• G
    Add enable_partial_send_recv switch in pipeline_configs (#46992) · b9a2f29c
    Ghost Screaming 提交于
    * Fix bug of reduce_sum op. When input.numel() > INT32_MAX, its result
    is wrong.
    
    * Support allow_partial switch, which can be configure in
    pipeline_configs. If sent tensor are not the same from
    different hosts, they shouldn't been sent partially and
    then concated as a whole tensor.
    
    * Change name allow_partial to enable_partial_send_recv.
    
    * Add global variable _enable_partial_send_recv
    b9a2f29c
pipeline_parallel.py 29.0 KB