leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

v4.1 release update v2. (#2481)

Browse Source
This commit is contained in:
Junkai-Wu
2025-07-22 10:03:55 +08:00
committed by GitHub
parent 9baa06dd57
commit fd6cfe1ed0
179 changed files with 7878 additions and 1286 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/13_two_tensor_op_fusion/kernel/b2b_gemm.h
View File

@@ -248,7 +248,7 @@ struct B2bGemm {
typename Epilogue::OutputTileIterator::TensorRef* ref_C1;
typename Epilogue::OutputTileIterator::TensorRef* ref_D1;
// Epilogue params remain constant across all problmes in the group. Thus,
// Epilogue params remain constant across all problems in the group. Thus,
// the parameter here is not a pointer.
typename OutputOp0::Params epilogue0;
typename OutputOp1::Params epilogue1;
@@ -402,7 +402,7 @@ struct B2bGemm {
typename Epilogue::OutputTileIterator::TensorRef* ref_C1;
typename Epilogue::OutputTileIterator::TensorRef* ref_D1;
// Epilogue params remain constant across all problmes in the group. Thus,
// Epilogue params remain constant across all problems in the group. Thus,
// the parameter here is not a pointer.
typename OutputOp0::Params output_op_0;
typename OutputOp1::Params output_op_1;
@@ -434,7 +434,7 @@ struct B2bGemm {
// Only row-major outputs are currently supported, so no transpose is performed
}
/// Returns non-grouped paramaters to be used as input to the kernel-level
/// Returns non-grouped parameters to be used as input to the kernel-level
/// operator for the problem indicated by problem_visitor.
CUTLASS_HOST_DEVICE
Params to_single_params(const ProblemVisitor& problem_visitor) const {

2
examples/13_two_tensor_op_fusion/kernel/default_b2b_conv2d_fprop_sm80.h
View File

@@ -560,7 +560,7 @@ struct DefaultB2bConv2dFprop <
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Defines a kernel for Conv2dFprop specialization for Optimzed IteratorAlgorithm and
/// Defines a kernel for Conv2dFprop specialization for Optimized IteratorAlgorithm and
// multistage pipeline with interleaved layout.
template <
typename ElementA,

2
examples/13_two_tensor_op_fusion/kernel/default_b2b_conv2d_fprop_smem_accumulator_sm80.h
View File

@@ -606,7 +606,7 @@ struct DefaultB2bConv2dFprop <
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Defines a kernel for Conv2dFprop specialization for Optimzed IteratorAlgorithm and
/// Defines a kernel for Conv2dFprop specialization for Optimized IteratorAlgorithm and
// multistage pipeline with interleaved layout.
/// Accumulator will be staged in shared memory.
template <