MCPcopy Create free account
hub / github.com/apache/tvm / CpAsyncBulkNamespace

Class CpAsyncBulkNamespace

python/tvm/backend/cuda/script.py:158–171  ·  view source on GitHub ↗

The CpAsyncBulk instruction submodule.

Source from the content-addressed store, hash-verified

156
157
158class CpAsyncBulkNamespace:
159 """The CpAsyncBulk instruction submodule."""
160
161 def __init__(self):
162 self.commit_group = _op_wrapper(_cuda_op.ptx_cp_async_bulk_commit_group)
163 self.wait_group = _op_wrapper(_cuda_op.ptx_cp_async_bulk_wait_group)
164 self.tensor = CpAsyncBulkTensorNamespace()
165 self.s2c = _op_wrapper(_cuda_op.ptx_cp_async_bulk_shared_to_cluster)
166
167 def __call__(self, *args, **kwds):
168 return _dtype_forward(_cuda_op.ptx_cp_async_bulk)(*args, **kwds)
169
170 # __call__ corresponds to ptx_cp_async_bulk
171 __tir_call_op_name__ = "ptx_cp_async_bulk"
172
173
174class CpAsyncBulkTensorNamespace:

Callers 1

__init__Method · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…