Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to use the UCX with RCCL and rccl-rdma-sharp-plugins? #29

Open
shanleo2024 opened this issue Nov 7, 2023 · 0 comments
Open

How to use the UCX with RCCL and rccl-rdma-sharp-plugins? #29

shanleo2024 opened this issue Nov 7, 2023 · 0 comments

Comments

@shanleo2024
Copy link

Hi, I just try to use the UCX with RCCL and rccl-rdma-sharp-plugins, but unfortunately they cannot work, as there always an error like that:
ucp_mm.c:855 Assertion `memh->md_map != 0' failed

And I noticed that the function nccl_ucx_regmr in rccl-rdma-sharp-plugins, only input those params:
mmap_params.field_mask = UCP_MEM_MAP_PARAM_FIELD_ADDRESS |
UCP_MEM_MAP_PARAM_FIELD_LENGTH;
mmap_params.address = (void*)reg_addr;
mmap_params.length = reg_size;
mh->mem_type = (type == NCCL_PTR_HOST)? UCS_MEMORY_TYPE_HOST: UCS_MEMORY_TYPE_CUDA;
mmap_params.field_mask |= UCP_MEM_MAP_PARAM_FIELD_MEMORY_TYPE;
mmap_params.memory_type = mh->mem_type;

And then calll the ucp_mem_map, as there is no flag to input, the ucp_mem_map will report this error.
So could someone who is so kind can help me to anslysis how to use the API correctlly?
Thanks a lot!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant