CUTLASS: cutlass::gemm::warp Namespace Reference

| | CUTLASS

CUDA Templates for Linear Algebra Subroutines and Solvers |

• Main Page

• Modules

• Namespaces

• Classes

• Files

• Namespace List

• Namespace Members

• cutlass

• gemm

• warp

Classes

cutlass::gemm::warp Namespace Reference

|

Classes

| | struct | DefaultMmaTensorOp | | | Partial specialization for m-by-n-by-kgroup. More...
| | | | class | MmaComplexTensorOp | | | | class | MmaComplexTensorOp< Shape_, complex< RealElementA >, LayoutA_, complex< RealElementB >, LayoutB_, complex< RealElementC >, LayoutC_, Policy_, TransformA, TransformB, Enable > | | | Partial specialization for complex*complex+complex => complex using real-valued TensorOps. More...
| | | | class | MmaSimt | | | Structure to compute the matrix product targeting CUDA cores and SIMT math instructions. More...
| | | | struct | MmaSimtPolicy | | | Describes the arrangement and configuration of per-lane operations in warp-level matrix multiply. More...
| | | | class | MmaSimtTileIterator | | | | class | MmaSimtTileIterator< Shape_, Operand::kA, Element_, layout::ColumnMajor, Policy_, PartitionsK, PartitionGroupSize > | | | | class | MmaSimtTileIterator< Shape_, Operand::kA, Element_, layout::ColumnMajorInterleaved< 4 >, Policy_, PartitionsK, PartitionGroupSize > | | | | class | MmaSimtTileIterator< Shape_, Operand::kB, Element_, layout::RowMajor, Policy_, PartitionsK, PartitionGroupSize > | | | | class | MmaSimtTileIterator< Shape_, Operand::kB, Element_, layout::RowMajorInterleaved< 4 >, Policy_, PartitionsK, PartitionGroupSize > | | | | class | MmaSimtTileIterator< Shape_, Operand::kC, Element_, layout::ColumnMajor, Policy_ > | | | | class | MmaSimtTileIterator< Shape_, Operand::kC, Element_, layout::RowMajor, Policy_ > | | | | class | MmaTensorOp | | | Structure to compute the matrix product targeting CUDA cores and SIMT math instructions. More...
| | | | class | MmaTensorOpAccumulatorTileIterator | | | | class | [MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajor, InstructionShape_, OpDelta_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape 00_01Element 008f607b871a2b3d854eb4def64712c042.html) | | | | class | [MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::ColumnMajorInterleaved< InterleavedN >, InstructionShape_, OpDelta_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape 00_01Element 00027dabdc144edd6276f664ca74088510.html) | | | | class | [MmaTensorOpAccumulatorTileIterator< Shape_, Element_, cutlass::layout::RowMajor, InstructionShape_, OpDelta_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpAccumulatorTileIterator_3_01Shape 00_01Element 006c39f57875e0aa9d0ad82c8043ed8b98.html) | | | | class | MmaTensorOpMultiplicandTileIterator | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::ColumnMajorTensorOpMultiplicandCongruous< sizeof_bits< Element_ >::value, int(128/sizeof(Element_))>, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 0b84f53cd44b339eccc12067c9f86e11c.html) | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::ColumnMajorTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Crosswise >, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 0e52ad425e1ee3e68544873f66733237b.html) | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::RowMajorTensorOpMultiplicandCongruous< sizeof_bits< Element_ >::value, int(128/sizeof(Element_))>, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 039819fb3ccd43786d556c2c9669508ef.html) | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::RowMajorTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Crosswise >, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 0352e0dcab42bc8360606874e00173556.html) | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCongruous< sizeof_bits< Element_ >::value, 64 >, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 0ed7daaeba1c095e77f68533d4d2c475c.html) | | | | class | [MmaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::TensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, Crosswise >, InstructionShape_, OpDelta_, 32, PartitionsK_ >](classcutlass_1_1gemm_1_1warp_1_1MmaTensorOpMultiplicandTileIterator_3_01Shape 00_01Operand 0c7d419c589d601ce4eb603be566fea21.html) | | | | struct | MmaTensorOpPolicy | | | Policy. More...
| | | | class | MmaVoltaTensorOp | | | Structure to compute the matrix product targeting CUDA cores and SIMT math instructions. More...
| | | | class | MmaVoltaTensorOpAccumulatorTileIterator | | | | class | MmaVoltaTensorOpMultiplicandTileIterator | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand::kA, Element_, cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCongruous< sizeof_bits< Element_ >::value >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand::kA, Element_, cutlass::layout::VoltaTensorOpMultiplicandCongruous< sizeof_bits< Element_ >::value >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand::kB, Element_, cutlass::layout::RowMajorVoltaTensorOpMultiplicandBCongruous< sizeof_bits< Element_ >::value >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand::kB, Element_, cutlass::layout::VoltaTensorOpMultiplicandBCongruous< sizeof_bits< Element_ >::value >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::ColumnMajorVoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, KBlock >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::RowMajorVoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, KBlock >, InstructionShape_, OpDelta_, 32 > | | | | class | MmaVoltaTensorOpMultiplicandTileIterator< Shape_, Operand_, Element_, cutlass::layout::VoltaTensorOpMultiplicandCrosswise< sizeof_bits< Element_ >::value, KBlock >, InstructionShape_, OpDelta_, 32 > | | | | struct | WarpSize | | | Query the number of threads per warp. More...
| | |

---

Generated by 1.8.11