hongfangyu
Results
3
comments of
hongfangyu
just add ADD_DEFINITIONS(-DCV__ENABLE_C_API_CTORS) in cmakelist
same error with ROCE. 2 nodes 16 GPUs under the same TOR is OK, but 4 nodes 32 gpus fail.
this repo is ok，about 45GB/s https://github.com/Infrawaves/DeepEP_ibrc_dual-ports_multiQP