Segmentation Fault While Running CUDA or OpenACC Applications With UCX Support

Owned by Cristian Di Pietrantonio
Jun 14, 2022
1 min read

Problem

Segmentation fault can occur if applications are either statically linked to CUDA libraries or memory allocated before MPI_Init.

Solution

As a workaround, disable memory type cache by exporting:

UCX_MEMTYPE_CACHE=n


Filter by label

There are no items with the selected labels at this time.