Enumerations

Enumerations
enum class	slate::Target : char {   slate::Target::Host = 'H' , slate::Target::HostTask = 'T' , slate::Target::HostNest = 'N' , slate::Target::HostBatch = 'B' ,   slate::Target::Devices = 'D' }
	Location and method of computation. More...
enum class	slate::Option : char {   slate::Option::ChunkSize , slate::Option::Lookahead , slate::Option::BlockSize , slate::Option::InnerBlocking ,   slate::Option::MaxPanelThreads , slate::Option::Tolerance , slate::Option::Target , slate::Option::HoldLocalWorkspace ,   slate::Option::Depth , slate::Option::MaxIterations , slate::Option::UseFallbackSolver , slate::Option::PivotThreshold ,   slate::Option::PrintVerbose = 50 , slate::Option::PrintEdgeItems , slate::Option::PrintWidth , slate::Option::PrintPrecision ,   slate::Option::MethodCholQR = 60 , slate::Option::MethodEig , slate::Option::MethodGels , slate::Option::MethodGemm ,   slate::Option::MethodHemm , slate::Option::MethodLU , slate::Option::MethodTrsm , slate::Option::MethodSVD }
	Keys for options to pass to SLATE routines. More...
enum class	slate::LayoutConvert : char { slate::LayoutConvert::ColMajor = 'C' , slate::LayoutConvert::RowMajor = 'R' , slate::LayoutConvert::None = 'N' }
	To convert matrix between column-major and row-major. More...
enum class	slate::NormScope : char { slate::NormScope::Columns = 'C' , slate::NormScope::Rows = 'R' , slate::NormScope::Matrix = 'M' }
	Whether computing matrix norm, column norms, or row norms. More...
enum class	slate::GridOrder : char { slate::GridOrder::Col = 'C' , slate::GridOrder::Row = 'R' , slate::GridOrder::Unknown = 'U' }
	Order to map MPI processes to tile grid. More...
enum class	slate::TileKind : char { slate::TileKind::Workspace = 'w' , slate::TileKind::SlateOwned = 'o' , slate::TileKind::UserOwned = 'u' }
	Whether a tile is workspace or origin (local non-workspace), and who owns (allocated, deallocates) the data. More...

Variables
const int	slate::MinOmpActiveLevels = 4
	Minimum number of OpenMP active parallel levels to allow multi-threaded panel implementation.

Detailed Description

---

---

Enumeration Type Documentation

GridOrder

enum class slate::GridOrder : char

strong

Order to map MPI processes to tile grid.

Enumerator
Col	Column major.
Row	Row major.
Unknown	Unknown (e.g., if using lambda functions)

LayoutConvert

enum class slate::LayoutConvert : char

strong

To convert matrix between column-major and row-major.

Enumerator
ColMajor	Convert to column-major.
RowMajor	Convert to row-major.
None	No conversion.

NormScope

enum class slate::NormScope : char

strong

Whether computing matrix norm, column norms, or row norms.

Enumerator
Columns	Compute column norms.
Rows	Compute row norms.
Matrix	Compute matrix norm.

Option

enum class slate::Option : char

strong

Keys for options to pass to SLATE routines.

Enumerator
ChunkSize	chunk size, >= 1
Lookahead	lookahead depth, >= 0
BlockSize	block size, >= 1
InnerBlocking	inner blocking size, >= 1
MaxPanelThreads	max number of threads for panel, >= 1
Tolerance	tolerance for iterative methods, default epsilon
Target	computation method ( See also Target)
HoldLocalWorkspace	do not erase local workspace tiles for enabling resue of the tiles by the next routine
Depth	depth for the RBT solver
MaxIterations	maximum iteration count
UseFallbackSolver	whether to fallback to a robust solver if iterations do not converge
PivotThreshold	threshold for pivoting, >= 0, <= 1
PrintVerbose	verbose, 0: no printing, verbose, 1: print metadata only (dimensions, uplo, etc.) verbose, 2: print first & last PrintEdgeItems rows & cols from the four corner tiles verbose, 3: print 4 corner elements of every tile verbose, 4: print full matrix
PrintEdgeItems	edgeitems: number of first & last rows & cols of matrix to print
PrintWidth	width print format specifier
PrintPrecision	precision print format specifier For correct printing, PrintWidth = PrintPrecision + 6.
MethodCholQR	Select the algorithm to compute A^H A.
MethodEig	Select the algorithm to compute eigenpairs of tridiagonal matrix.
MethodGels	Select the gels algorithm.
MethodGemm	Select the gemm algorithm.
MethodHemm	Select the hemm algorithm.
MethodLU	Select the LU (getrf) algorithm.
MethodTrsm	Select the trsm algorithm.
MethodSVD	Select the algorithm to compute singular values of bidiagonal matrix.

Target

enum class slate::Target : char

strong

Location and method of computation.

Enumerator
Host	data resides on host
HostTask	computation using OpenMP nested tasks on host
HostNest	computation using OpenMP nested parallel for loops on host
HostBatch	computation using batch BLAS on host (Intel MKL)
Devices	computation using batch BLAS on devices (cuBLAS)

TileKind

enum class slate::TileKind : char

strong

Whether a tile is workspace or origin (local non-workspace), and who owns (allocated, deallocates) the data.

Enumerator
Workspace	SLATE allocated workspace tile.
SlateOwned	SLATE allocated origin tile.
UserOwned	User owned origin tile.