frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Delete a bunch of type-ignores (#113990)

Browse Source 
* Replaced `ignore[import]` by mypy config file entries
* Removed a bunch of ignores around previously-fixed attr-defined /
  call-arg issues
* Fixed some invalid / undefined types; added a few more type-ignores to
  squelch the downstream errors this exposed

Pull Request resolved: https://github.com/pytorch/pytorch/pull/113990
Approved by: https://github.com/eellison, https://github.com/Skylion007
ghstack dependencies: #113979
This commit is contained in:
Jez Ng
2023-11-17 15:33:34 -08:00
committed by PyTorch MergeBot
parent 47220bc72a
commit 4667e20b3f
24 changed files with 72 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
torch/_inductor/comm_analysis.py
View File

@ -1,12 +1,17 @@
import math
from enum import IntEnum
from typing import TYPE_CHECKING
import torch
from . import ir
from .utils import get_dtype_size, sympy_product
from .virtualized import V
if TYPE_CHECKING:
from torch._inductor.scheduler import BaseSchedulerNode
class NCCL_COLL(IntEnum):
ALL_REDUCE = 0
@ -33,7 +38,7 @@ def get_gpu_type() -> NVIDIA_GPU_TYPE:
return NVIDIA_GPU_TYPE.AMPERE
def get_collective_type(snode: "BaseSchedulerNode") -> NCCL_COLL: # type: ignore[name-defined]
def get_collective_type(snode: "BaseSchedulerNode") -> NCCL_COLL:
if isinstance(snode.node, (ir.AllReduce, ir.AllReduceCoalesced)):
return NCCL_COLL.ALL_REDUCE
elif isinstance(
@ -136,7 +141,7 @@ llMaxBws = torch.tensor(
)
def estimate_nccl_collective_runtime(snode: "BaseSchedulerNode") -> float: # type: ignore[name-defined]
def estimate_nccl_collective_runtime(snode: "BaseSchedulerNode") -> float:
"""
Returns estimated NCCL collective runtime in nanoseconds (ns).
@ -158,7 +163,7 @@ def estimate_nccl_collective_runtime(snode: "BaseSchedulerNode") -> float: # ty
# Currently assumes each node has 8 gpus. And when >1 node is used, assumes each node uses all 8 gpus.
# TODO: Need to find a way to get accurate "gpus per node" and "# nodes" info.
num_gpus_per_node = 8
_, _, group_size = snode.node.constant_args
_, _, group_size = snode.node.constant_args # type: ignore[attr-defined]
nNodes = math.ceil(group_size / num_gpus_per_node)
nRanks = group_size # this is total # of gpus globally that participate in this collective op