GitLab

For the SMs which calculate metadata in notify_dispatch, each warp in the SM is used to calculate the metadata of one channel. The default configuration is 8 warps for 10 channels, which needs two rounds of loop. Maybe the number of warps can be configured to the number of the channels so that one loop is enough.

Super tiny fix typo

Support zero-copy for low-latency combine

Low latency kernels use rdma atomic to support AR

Allow passing output tensor in low_latency_combine

Update NVSHMEM to v3.2.5.

Fix AR bugs for normal kernels

Update path

fix installation