chore: Allow control dependent LICM in ACIR

[vezenovm]

Description

Problem*

No issue. I just wanted to test the effects of activating control dependent on Brillig only and then also ACIR.

Summary*

Similar to #6785 I am expecting some kind of improvement in ACIR compilation. I expect an improvement as we will be potentially hoisting more instructions out of loops, thus reducing the work of loop unrolling which is required in the ACIR runtime. We do not expect any gate count improvements as ACIR ultimately unrolls all loop. The only real benefit is potentially unrolling less instructions.

The initial implementation of #7660 was quite inefficient and led to large compilation time regressions. After switching to the usage of post-dominance frontiers there are no longer an compilation regression in #7660 so I want to see the benefits we get from activating control dependent LICM on ACIR as well.

Additional Context

Good memory reductions on the rollup circuits:

rollup-block-root-single-tx 7190 MB 7860 MB 0.91
rollup-block-root 7190 MB 7870 MB 0.91

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.

[github-actions]

Execution Time

Details

Benchmark suite	Current: d6a3bf7	Previous: 58be0cd	Ratio
private-kernel-inner	0.069 s	0.069 s	1
private-kernel-reset	0.287 s	0.283 s	1.01
private-kernel-tail	0.026 s	0.026 s	1
rollup-base-private	0.737 s	0.73 s	1.01
rollup-base-public	0.487 s	0.487 s	1
rollup-block-root	16.7 s	16.7 s	1
rollup-merge	0.006 s	0.006 s	1
rollup-root	0.025 s	0.025 s	1

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

[github-actions]

Compilation Time

Details

Benchmark suite	Current: d6a3bf7	Previous: 58be0cd	Ratio
regression_4709	0.729 s	0.67 s	1.09
ram_blowup_regression	14.9 s	14.6 s	1.02
global_var_regression_entry_points	0.5 s	0.484 s	1.03
private-kernel-inner	2.278 s	2.388 s	0.95
private-kernel-reset	6.85 s	6.782 s	1.01
private-kernel-tail	1.102 s	1.104 s	1.00
rollup-base-private	16.86 s	15.2 s	1.11
rollup-base-public	12.26 s	11.34 s	1.08
rollup-block-root-empty	0.963 s	0.933 s	1.03
rollup-block-root-single-tx	124 s	129 s	0.96
rollup-block-root	126 s	133 s	0.95
rollup-merge	0.961 s	0.928 s	1.04
rollup-root	1.558 s	1.57 s	0.99

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

[github-actions]

Test Suite Duration

Details

Benchmark suite	Current: d6a3bf7	Previous: 58be0cd	Ratio
AztecProtocol_aztec-packages_noir-projects_aztec-nr	37 s	37 s	1
AztecProtocol_aztec-packages_noir-projects_noir-contracts	83 s	78 s	1.06
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_blob	49 s	44 s	1.11
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_private-kernel-lib	185 s	170 s	1.09
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_reset-kernel-lib	10 s	9 s	1.11
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_rollup-lib	159 s	168 s	0.95
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_types	53 s	54 s	0.98
noir-lang_noir-bignum_	80 s	86 s	0.93
noir-lang_noir_bigcurve_	215 s	209 s	1.03
noir-lang_noir_json_parser_	9 s	8 s	1.13
noir-lang_sha512_	25 s	23 s	1.09

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

[github-actions]

Compilation Memory

Details

Benchmark suite	Current: d6a3bf7	Previous: 58be0cd	Ratio
private-kernel-inner	301.67 MB	301.63 MB	1.00
private-kernel-reset	611.54 MB	611.54 MB	1
private-kernel-tail	228.23 MB	228.23 MB	1
rollup-base-private	1250 MB	1250 MB	1
rollup-base-public	1330 MB	1330 MB	1
rollup-block-root-empty	305.08 MB	305.11 MB	1.00
rollup-block-root-single-tx	7860 MB	7860 MB	1
rollup-block-root	7870 MB	7870 MB	1
rollup-merge	303.51 MB	303.53 MB	1.00
rollup-root	351.06 MB	351.05 MB	1.00

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

[github-actions]

Execution Memory

Details

Benchmark suite	Current: d6a3bf7	Previous: 58be0cd	Ratio
private-kernel-inner	241.44 MB	241.44 MB	1
private-kernel-reset	275.87 MB	275.87 MB	1
private-kernel-tail	215.09 MB	215.09 MB	1
rollup-base-private	509.27 MB	509.27 MB	1
rollup-base-public	418.07 MB	418.07 MB	1
rollup-block-root	1420 MB	1420 MB	1
rollup-merge	289.01 MB	289 MB	1.00
rollup-root	295.48 MB	295.48 MB	1

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

[vezenovm]

rollup-block-root-single-tx 122 s 126 s 0.97
rollup-block-root 134 s 125 s 1.07

So compilation time seems to grow slightly for rollup-block-root

We look to drop by 680 MB in peak memory.

rollup-block-root-single-tx 7190 MB 7860 MB 0.91
rollup-block-root 7190 MB 7870 MB 0.91

I have not profiled memory consumption on these programs yet, but it is mostly likely due to the peak memory usage occurring during loop unrolling and LICM reducing the number of instructions we need to unroll.

EDIT: It looks like we no longer have much of a compilation time regression (and in the case of rollup-block-root an improvement). So I am also marking this ready for review.

rollup-block-root-single-tx 127 s 121 s 1.05
rollup-block-root 125 s 134 s 0.93

[vezenovm]

rollup-base-private 16.54 s 15.36 s 1.08
rollup-base-public 12.2 s 11.02 s 1.11

That main regression is in the rollup base circuits.