8000 Decorators like `torch.compiler.allow_in_graph` doesn't account for id reuse · Issue #147777 · pytorch/pytorch · GitHub
[go: up one dir, main page]
Skip to content

Decorators like torch.compiler.allow_in_graph doesn't account for id reuse #147777

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Closed
StrongerXi opened this issue Feb 24, 2025 · 0 comments
Closed

Decorators like torch.compiler.allow_in_graph doesn't account for id reuse #147777

StrongerXi opened this issue Feb 24, 2025 · 0 comments
Assignees
@StrongerXi
Labels
module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Comments

@StrongerXi
Copy link
Contributor
StrongerXi commented Feb 24, 2025 •
edited by pytorch-bot bot
Loading

🐛 Describe the bug

Context: https://github.com/pytorch/pytorch/pull/146367/files#r1964644166

Repro:

import torch

@torch.compiler.allow_in_graph
def f(x):
    return x + 1
del f

def g(x):
    return x + 2

@torch.compile(fullgraph=True, backend="eager")
def fn(x):
    return g(x)

fn(torch.ones(1))

Run it with TORCH_LOGS="graph_code":

output_graph.py:1385] [0/0] [__graph_code] TRACED GRAPH
output_graph.py:1385] [0/0] [__graph_code]  ===== __compiled_fn_1 =====
output_graph.py:1385] [0/0] [__graph_code]  /Users/ryanguo99/Documents/work/pytorch/torch/fx/_lazy_graph_module.py class GraphModule(torch.nn.Module):
output_graph.py:1385] [0/0] [__graph_code]     def forward(self, L_x_: "f32[1][1]cpu"):
output_graph.py:1385] [0/0] [__graph_code]         l_x_ = L_x_
output_graph.py:1385] [0/0] [__graph_code]
output_graph.py:1385] [0/0] [__graph_code]          # File: /Users/ryanguo99/Documents/work/scratch/allow-in-graph.py:14 in fn, code: return g(x)
output_graph.py:1385] [0/0] [__graph_code]         g: "f32[1][1]cpu" = __main___g(l_x_);  l_x_ = None
output_graph.py:1385] [0/0] [__graph_code]         return (g,)

Commenting out del f and rerun:

output_graph.py:1385] [0/0] [__graph_code] TRACED GRAPH
output_graph.py:1385] [0/0] [__graph_code]  ===== __compiled_fn_1 =====
output_graph.py:1385] [0/0] [__graph_code]  /Users/ryanguo99/Documents/work/pytorch/torch/fx/_lazy_graph_module.py class GraphModule(torch.nn.Module):
output_graph.py:1385] [0/0] [__graph_code]     def forward(self, L_x_: "f32[1][1]cpu"):
output_graph.py:1385] [0/0] [__graph_code]         l_x_ = L_x_
output_graph.py:1385] [0/0] [__graph_code]
output_graph.py:1385] [0/0] [__graph_code]          # File: /Users/ryanguo99/Documents/work/scratch/allow-in-graph.py:9 in g, code: return x + 2
output_graph.py:1385] [0/0] [__graph_code]         add: "f32[1][1]cpu" = l_x_ + 2;  l_x_ = None
output_graph.py:1385] [0/0] [__graph_code]         return (add,)

Error logs

No response

Versions

main

cc @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @chenyang78 @kadeng @amjames
@StrongerXi StrongerXi mentioned this issue Feb 24, 2025
Closed
@StrongerXi StrongerXi self-assigned this Feb 24, 2025
@desertfire desertfire added the triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module label Feb 25, 2025
@StrongerXi StrongerXi mentioned this issue Mar 4, 2025
Closed
@StrongerXi StrongerXi linked a pull request Mar 4, 2025 that will close this issue
[dynamo] Account for function id reuse in relevant Dynamo decorators #148385
Closed
pytorchmergebot pushed a commit that referenced this issue Mar 5, 2025
@StrongerXi @pytorchmergebot
[dynamo] Account for function id reuse in relevant Dynamo decorators
cb317a4 
This fixes a recent series of flaky failure from `nonstrict_trace` unit
tests: #148166, #148056, #148055, #148054, #148034, #148033, #148032, #148031.

For now we don't need to worry about the other decorators because they
are either meant for builtin/numpy functions (which should never
deallocate in practice), or used for polyfills which keeps the function
object in `get_torch_obj_rule_map()`.

Fixes #147777.

ghstack-source-id: d9bea5f
Pull Request resolved: #148385
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@StrongerXi StrongerXi

Labels
module: dynamo oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module
Projects
None yet
Milestone
No milestone
38F8
Development

Successfully merging a pull request may close this issue.

[dynamo] Account for function id reuse in relevant Dynamo decorators pytorch/pytorch
2 participants
@desertfire @StrongerXi
0