Nargo panics if you specify an expression-width < 3 during compile #5560
Assignees
Labels
bug
Something isn't working
Comments
5 tasks
github-merge-queue bot
pushed a commit
that referenced
this issue
Oct 17, 2024
# Description ## Problem\* Resolves #5560 ## Summary\* Rejects an invalid `--expression-value` in the CLI, which would later cause a panic in `acvm::compiler::transformers::csat`. An example error message looks like: ```text error: invalid value '1' for '--expression-width <EXPRESSION_WIDTH>': minimum value is 3 ``` ## Additional Context The issue suggests that [CSatTransformer::new](https://github.com/noir-lang/noir/blob/ae87d287ab1fae0f999dfd0d1166fbddb927ba97/acvm-repo/acvm/src/compiler/transformers/csat.rs#L24-L27) should return a `Result` explaining the problem it has with `width`, rather than doing an `assert!` (without any message) and crashing the program. I agree, however looking at the chain of calls, none of the half-dozen functions we go through to reach this use `Result`, and the `CSatTransformer` is the only option we have. This suggests to me that this limitation is perhaps supposed to be well-known to the user, ie. it's not the case that one transformer has X limit and another has Y limit. For this reason I added a `pub const MIN_EXPRESSION_WIDTH: usize = 3;` to the `acvm::compiler` module and using that as a common value for the assertion as well as the validation in the CLI. Should the assumption of a single global value change, removing that will force us to update the validation logic as well. That said if you prefer going the `Result` route I'm not against it, it just seemed like an overkill for this single use case. ## Documentation\* Check one: - [x] No documentation needed. - [ ] Documentation included in this PR. - [ ] **[For Experimental Features]** Documentation to be submitted in a separate PR. # PR Checklist\* - [x] I have tested the changes locally. - [ ] I have formatted the changes with [Prettier](https://prettier.io/) and/or `cargo fmt` on default settings.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Aim
When compiling a noir program and specifying the expression-width (amount of witnesses in an expression), the compiler expects a width > 2. This is most likely the correct behaviour, but nargo panics with the following message:
Expected Behavior
The program should terminate and tell me that nargo expects an expression-width > 2, instead of a panic and indicating that there is a bug.
Bug
The problem is here:
Here the CSat Solver asserts that the expression-width is > 2. It should return an Result instead telling the user to provide an expression-width > 2
To Reproduce
nargo new wrong_expr_width && cd wrong_expr_widthnargo compile --expression-width 1Project Impact
None
Impact Context
No response
Workaround
None
Workaround Description
No response
Additional Context
No response
Installation Method
Binary (
noirupdefault)Nargo Version
noirc version = 0.30.0+af57471035e4fa7eaffa71693219df6d029dbcde
NoirJS Version
No response
Would you like to submit a PR for this Issue?
Maybe
Support Needs
No response
The text was updated successfully, but these errors were encountered: