squall/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
2e07c4cc2f
cutlass/test/unit/conv/device
History
Manish Gupta 2e07c4cc2f
CUTLASS 2.7 (#318) 
CUTLASS 2.7

Mainloop fusion for GEMM: summation over A or B
Strided DGRAD (optimized iterators)
Half-precision GELU_taylor activation functions
Use these when accumulation and epilogue compute types are all cutlass::half_t
Tuning and bug fixes to fused GEMM + GEMM example
Support for smaller than 128b aligned Convolutions: see examples
Caching of results to accelerate Convolution unit tests
Can be enabled or disabled by running cmake .. -DCUTLASS_TEST_ENABLE_CACHED_RESULTS=OFF
Corrections and bug fixes reported by the CUTLASS community
Thank you for filing these issues!

authored-by: Haicheng Wu haichengw@nvidia.com, Manish Gupta manigupta@nvidia.com, Dustyn Blasig dblasig@nvidia.com, Andrew Kerr akerr@nvidia.com
2021-09-20 11:02:22 -07:00
..
cache_testbed_output.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
CMakeLists.txt CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_dgrad_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm50.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_dgrad_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_dgrad_implicit_gemm_f16nhwc_f16nhwc_f16nhwc_tensor_op_f16_sm80.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_dgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm70.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_dgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm75.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_dgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_dgrad_implicit_gemm_f32nhwc_f32nhwc_f32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_dgrad_implicit_gemm_tf32nhwc_tf32nhwc_f32nhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm50.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f16nhwc_simt_f16_sm60.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f16nhwc_tensor_op_f16_sm80.cu CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f16nhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm70.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm75.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_fprop_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f32nhwc_f32nhwc_f32nhwc_simt_f32_sm50.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_f32nhwc_f32nhwc_f32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_qf32nhwc_qf32nhwc_qf32nhwc_simt_f32_sm50.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s4ncxhwx_s4cxrskx_s4ncxhwx_tensor_op_s32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s4ncxhwx_s4cxrskx_s4ncxhwx_tensor_op_s32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s4nhwc_s4nhwc_s32nhwc_tensor_op_s32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s4nhwc_s4nhwc_s32nhwc_tensor_op_s32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s8ncxhwx_s8cxrskx_s8ncxhwx_tensor_op_s32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s8ncxhwx_s8cxrskx_s8ncxhwx_tensor_op_s32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s8nhwc_s8nhwc_s32nhwc_tensor_op_s32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_s8nhwc_s8nhwc_s32nhwc_tensor_op_s32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_implicit_gemm_tf32nhwc_tf32nhwc_f32nhwc_tensor_op_f32_sm80.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_fprop_with_broadcast_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_fprop_with_reduction_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_problems.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_strided_dgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm80.cu CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_testbed_interleaved.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_testbed.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_wgrad_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm50.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_wgrad_implicit_gemm_cf32nhwc_cf32nhwc_cf32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_wgrad_implicit_gemm_f16nhwc_f16nhwc_f16nhwc_tensor_op_f16_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_wgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm70.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_wgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm75.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_wgrad_implicit_gemm_f16nhwc_f16nhwc_f32nhwc_tensor_op_f32_sm80.cu refine the implementation 2021-09-08 13:14:08 +00:00
conv2d_wgrad_implicit_gemm_f32nhwc_f32nhwc_f32nhwc_simt_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_wgrad_implicit_gemm_tf32nhwc_tf32nhwc_f32nhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv2d_with_broadcast_testbed.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv2d_with_reduction_testbed.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv3d_dgrad_implicit_gemm_f16ndhwc_f16ndhwc_f32ndhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_dgrad_implicit_gemm_tf32ndhwc_tf32ndhwc_f32ndhwc_tensor_op_f32_sm80.cu CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv3d_fprop_implicit_gemm_f16ndhwc_f16ndhwc_f32ndhwc_tensor_op_f32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_fprop_implicit_gemm_f16ndhwc_f16ndhwc_f32ndhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_fprop_implicit_gemm_tf32ndhwc_tf32ndhwc_f32ndhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_problems.h Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_testbed.h CUTLASS 2.7 (#318) 2021-09-20 11:02:22 -07:00
conv3d_wgrad_implicit_gemm_f16ndhwc_f16ndhwc_f32ndhwc_tensor_op_f32_sm75.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_wgrad_implicit_gemm_f16ndhwc_f16ndhwc_f32ndhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00
conv3d_wgrad_implicit_gemm_tf32ndhwc_tf32ndhwc_f32ndhwc_tensor_op_f32_sm80.cu Cutlass 2.6 Update 1 (#301) 2021-07-27 17:58:30 -07:00