feat(perf): Cache commutative instructions in constant folding

[vezenovm]

Description

Problem*

Resolves

Summary*

Keeping a draft as I want to see how it runs in the code size benchmarks

Additional Context

Documentation*

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist*

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[jfecher]

Instead of this hard check here, we should sort the operands when creating a commutative binary op. E.g. lower ValueIds are always the lhs.

[vezenovm]

Done. However, we seem to be getting similar results as before.

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: 6605711922cbb8ee068b8cdbbd9b00c869b47fcf, compared to commit: 3c778b73d9458ab708df21c850468d708676cde4

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
binary_operator_overloading	-2 ✅	-0.44%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
binary_operator_overloading	450 (-2)	-0.44%

[github-actions]

Changes to circuit sizes

Generated at commit: 6605711922cbb8ee068b8cdbbd9b00c869b47fcf, compared to commit: 3c778b73d9458ab708df21c850468d708676cde4

🧾 Summary (10% most significant diffs)

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
regression_5252	+72 ❌	+0.09%	+72 ❌	+0.07%
slices	+1 ❌	+0.32%	+1 ❌	+0.03%

---

Full diff report 👇

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
regression_5252	77,070 (+72)	+0.09%	97,311 (+72)	+0.07%
slices	312 (+1)	+0.32%	3,381 (+1)	+0.03%
hashmap	200,183 (+10)	+0.00%	361,213 (+10)	+0.00%
conditional_1	4,784 (-3)	-0.06%	38,799 (0)	0.00%
merkle_insert	1,780 (+1)	+0.06%	29,032 (0)	0.00%
slice_dynamic_index	1,201 (+1)	+0.08%	6,420 (0)	0.00%

[jfecher]

I don't understand the regression_5252 size increase

[vezenovm]

I don't understand the regression_5252 size increase

Yeah I don't either. I guess we can close this PR as it doesn't seem to be providing many benefits.

[jfecher]

Oh, I meant sorting when the instructions are first created. Not during constant folding. Probably won't remove the size increase in the tests though.

[vezenovm]

Going to close this PR as the benefits seem minimal in Brillig code size and actually cause regressions for ACIR.