Commit 1fa274b1 authored by Ayush Dubey's avatar Ayush Dubey Committed by TensorFlower Gardener
Browse files

Better error checking and testing in NcclManager. 

After this change, we check the return value of every CUDA and NCCL call in
NcclManager.  If any call is unsuccessful, we call the NCCL callback with an
error status.

This change also re-enables NCCL tests.

PiperOrigin-RevId: 224038066
parent 2efc17d7
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment