Using torch expandable segments with megatron breaks vLLM refit on Ampere. #522
Description
Describe the bug
Running the refit_policy_generation operation with a Megatron policy that is running torch with 'expandable_segments:true' causes the colocated vLLM refit case to error out with:
(VllmGenerationWorker pid=36530) Error in VllmInternalWorkerExtension.update_weights_from_ipc_handles: pidfd_getfd: Operation not permitted
on Ampere. Seems to work fine on Hopper (Blackwell TBD).