Code
Hub
Workspaces
Connect
Indexed graphs
Engine
MCP
copy
hub
/
github.com/facebookresearch/dlrm
/ wait
Method
wait
extend_distributed.py:200–204 ·
view source on GitHub ↗
(self)
Source
from the content-addressed store, hash-verified
198
self.WaitFunction = All2All_Scatter_Wait
199
200
def
wait(self):
201
ret = self.WaitFunction.apply(*self.tensor)
202
self.req = None
203
self.tensor = None
204
return
ret
205
206
207
class
All2All_ScatterList_Req(Function):
Callers
7
distributed_forward
Method · 0.80
backward
Method · 0.80
forward
Method · 0.80
backward
Method · 0.80
forward
Method · 0.80
backward
Method · 0.80
forward
Method · 0.80
Calls
no outgoing calls
Tested by
no test coverage detected