy of the device vector.
| void culite::dns::XxVector< T_Scalar >::iscale | ( | const T_Scalar & | val | ) |
Scale the device vector in-place.
Multiplies all elements of the device vector by a scalar value.
| [in] | val | The scalar value to multiply by. |
| VirtualRowvec< T_Scalar > culite::dns::XxVector< T_Scalar >::transpose | ( | ) | const |
Transpose the vector.
Returns a row vector view of this column vector.
| VirtualRowvec< T_Scalar > culite::dns::XxVector< T_Scalar >::ctranspose | ( | ) | const |
Conjugate transpose the vector.
Returns a conjugate transposed row vector view of this column vector.
| alias::VirtualConj_vec< T_Scalar > culite::dns::XxVector< T_Scalar >::conjugate | ( | ) | const |
Compute the complex conjugate.
Returns a device vector containing the complex conjugate of each element.
| void culite::dns::XxVector< T_Scalar >::iconjugate | ( | ) |
Conjugate the device vector in-place.
Replaces all elements with their complex conjugates.
| T_RScalar culite::dns::XxVector< T_Scalar >::normOne | ( | ) | const |
Compute the 1-norm.
Computes the sum of absolute values of all elements.
| T_RScalar culite::dns::XxVector< T_Scalar >::normInf | ( | ) | const |
Compute the infinity norm.
Computes the maximum absolute value of all elements.
| T_RScalar culite::dns::XxVector< T_Scalar >::normEuc | ( | ) | const |
Compute the Euclidean norm (2-norm).
Computes the square root of the sum of squared absolute values.
| XxVector< T_Scalar > culite::dns::XxVector< T_Scalar >::block | ( | int_t | ibgn, |
| int_t | ni ) const |
Extract a block as a new device vector.
Creates a new device vector containing a deep copy of a contiguous block of elements.
| [in] | ibgn | The starting index of the block. |
| [in] | ni | The number of elements in the block. |
| XxVector< T_Scalar > culite::dns::XxVector< T_Scalar >::rblock | ( | int_t | ibgn, |
| int_t | ni ) |
Extract a reference block.
Creates a device vector that references a contiguous block of this vector's device memory.
| [in] | ibgn | The starting index of the block. |
| [in] | ni | The number of elements in the block. |
| Guard< XxVector< T_Scalar > > culite::dns::XxVector< T_Scalar >::rblock | ( | int_t | ibgn, |
| int_t | ni ) const |
Extract a guarded reference block.
Creates a guarded device vector that references a contiguous block of this vector's device memory.
| [in] | ibgn | The starting index of the block. |
| [in] | ni | The number of elements in the block. |
| void culite::dns::XxVector< T_Scalar >::setBlock | ( | int_t | ibgn, |
| const XxVector< T_Scalar > & | src ) |
Set a block of elements.
Copies elements from the source device vector to a contiguous block of this vector.
| [in] | ibgn | The starting index where the block will be written. |
| [in] | src | The source device vector to copy from. |