c3sr / comm_scope

NUMA-aware multi-CPU multi-GPU data transfer benchmarks
https://github.com/c3sr/scope
Apache License 2.0
21 stars 3 forks source link

ThetaGPU: error in src/cudaMemcpyPeerAsync_Duplex_GPUGPUPeer: a PTX JIT compilation failed #42

Open cwpearson opened 3 years ago

cwpearson commented 3 years ago

the workaround for now is export CUDAFLAGS=-arch=sm_80 before cmake.