MCPcopy Create free account

hub / github.com/NVIDIA/cutlass / endpoints

Endpoints26 in github.com/NVIDIA/cutlass

⨍ Functions 19,553◇ Types & classes 10,397↳ Endpoints 26
Route__add__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__and__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__eq__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__floordiv__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__ge__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__gt__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__le__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__lshift__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__lt__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__mod__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__mul__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__ne__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__or__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__rshift__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__sub__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__truediv__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Route__xor__
_dispatch_to_rhs_r_op
python/CuTeDSL/cutlass/base_dsl/_mlir_helpers/arith.py:None
Routeget_num_tmem_alloc_cols
deprecated("API is deprecated, use cutlass.utils.get_num_tmem_alloc_cols instead")
python/CuTeDSL/cutlass/utils/blackwell_helpers.py:None
Routetest_bench
pytest.mark.parametrize( "mnkl, tile_shape_mn, cluster_shape_mn, mma_promotion_interval, label", [
test/examples/CuTeDSL/hopper/test_dense_gemm_fp8_2xacc.py:None
Routetest_gemm_equivalence_f16_f16_f16_f16_f16_ttt_4_4_4
unittest.skipIf(device_cc() < 70, "Device compute capability is insufficient for FP16 Tensor Core tests.")
test/python/cutlass/interface/gemm_interface.py:None
Routetest_gemm_equivalence_f16_f16_f16_f16_f16_ttt_8_8_8
unittest.skipIf(device_cc() < 70, "Device compute capability is insufficient for FP16 Tensor Core tests.")
test/python/cutlass/interface/gemm_interface.py:None
Routetest_gemm_equivalence_f16_f16_f16_f16_f32_ntn_8_8_8
unittest.skipIf(device_cc() < 70, "Device compute capability is insufficient for FP16 Tensor Core tests.")
test/python/cutlass/interface/gemm_interface.py:None
Routetest_gemm_equivalence_f64_f64_f64_f64_f64_tnt_1_1_1
unittest.skipIf(device_cc() < 80, "Device compute capability is insufficient for F64 Tensor Core tests.")
test/python/cutlass/interface/gemm_interface.py:None
Routetest_l0_dtypes
pytest.mark.parametrize( "a_dtype, b_dtype, c_dtype, acc_dtype, problem_sizes_mnkl", [ # fp16
test/examples/CuTeDSL/hopper/test_grouped_gemm.py:None
Routetest_l0_major_modes
pytest.mark.parametrize( "a_major, b_major, c_major, problem_sizes_mnkl, tile_shape_mn", [ # k
test/examples/CuTeDSL/hopper/test_grouped_gemm.py:None
Routetest_opclass_switch
unittest.skipIf(device_cc() < 70, "Device compute capability is insufficient for F16 Tensor Core tests.")
test/python/cutlass/interface/gemm_interface.py:None