fix(ssa): Do not remove array operations that can be OOB during DIE #9232
Merged
TomAFrench merged 6 commits intotf/optimize-array-access-checksfrom Jul 22, 2025
Merged
Conversation
vezenovm
changed the title
Contributor
Changes to Brillig bytecode sizes
🧾 Summary (10% most significant diffs)
Full diff report 👇
|
Contributor
Changes to number of Brillig opcodes executed
🧾 Summary (10% most significant diffs)
Full diff report 👇
|
Contributor
Changes to circuit sizes
🧾 Summary (10% most significant diffs)
Full diff report 👇
|
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
⚠️ Performance Alert ⚠️
Possible performance regression was detected for benchmark 'Opcode count'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.10.
| Benchmark suite | Current: efbe355 | Previous: 39a8379 | Ratio |
|---|---|---|---|
private-kernel-reset |
83598 opcodes |
68865 opcodes |
1.21 |
This comment was automatically generated by workflow using github-action-benchmark.
CC: @TomAFrench
Contributor
There was a problem hiding this comment.
⚠️ Performance Alert ⚠️
Possible performance regression was detected for benchmark 'Execution Time'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.
| Benchmark suite | Current: 4a1d68d | Previous: 3c18e33 | Ratio |
|---|---|---|---|
sha512-100-bytes |
0.115 s |
0.056 s |
2.05 |
This comment was automatically generated by workflow using github-action-benchmark.
CC: @TomAFrench
Contributor
There was a problem hiding this comment.
⚠️ Performance Alert ⚠️
Possible performance regression was detected for benchmark 'Test Suite Duration'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.
| Benchmark suite | Current: 4a1d68d | Previous: 3c18e33 | Ratio |
|---|---|---|---|
test_report_AztecProtocol_aztec-packages_noir-projects_aztec-nr |
132 s |
5 s |
26.40 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-contracts |
151 s |
42 s |
3.60 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_blob |
200 s |
3 s |
66.67 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_private-kernel-lib |
263 s |
1 s |
263 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_reset-kernel-lib |
33 s |
2 s |
16.50 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_rollup-lib |
666 s |
3 s |
222 |
test_report_AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_types |
126 s |
1 s |
126 |
test_report_zkpassport_noir-ecdsa_ |
112 s |
3 s |
37.33 |
This comment was automatically generated by workflow using github-action-benchmark.
CC: @TomAFrench
This was referenced Jul 17, 2025
Co-authored-by: Tom French <15848336+TomAFrench@users.noreply.github.com>
Member
|
I took a look at the cause of the regressions in this PR and it seems to be down to us now having a lot of instructions would have been removed by die which are sitting under a constant zero predicate. It's a bit annoying to deal with these in die (as we'd need to do lookahead to determine the predicate). Chucking a |
Contributor
Author
Which ones do you see are sitting under a constant zero? Some of the circuit regressions look to be unavoidable without extra handling in DIE. We may have to bring back some form of the OOB checks inserted during DIE that were removed in #7995. For example this program taken from pub struct Data {
fields: [Field; 1],
counter: u32,
}
fn main(array: [Data; 1], x: bool) {
let index = if x { 0 } else { 1 };
if index != 0 {
assert(array[index - 1].counter < 3);
}
}Produces this SSA: Previously we would remove |
Member
|
More details in #9275, tl;dr is that this is a no_predicates issue. |
5 tasks
Contributor
Author
|
I'm looking at bring back removing unused array get groups during DIE. |
Contributor
Author
This is showing to be promising. |
Co-authored-by: Maxim Vezenov <mvezenov@gmail.com>
Member
|
huh, sorry I seem to have broken this. |
5 tasks
TomAFrench
merged commit 59fcd88
into
tf/optimize-array-access-checks
118 checks passed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Problem*
Resolves #9223
Summary*
Builds off of #9200
This replaces #9220 which branches off master. The motivation to have these changes branch off one another is that we have several test failures triggerd by #9200 (the original parent PR). We can more easily get a passing CI but having any fixes built off one another.
Additional Context
Documentation*
Check one:
PR Checklist*
cargo fmton default settings.