fix: Revert "feat(mem2reg): address last known value is independent of its…

[asterite]

… aliases (#9613)"

This reverts commit 92aa75d.

Description

Problem

Unblocks the Noir sync on Aztec-Packages

Summary

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]

@asterite if you aren't already working on it, can you make sure we have a regression for this so it is caught in the future?

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: 888deeb48bf5f37e0946c2734d7f3cf1bdbf4a1f, compared to commit: 22ceb270944bf6688828592d845d49527609b3eb

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
regression_9303_inliner_max	+1 ❌	+0.95%
regression_9303_inliner_min	+1 ❌	+0.95%
regression_9303_inliner_zero	+1 ❌	+0.95%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
regression_9303_inliner_max	106 (+1)	+0.95%
regression_9303_inliner_min	106 (+1)	+0.95%
regression_9303_inliner_zero	106 (+1)	+0.95%
reference_counts_slices_inliner_0_inliner_min	1,212 (+11)	+0.92%
reference_counts_slices_inliner_0_inliner_zero	1,001 (+7)	+0.70%
regression_6674_2_inliner_min	244 (+1)	+0.41%
reference_counts_inliner_max_inliner_zero	893 (+3)	+0.34%
reference_counts_inliner_0_inliner_zero	893 (+3)	+0.34%
reference_counts_inliner_min_inliner_zero	893 (+3)	+0.34%
regression_6674_3_inliner_min	674 (+2)	+0.30%
reference_counts_inliner_0_inliner_min	1,044 (+3)	+0.29%
reference_counts_inliner_min_inliner_min	1,044 (+3)	+0.29%
reference_counts_inliner_max_inliner_min	1,044 (+3)	+0.29%
uhashmap_inliner_zero	6,736 (+3)	+0.04%
uhashmap_inliner_min	7,150 (+3)	+0.04%
hashmap_inliner_zero	7,835 (+1)	+0.01%
hashmap_inliner_min	8,873 (+1)	+0.01%

[github-actions]

Changes to number of Brillig opcodes executed

Generated at commit: 888deeb48bf5f37e0946c2734d7f3cf1bdbf4a1f, compared to commit: 22ceb270944bf6688828592d845d49527609b3eb

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
regression_9303_inliner_max	+2 ❌	+1.26%
regression_9303_inliner_min	+2 ❌	+1.26%
regression_9303_inliner_zero	+2 ❌	+1.26%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
regression_9303_inliner_max	161 (+2)	+1.26%
regression_9303_inliner_min	161 (+2)	+1.26%
regression_9303_inliner_zero	161 (+2)	+1.26%
regression_6674_3_inliner_min	1,533 (+2)	+0.13%
regression_6674_2_inliner_min	864 (+1)	+0.12%
hashmap_inliner_zero	72,712 (+57)	+0.08%
hashmap_inliner_min	79,587 (+57)	+0.07%
uhashmap_inliner_zero	166,316 (+97)	+0.06%
uhashmap_inliner_min	174,205 (+97)	+0.06%

[asterite]

@jfecher I'm working on it. I see that the PR caused a difference in SSA but it's a bit SSA (luckily not that big, around 90K and only a few changes here and there) so I need to check if those differences are okay or if there's something wrong, then I could create a regression test.

[github-actions]

⚠️ 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: e24be34	Previous: 22ceb27	Ratio
test_report_noir-lang_noir-bignum_	504 s	379 s	1.33

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench

[jfecher]

@asterite sounds good - just wanted to make sure this isn't lost particularly since we're continuing to work on more mem2reg changes