MLIR  19.0.0git
Typedefs | Enumerations | Functions
SparseTensor.h File Reference
#include "mlir-c/AffineMap.h"
#include "mlir-c/IR.h"
#include "mlir/Dialect/SparseTensor/Transforms/Passes.capi.h.inc"

Go to the source code of this file.

Typedefs

typedef uint64_t MlirSparseTensorLevelType
 Dimension level types (and properties) that define sparse tensors. More...
 

Enumerations

enum  MlirSparseTensorLevelFormat {
  MLIR_SPARSE_TENSOR_LEVEL_DENSE = 0x000000010000 , MLIR_SPARSE_TENSOR_LEVEL_BATCH = 0x000000020000 , MLIR_SPARSE_TENSOR_LEVEL_COMPRESSED = 0x000000040000 , MLIR_SPARSE_TENSOR_LEVEL_SINGLETON = 0x000000080000 ,
  MLIR_SPARSE_TENSOR_LEVEL_LOOSE_COMPRESSED = 0x000000100000 , MLIR_SPARSE_TENSOR_LEVEL_N_OUT_OF_M = 0x000000200000
}
 
enum  MlirSparseTensorLevelPropertyNondefault { MLIR_SPARSE_PROPERTY_NON_UNIQUE = 0x0001 , MLIR_SPARSE_PROPERTY_NON_ORDERED = 0x0002 }
 

Functions

 MLIR_DECLARE_CAPI_DIALECT_REGISTRATION (SparseTensor, sparse_tensor)
 
MLIR_CAPI_EXPORTED bool mlirAttributeIsASparseTensorEncodingAttr (MlirAttribute attr)
 Checks whether the given attribute is a sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED MlirAttribute mlirSparseTensorEncodingAttrGet (MlirContext ctx, intptr_t lvlRank, MlirSparseTensorLevelType const *lvlTypes, MlirAffineMap dimToLvl, MlirAffineMap lvlTodim, int posWidth, int crdWidth)
 Creates a sparse_tensor.encoding attribute with the given parameters. More...
 
MLIR_CAPI_EXPORTED intptr_t mlirSparseTensorEncodingGetLvlRank (MlirAttribute attr)
 Returns the level-rank of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED MlirSparseTensorLevelType mlirSparseTensorEncodingAttrGetLvlType (MlirAttribute attr, intptr_t lvl)
 Returns a specified level-type of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED enum MlirSparseTensorLevelFormat mlirSparseTensorEncodingAttrGetLvlFmt (MlirAttribute attr, intptr_t lvl)
 Returns a specified level-format of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED MlirAffineMap mlirSparseTensorEncodingAttrGetDimToLvl (MlirAttribute attr)
 Returns the dimension-to-level mapping of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED MlirAffineMap mlirSparseTensorEncodingAttrGetLvlToDim (MlirAttribute attr)
 Returns the level-to-dimension mapping of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED int mlirSparseTensorEncodingAttrGetPosWidth (MlirAttribute attr)
 Returns the position bitwidth of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED int mlirSparseTensorEncodingAttrGetCrdWidth (MlirAttribute attr)
 Returns the coordinate bitwidth of the sparse_tensor.encoding attribute. More...
 
MLIR_CAPI_EXPORTED unsigned mlirSparseTensorEncodingAttrGetStructuredN (MlirSparseTensorLevelType lvlType)
 
MLIR_CAPI_EXPORTED unsigned mlirSparseTensorEncodingAttrGetStructuredM (MlirSparseTensorLevelType lvlType)
 
MLIR_CAPI_EXPORTED MlirSparseTensorLevelType mlirSparseTensorEncodingAttrBuildLvlType (enum MlirSparseTensorLevelFormat lvlFmt, const enum MlirSparseTensorLevelPropertyNondefault *properties, unsigned propSize, unsigned n, unsigned m)
 

Typedef Documentation

◆ MlirSparseTensorLevelType

typedef uint64_t MlirSparseTensorLevelType

Dimension level types (and properties) that define sparse tensors.

See the documentation in SparseTensorAttrDefs.td for their meaning.

These correspond to SparseTensorEncodingAttr::LevelType in the C++ API. If updating, keep them in sync and update the static_assert in the impl file.

Definition at line 28 of file SparseTensor.h.

Enumeration Type Documentation

◆ MlirSparseTensorLevelFormat

Enumerator
MLIR_SPARSE_TENSOR_LEVEL_DENSE 
MLIR_SPARSE_TENSOR_LEVEL_BATCH 
MLIR_SPARSE_TENSOR_LEVEL_COMPRESSED 
MLIR_SPARSE_TENSOR_LEVEL_SINGLETON 
MLIR_SPARSE_TENSOR_LEVEL_LOOSE_COMPRESSED 
MLIR_SPARSE_TENSOR_LEVEL_N_OUT_OF_M 

Definition at line 30 of file SparseTensor.h.

◆ MlirSparseTensorLevelPropertyNondefault

Enumerator
MLIR_SPARSE_PROPERTY_NON_UNIQUE 
MLIR_SPARSE_PROPERTY_NON_ORDERED 

Definition at line 39 of file SparseTensor.h.

Function Documentation

◆ MLIR_DECLARE_CAPI_DIALECT_REGISTRATION()

MLIR_DECLARE_CAPI_DIALECT_REGISTRATION ( SparseTensor  ,
sparse_tensor   
)

◆ mlirAttributeIsASparseTensorEncodingAttr()

MLIR_CAPI_EXPORTED bool mlirAttributeIsASparseTensorEncodingAttr ( MlirAttribute  attr)

Checks whether the given attribute is a sparse_tensor.encoding attribute.

Definition at line 43 of file SparseTensor.cpp.

References unwrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrBuildLvlType()

MLIR_CAPI_EXPORTED MlirSparseTensorLevelType mlirSparseTensorEncodingAttrBuildLvlType ( enum MlirSparseTensorLevelFormat  lvlFmt,
const enum MlirSparseTensorLevelPropertyNondefault properties,
unsigned  propSize,
unsigned  n,
unsigned  m 
)

◆ mlirSparseTensorEncodingAttrGet()

MLIR_CAPI_EXPORTED MlirAttribute mlirSparseTensorEncodingAttrGet ( MlirContext  ctx,
intptr_t  lvlRank,
MlirSparseTensorLevelType const *  lvlTypes,
MlirAffineMap  dimToLvl,
MlirAffineMap  lvlTodim,
int  posWidth,
int  crdWidth 
)

Creates a sparse_tensor.encoding attribute with the given parameters.

Definition at line 48 of file SparseTensor.cpp.

References mlir::get(), unwrap(), and wrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetCrdWidth()

MLIR_CAPI_EXPORTED int mlirSparseTensorEncodingAttrGetCrdWidth ( MlirAttribute  attr)

Returns the coordinate bitwidth of the sparse_tensor.encoding attribute.

Definition at line 90 of file SparseTensor.cpp.

References unwrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetDimToLvl()

MLIR_CAPI_EXPORTED MlirAffineMap mlirSparseTensorEncodingAttrGetDimToLvl ( MlirAttribute  attr)

Returns the dimension-to-level mapping of the sparse_tensor.encoding attribute.

Definition at line 61 of file SparseTensor.cpp.

References unwrap(), and wrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetLvlFmt()

MLIR_CAPI_EXPORTED enum MlirSparseTensorLevelFormat mlirSparseTensorEncodingAttrGetLvlFmt ( MlirAttribute  attr,
intptr_t  lvl 
)

Returns a specified level-format of the sparse_tensor.encoding attribute.

Definition at line 74 of file SparseTensor.cpp.

References unwrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetLvlToDim()

MLIR_CAPI_EXPORTED MlirAffineMap mlirSparseTensorEncodingAttrGetLvlToDim ( MlirAttribute  attr)

Returns the level-to-dimension mapping of the sparse_tensor.encoding attribute.

Definition at line 65 of file SparseTensor.cpp.

References unwrap(), and wrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetLvlType()

MLIR_CAPI_EXPORTED MlirSparseTensorLevelType mlirSparseTensorEncodingAttrGetLvlType ( MlirAttribute  attr,
intptr_t  lvl 
)

Returns a specified level-type of the sparse_tensor.encoding attribute.

Definition at line 74 of file SparseTensor.cpp.

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetPosWidth()

MLIR_CAPI_EXPORTED int mlirSparseTensorEncodingAttrGetPosWidth ( MlirAttribute  attr)

Returns the position bitwidth of the sparse_tensor.encoding attribute.

Definition at line 86 of file SparseTensor.cpp.

References unwrap().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetStructuredM()

MLIR_CAPI_EXPORTED unsigned mlirSparseTensorEncodingAttrGetStructuredM ( MlirSparseTensorLevelType  lvlType)

Definition at line 113 of file SparseTensor.cpp.

References mlir::sparse_tensor::getM().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingAttrGetStructuredN()

MLIR_CAPI_EXPORTED unsigned mlirSparseTensorEncodingAttrGetStructuredN ( MlirSparseTensorLevelType  lvlType)

Definition at line 108 of file SparseTensor.cpp.

References mlir::sparse_tensor::getN().

Referenced by populateDialectSparseTensorSubmodule().

◆ mlirSparseTensorEncodingGetLvlRank()

MLIR_CAPI_EXPORTED intptr_t mlirSparseTensorEncodingGetLvlRank ( MlirAttribute  attr)

Returns the level-rank of the sparse_tensor.encoding attribute.

Definition at line 69 of file SparseTensor.cpp.

References unwrap().

Referenced by populateDialectSparseTensorSubmodule().