trunk/6fb6ef21931fa1be2b171f5f5542730663afcf86: [FSDP2] Join pending all-gather with device stream (#199677)

发生了什么

FSDPParamGroup.finalize_backward may clean up an all-gather launched by backward prefetch but never consumed. It recorded a device-specific event and waited through torch.accelerator.current_stream(), whose generic Stream.wait_event does not handle that device-specific event correctly during CUDA graph capture. The all-gather stream remained unjoined, so capture failed with cudaErrorStreamCaptureUnjoined. Use FSDP d…

摘要按规则整理自下方来源原文

来源