Thanks to visit codestin.com
Credit goes to github.com

Skip to content

We should include where specialization happens when we throw a constraint violation error #152918

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
Open
bobrenjc93 opened this issue May 6, 2025 · 0 comments
Open

We should include where specialization happens when we throw a constraint violation error #152918

bobrenjc93 opened this issue May 6, 2025 · 0 comments
Assignees
@bobrenjc93
Labels
module: dynamic shapes oncall: pt2 triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module

Comments

@bobrenjc93
Copy link
Contributor

bobrenjc93 commented May 6, 2025 •
edited by pytorch-bot bot
Loading

Suggested by @Chillee

Here's an example repro: https://gist.github.com/bobrenjc93/bdbd0e16bcccb9280ba9113b5f647138

Which prints out this stack trace: https://gist.github.com/bobrenjc93/a9eb9b7a3ab6659e6238869eb72fedfc

But it doesn't point to the user LOC that actually caused the specialization

cc @chauhang @penguinwu @ezyang
@bobrenjc93 bobrenjc93 self-assigned this May 6, 2025
bobrenjc93 added a commit that referenced this issue May 6, 2025
@bobrenjc93
include user stacks with constraint violation error message
eaba9fc 
Fixes #152918

Before:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.
```

After:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.

User stack:
  File "/home/bobren/local/a/pytorch/error.py", line 5, in foo
    return torch.randn(5) * x
```

[ghstack-poisoned]
@bobrenjc93 bobrenjc93 mentioned this issue May 6, 2025
Open
bobrenjc93 added a commit that referenced this issue May 6, 2025
@bobrenjc93
include user stacks with constraint violation error message
bb4131f 
Fixes #152918

Before:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.
```

After:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.

User stack:
  File "/home/bobren/local/a/pytorch/error.py", line 5, in foo
    return torch.randn(5) * x
```

ghstack-source-id: 8008f98
Pull Request resolved: #152924
@drisspg drisspg added triaged This issue has been looked at a team member, and triaged and prioritized into an appropriate module module: dynamic shapes labels May 6, 2025
bobrenjc93 added a commit that referenced this issue May 8, 2025
@bobrenjc93
Update on "include user stacks with constraint violation error message"
4951f50 
Fixes #152918

Before:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.
```

After:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.

User stack:
  File "/home/bobren/local/a/pytorch/error.py", line 5, in foo
    return torch.randn(5) * x
```

cc ezyang SherlockNoMad EikanWang jgong5 wenzhe-nrv

[ghstack-poisoned]
bobrenjc93 added a commit that referenced this issue May 8, 2025
@bobrenjc93
include user stacks with constraint violation error message
1e344d2 
Fixes #152918

Before:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.
```

After:

```
File "/data/users/bobren/a/pytorch/torch/fx/experimental/symbolic_shapes.py", line 5588, in produce_guards_verbose
    raise ConstraintViolationError(
torch.fx.experimental.symbolic_shapes.ConstraintViolationError: Constraints violated (L['x'].size()[0])! For more information, run with TORCH_LOGS="+dynamic".
  - You marked L['x'].size()[0] as dynamic but your code specialized it to be a constant (5). Either remove the mark_dynamic or use a less strict API such as maybe_mark_dynamic or Dim.AUTO.

User stack:
  File "/home/bobren/local/a/pytorch/error.py", line 5, in foo
    return torch.randn(5) * x
```

ghstack-source-id: 0397f89
Pull Request resolved: #152924
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@bobrenjc93 bobrenjc93

Labels
module: dynamic shapes 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
Development

No branches or pull requests
2 participants
@bobrenjc93 @drisspg