Questions About AR Support for Low-latency

[michaelchen1996]

Commit: 2d0cf41dd1900b105d74cb071f4cac35e3fb6f47​
Low latency kernels use rdma atomic to support AR.​

This commit replaces two functions:​

• nvshmemi_ibgda_rma_p​
• nvshmemi_ibgda_poll_recv​

with a new function: nvshmemi_ibgda_amo_nonfetch_add
​
From code inspection, nvshmemi_ibgda_rma_p appears to use RDMA Write with Immediate Data.​
Does RDMA Write with IMM not support AR? Why we have to use nvshmemi_ibgda_amo_nonfetch_add to add AR support for low-latency ?​
Thank you!

[michaelchen1996]

Additionally, may I ask that if this is because NVSHMEM does not have a write with imm operation, why is nvshmemi_ibgda_rma_p with the imm parameter still retained?