ColossalAI/colossalai/gemini/tensor_utils.py

119 lines
3.9 KiB
Python
Raw Normal View History

2022-04-01 01:22:33 +00:00
import torch
from colossalai.gemini.stateful_tensor import StatefulTensor
2022-04-01 01:22:33 +00:00
from typing import Union, Tuple
def is_storage_empty(tensor: torch.Tensor) -> bool:
return tensor.storage().size() == 0
def free_storage(tensor: torch.Tensor) -> None:
if not is_storage_empty(tensor):
tensor.storage().resize_(0)
def alloc_storage(tensor: torch.Tensor) -> None:
if is_storage_empty(tensor):
tensor.storage().resize_(tensor.numel())
2022-04-01 01:22:33 +00:00
def colo_tensor_mem_usage(tensor: Union[torch.Tensor, StatefulTensor]) -> Tuple[int, int]:
if isinstance(tensor, StatefulTensor):
2022-04-01 01:22:33 +00:00
t = tensor.payload
elif isinstance(tensor, torch.Tensor):
t = tensor
else:
return 0, 0
cuda_use, cpu_use = 0, 0
mem_use = t.storage().size() * t.element_size()
2022-04-01 01:22:33 +00:00
if t.device.type == 'cuda':
cuda_use += mem_use
elif t.device.type == 'cpu':
cpu_use += mem_use
return cuda_use, cpu_use
def colo_model_data_tensor_move(src_t: Union[StatefulTensor, torch.Tensor], tgt_t: Union[StatefulTensor,
torch.Tensor]) -> None:
"""
A colossal API for model data tensor move.
2022-04-01 01:22:33 +00:00
The src and target tensors could be resident on both CPU and GPU.
2022-04-01 01:22:33 +00:00
NOTE() The source tensor payload will be removed after this function.
2022-04-01 01:22:33 +00:00
The function will record the communication volume between CPU and GPU.
Args:
src_t (Union[StatefulTensor, torch.Tensor]): source tensor
2022-04-01 01:22:33 +00:00
tgt_t (Union[StatefulTensor, torch.Tensor]): target tensor
"""
if isinstance(src_t, StatefulTensor):
2022-04-01 01:22:33 +00:00
src_t_payload = src_t.payload
else:
src_t_payload = src_t.data
src_dev = src_t_payload.device
if isinstance(tgt_t, StatefulTensor):
2022-04-01 01:22:33 +00:00
tgt_t_payload = tgt_t.payload
else:
tgt_t_payload = tgt_t.data
tgt_t_payload.copy_(src_t_payload)
# remove payload of src_t
if isinstance(src_t, StatefulTensor):
src_t.set_null()
2022-04-01 01:22:33 +00:00
else:
src_t.data = torch.empty(0, device=src_dev, dtype=src_t_payload.dtype)
2022-04-01 01:22:33 +00:00
def colo_model_data_tensor_move_inline(t: Union[StatefulTensor, torch.Tensor], target_device: Union[torch.device,
int]) -> None:
"""
2022-04-01 01:22:33 +00:00
move a tensor to the target_device
Args:
t (Union[StatefulTensor, torch.Tensor]): the tensor be moved
target_device: a traget device, if type is int, it the index of cuda card.
"""
if not isinstance(target_device, torch.device):
target_device = torch.device(f'cuda:{target_device}')
if isinstance(t, torch.Tensor):
t.data = t.data.to(target_device)
elif isinstance(t, StatefulTensor):
t.move_to(target_device)
else:
raise TypeError(f'colo_model_data_tensor_move_inline dose not accept type {type(t)}')
2022-04-01 01:22:33 +00:00
def colo_model_data_move_to_cpu(t: Union[StatefulTensor, torch.Tensor]) -> None:
"""colo_model_data_move_to_cpu
2022-04-01 01:22:33 +00:00
move a model data tensor from gpu to cpu
Args:
t (Union[StatefulTensor, torch.Tensor]): _description_
"""
# TODO() optimize the tensor moving with non-blocking
if isinstance(t, torch.Tensor):
t.data = t.data.cpu()
elif isinstance(t, StatefulTensor):
t.move_to(torch.device('cpu'))
else:
raise TypeError(f'colo_model_data_move_to_cpu dose not accept type {type(t)}')
2022-04-01 01:22:33 +00:00
def colo_model_tensor_clone(t: Union[StatefulTensor, torch.Tensor], target_device: torch.device) -> torch.Tensor:
"""
Clone a model data tensor
Args:
t (Union[StatefulTensor, torch.Tensor]): a model data tensor
target_device (torch.device): the target device
Returns:
torch.Tensor: a cloned torch tensor
"""
# TODO() rename this function
colo_model_data_tensor_move_inline(t, target_device)
t_payload = t.payload if isinstance(t, StatefulTensor) else t
return t_payload