leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

CUTLASS 2.0 (#62)

Browse Source 
CUTLASS 2.0

Substantially refactored for

- Better performance, particularly for native Turing Tensor Cores
- Robust and durable templates spanning the design space
- Encapsulated functionality embodying modern C++11 programming techniques
- Optimized containers and data types for efficient, generic, portable device code

Updates to:
- Quick start guide
- Documentation
- Utilities
- CUTLASS Profiler

Native Turing Tensor Cores
- Efficient GEMM kernels targeting Turing Tensor Cores
- Mixed-precision floating point, 8-bit integer, 4-bit integer, and binarized operands

Coverage of existing CUTLASS functionality:
- GEMM kernels targeting CUDA and Tensor Cores in NVIDIA GPUs
- Volta Tensor Cores through native mma.sync and through WMMA API
- Optimizations such as parallel reductions, threadblock rasterization, and intra-threadblock reductions
- Batched GEMM operations
- Complex-valued GEMMs

Note: this commit and all that follow require a host compiler supporting C++11 or greater.
This commit is contained in:
Andrew Kerr
2019-11-19 16:55:34 -08:00
committed by GitHub
parent b5cab177a9
commit fb335f6a5f
5434 changed files with 599799 additions and 250176 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/search/namespaces_0.js
+34 -2
View File
@@ -1,9 +1,41 @@
var searchData=
[
['arch',['arch',['../namespacecutlass_1_1arch.html',1,'cutlass']]],
['cutlass',['cutlass',['../namespacecutlass.html',1,'']]],
['debug',['debug',['../namespacecutlass_1_1debug.html',1,'cutlass']]],
['detail',['detail',['../namespacecutlass_1_1detail.html',1,'cutlass']]],
['detail',['detail',['../namespacecutlass_1_1gemm_1_1thread_1_1detail.html',1,'cutlass::gemm::thread']]],
['detail',['detail',['../namespacecutlass_1_1reference_1_1host_1_1detail.html',1,'cutlass::reference::host']]],
['detail',['detail',['../namespacecutlass_1_1epilogue_1_1threadblock_1_1detail.html',1,'cutlass::epilogue::threadblock']]],
['detail',['detail',['../namespacecutlass_1_1gemm_1_1threadblock_1_1detail.html',1,'cutlass::gemm::threadblock']]],
['detail',['detail',['../namespacecutlass_1_1reference_1_1detail.html',1,'cutlass::reference']]],
['detail',['detail',['../namespacecutlass_1_1gemm_1_1kernel_1_1detail.html',1,'cutlass::gemm::kernel']]],
['detail',['detail',['../namespacecutlass_1_1reference_1_1device_1_1detail.html',1,'cutlass::reference::device']]],
['detail',['detail',['../namespacecutlass_1_1reference_1_1device_1_1kernel_1_1detail.html',1,'cutlass::reference::device::kernel']]],
['device',['device',['../namespacecutlass_1_1gemm_1_1device.html',1,'cutlass::gemm']]],
['device',['device',['../namespacecutlass_1_1reference_1_1device.html',1,'cutlass::reference']]],
['device_5fmemory',['device_memory',['../namespacecutlass_1_1device__memory.html',1,'cutlass']]],
['epilogue',['epilogue',['../namespacecutlass_1_1epilogue.html',1,'cutlass']]],
['gemm',['gemm',['../namespacecutlass_1_1gemm.html',1,'cutlass']]],
['matrixlayout',['MatrixLayout',['../namespacecutlass_1_1MatrixLayout.html',1,'cutlass']]],
['host',['host',['../namespacecutlass_1_1reference_1_1host.html',1,'cutlass::reference']]],
['kernel',['kernel',['../namespacecutlass_1_1reduction_1_1kernel.html',1,'cutlass::reduction']]],
['kernel',['kernel',['../namespacecutlass_1_1gemm_1_1kernel.html',1,'cutlass::gemm']]],
['kernel',['kernel',['../namespacecutlass_1_1reference_1_1device_1_1kernel.html',1,'cutlass::reference::device']]],
['layout',['layout',['../namespacecutlass_1_1layout.html',1,'cutlass']]],
['library',['library',['../namespacecutlass_1_1library.html',1,'cutlass']]],
['platform',['platform',['../namespacecutlass_1_1platform.html',1,'cutlass']]],
['reduction',['reduction',['../namespacecutlass_1_1reduction.html',1,'cutlass']]]
['reduction',['reduction',['../namespacecutlass_1_1reduction.html',1,'cutlass']]],
['reference',['reference',['../namespacecutlass_1_1reference.html',1,'cutlass']]],
['thread',['thread',['../namespacecutlass_1_1gemm_1_1thread.html',1,'cutlass::gemm']]],
['thread',['thread',['../namespacecutlass_1_1reference_1_1device_1_1thread.html',1,'cutlass::reference::device']]],
['thread',['thread',['../namespacecutlass_1_1thread.html',1,'cutlass']]],
['thread',['thread',['../namespacecutlass_1_1reduction_1_1thread.html',1,'cutlass::reduction']]],
['thread',['thread',['../namespacecutlass_1_1epilogue_1_1thread.html',1,'cutlass::epilogue']]],
['thread',['thread',['../namespacecutlass_1_1transform_1_1thread.html',1,'cutlass::transform']]],
['threadblock',['threadblock',['../namespacecutlass_1_1epilogue_1_1threadblock.html',1,'cutlass::epilogue']]],
['threadblock',['threadblock',['../namespacecutlass_1_1gemm_1_1threadblock.html',1,'cutlass::gemm']]],
['threadblock',['threadblock',['../namespacecutlass_1_1transform_1_1threadblock.html',1,'cutlass::transform']]],
['transform',['transform',['../namespacecutlass_1_1transform.html',1,'cutlass']]],
['warp',['warp',['../namespacecutlass_1_1epilogue_1_1warp.html',1,'cutlass::epilogue']]],
['warp',['warp',['../namespacecutlass_1_1gemm_1_1warp.html',1,'cutlass::gemm']]]
];