Add IPC domain rank detection (#824)

Expose the number of ranks in each GPU IPC domain through Bootstrap and
Python bindings, using NVML fabric information on CUDA when available
and falling back to host-local ranks otherwise.

---------

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This commit is contained in:
Binyang Li
2026-06-30 09:47:14 -07:00
committed by GitHub
parent 9c6b28337d
commit 197d62c4b9
9 changed files with 148 additions and 0 deletions

View File

@@ -56,6 +56,7 @@ void register_core(nb::module_& m) {
.def("get_rank", &Bootstrap::getRank)
.def("get_n_ranks", &Bootstrap::getNranks)
.def("get_n_ranks_per_node", &Bootstrap::getNranksPerNode)
.def("get_n_ranks_per_ipc_domain", &Bootstrap::getNranksPerIpcDomain)
.def(
"send",
[](Bootstrap* self, uintptr_t ptr, size_t size, int peer, int tag) {