chore: PR 6782 w/ PR 6783

[vezenovm]

Description

Problem*

Resolves

Summary*

Pushed to see the bench report of #6782 with the #6783 optimization.

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.

[github-actions]

Changes to Brillig bytecode sizes

Generated at commit: 5f8cb04e452c26fe2fcdeed400fe9ab58ce0515a, compared to commit: 1b0dd4149d9249f0ea4fb5e2228c688e0135618f

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
to_le_bytes	-12 ✅	-8.96%
slices	-216 ✅	-9.11%
wildcard_type	-33 ✅	-10.06%
nested_array_dynamic	-324 ✅	-11.66%
reference_counts	-63 ✅	-11.89%
nested_array_in_slice	-189 ✅	-13.30%
slice_regex	-396 ✅	-15.06%
nested_array_dynamic_simple	-30 ✅	-22.39%
fold_complex_outputs	-153 ✅	-24.36%
brillig_nested_arrays	-198 ✅	-52.11%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
array_dedup_regression	259 (+3)	+1.17%
regression_4709	133,737 (-3)	-0.00%
modulus	1,762 (-6)	-0.34%
fold_numeric_generic_poseidon	754 (-3)	-0.40%
no_predicates_numeric_generic_poseidon	754 (-3)	-0.40%
fold_2_to_17	578 (-3)	-0.52%
debug_logs	5,124 (-36)	-0.70%
slice_dynamic_index	2,504 (-18)	-0.71%
pedersen_hash	367 (-3)	-0.81%
hashmap	21,693 (-206)	-0.94%
check_large_field_bits	292 (-3)	-1.02%
array_dynamic_blackbox_input	1,032 (-12)	-1.15%
to_be_bytes	209 (-3)	-1.42%
pedersen_commitment	206 (-3)	-1.44%
sha2_byte	2,743 (-42)	-1.51%
regression	939 (-15)	-1.57%
brillig_oracle	364 (-6)	-1.62%
simple_shield	899 (-15)	-1.64%
poseidon_bn254_hash_width_3	5,413 (-92)	-1.67%
poseidon_bn254_hash	5,413 (-92)	-1.67%
nested_dyn_array_regression_5782	172 (-3)	-1.71%
hint_black_box	336 (-6)	-1.75%
keccak256	2,158 (-39)	-1.78%
regression_4449	748 (-15)	-1.97%
7	144 (-3)	-2.04%
blake3	144 (-3)	-2.04%
uhashmap	13,887 (-291)	-2.05%
7_function	570 (-12)	-2.06%
hash_to_field	139 (-3)	-2.11%
aes128_encrypt	543 (-12)	-2.16%
u128	2,754 (-63)	-2.24%
6_array	392 (-9)	-2.24%
sha256_regression	6,851 (-162)	-2.31%
conditional_regression_661	124 (-3)	-2.36%
array_to_slice	715 (-18)	-2.46%
poseidonsponge_x5_254	4,245 (-108)	-2.48%
conditional_1	1,176 (-30)	-2.49%
ram_blowup_regression	957 (-27)	-2.74%
brillig_cow_regression	2,143 (-63)	-2.86%
regression_5252	4,591 (-135)	-2.86%
strings	914 (-27)	-2.87%
if_else_chain	101 (-3)	-2.88%
ecdsa_secp256k1	899 (-27)	-2.92%
sha256_var_padding_regression	5,057 (-153)	-2.94%
higher_order_functions	674 (-21)	-3.02%
pedersen_check	556 (-18)	-3.14%
brillig_pedersen	556 (-18)	-3.14%
brillig_constant_reference_regression	84 (-3)	-3.45%
conditional_regression_421	83 (-3)	-3.49%
sha256_var_witness_const_regression	1,287 (-51)	-3.81%
brillig_cow	372 (-15)	-3.88%
array_dynamic_nested_blackbox_input	892 (-36)	-3.88%
simple_radix	74 (-3)	-3.90%
conditional_regression_short_circuit	1,246 (-54)	-4.15%
sha256_brillig_performance_regression	1,650 (-72)	-4.18%
slice_loop	266 (-12)	-4.32%
6	1,173 (-54)	-4.40%
side_effects_constrain_array	129 (-6)	-4.44%
sha256	2,369 (-111)	-4.48%
bigint	2,165 (-102)	-4.50%
regression_6674_1	244 (-12)	-4.69%
global_consts	227 (-12)	-5.02%
brillig_rc_regression_6123	170 (-9)	-5.03%
merkle_insert	765 (-43)	-5.32%
brillig_uninitialized_arrays	51 (-3)	-5.56%
to_bits	201 (-12)	-5.63%
regression_6674_2	247 (-15)	-5.73%
regression_struct_array_conditional	570 (-36)	-5.94%
to_bytes_integration	88 (-6)	-6.38%
regression_5045	87 (-6)	-6.45%
databus_composite_calldata	390 (-30)	-7.14%
struct_inputs	258 (-21)	-7.53%
simple_2d_array	106 (-9)	-7.83%
sha256_var_size_regression	1,720 (-148)	-7.92%
to_bytes_consistent	135 (-12)	-8.16%
regression_6674_3	552 (-51)	-8.46%
to_le_bytes	122 (-12)	-8.96%
slices	2,156 (-216)	-9.11%
wildcard_type	295 (-33)	-10.06%
nested_array_dynamic	2,454 (-324)	-11.66%
reference_counts	467 (-63)	-11.89%
nested_array_in_slice	1,232 (-189)	-13.30%
slice_regex	2,233 (-396)	-15.06%
nested_array_dynamic_simple	104 (-30)	-22.39%
fold_complex_outputs	475 (-153)	-24.36%
brillig_nested_arrays	182 (-198)	-52.11%

[github-actions]

Changes to number of Brillig opcodes executed

Generated at commit: 5f8cb04e452c26fe2fcdeed400fe9ab58ce0515a, compared to commit: 1b0dd4149d9249f0ea4fb5e2228c688e0135618f

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
merkle_insert	+574 ❌	+17.97%
simple_shield	+316 ❌	+13.17%
wildcard_type	-51 ✅	-9.77%
slice_regex	-387 ✅	-10.02%
nested_array_in_slice	-177 ✅	-10.17%
reference_counts	-60 ✅	-13.13%
fold_complex_outputs	-150 ✅	-16.45%
nested_array_dynamic_simple	-30 ✅	-25.00%
array_dedup_regression	-413 ✅	-37.89%
brillig_nested_arrays	-198 ✅	-56.09%

---

Full diff report 👇

Program	Brillig opcodes (+/-)	%
merkle_insert	3,769 (+574)	+17.97%
simple_shield	2,716 (+316)	+13.17%
array_dynamic_blackbox_input	18,209 (+297)	+1.66%
sha2_byte	45,877 (+341)	+0.75%
sha256_brillig_performance_regression	22,814 (+68)	+0.30%
sha256_var_padding_regression	205,422 (+571)	+0.28%
ram_blowup_regression	721,038 (+815)	+0.11%
brillig_cow_regression	479,934 (+497)	+0.10%
fold_2_to_17	1,093,919 (-3)	-0.00%
modulus	19,172 (-6)	-0.03%
check_large_field_bits	2,740 (-3)	-0.11%
to_be_bytes	2,216 (-3)	-0.14%
brillig_oracle	2,433 (-6)	-0.25%
u128	24,831 (-63)	-0.25%
7	1,106 (-3)	-0.27%
blake3	1,106 (-3)	-0.27%
poseidonsponge_x5_254	183,645 (-501)	-0.27%
aes128_encrypt	4,326 (-12)	-0.28%
to_bytes_integration	2,065 (-6)	-0.29%
sha256_regression	116,817 (-363)	-0.31%
hash_to_field	908 (-3)	-0.33%
keccak256	32,474 (-111)	-0.34%
slice_dynamic_index	4,664 (-18)	-0.38%
7_function	2,333 (-12)	-0.51%
to_le_bytes	1,152 (-6)	-0.52%
regression	2,810 (-15)	-0.53%
regression_5252	914,986 (-5,028)	-0.55%
pedersen_hash	526 (-3)	-0.57%
global_consts	1,773 (-12)	-0.67%
debug_logs	5,136 (-36)	-0.70%
hint_black_box	717 (-6)	-0.83%
array_to_slice	1,667 (-18)	-1.07%
pedersen_commitment	246 (-3)	-1.20%
6_array	1,631 (-21)	-1.27%
brillig_cow	1,140 (-15)	-1.30%
fold_numeric_generic_poseidon	4,941 (-66)	-1.32%
no_predicates_numeric_generic_poseidon	4,941 (-66)	-1.32%
regression_6674_1	816 (-12)	-1.45%
strings	1,767 (-27)	-1.51%
uhashmap	144,661 (-2,287)	-1.56%
to_bytes_consistent	754 (-12)	-1.57%
higher_order_functions	1,215 (-21)	-1.70%
nested_dyn_array_regression_5782	171 (-3)	-1.72%
regression_6674_2	819 (-15)	-1.80%
sha256_var_size_regression	16,311 (-303)	-1.82%
slice_loop	939 (-18)	-1.88%
to_bits	614 (-12)	-1.92%
regression_struct_array_conditional	1,515 (-30)	-1.94%
brillig_pedersen	782 (-18)	-2.25%
pedersen_check	782 (-18)	-2.25%
poseidon_bn254_hash	162,443 (-4,097)	-2.46%
poseidon_bn254_hash_width_3	162,443 (-4,097)	-2.46%
conditional_regression_661	117 (-3)	-2.50%
hashmap	54,038 (-1,502)	-2.70%
ecdsa_secp256k1	6,540 (-197)	-2.92%
if_else_chain	98 (-3)	-2.97%
brillig_rc_regression_6123	290 (-9)	-3.01%
regression_6674_3	1,561 (-51)	-3.16%
conditional_regression_421	89 (-3)	-3.26%
brillig_constant_reference_regression	81 (-3)	-3.57%
struct_inputs	566 (-21)	-3.58%
conditional_1	5,225 (-200)	-3.69%
sha256_var_witness_const_regression	6,571 (-263)	-3.85%
simple_radix	69 (-3)	-4.17%
slices	3,335 (-159)	-4.55%
sha256	13,869 (-663)	-4.56%
array_dynamic_nested_blackbox_input	4,301 (-206)	-4.57%
databus_composite_calldata	617 (-30)	-4.64%
conditional_regression_short_circuit	7,030 (-394)	-5.31%
6	6,952 (-394)	-5.36%
brillig_uninitialized_arrays	44 (-3)	-6.38%
side_effects_constrain_array	86 (-6)	-6.52%
simple_2d_array	125 (-9)	-6.72%
nested_array_dynamic	3,417 (-303)	-8.15%
regression_4449	200,854 (-20,438)	-9.24%
wildcard_type	471 (-51)	-9.77%
slice_regex	3,474 (-387)	-10.02%
nested_array_in_slice	1,564 (-177)	-10.17%
reference_counts	397 (-60)	-13.13%
fold_complex_outputs	762 (-150)	-16.45%
nested_array_dynamic_simple	90 (-30)	-25.00%
array_dedup_regression	677 (-413)	-37.89%
brillig_nested_arrays	155 (-198)	-56.09%

[github-actions]

Peak Memory Sample

Program	Peak Memory	%
keccak256	78.17M	0%
workspace	121.86M	-1%
regression_4709	294.72M	0%
ram_blowup_regression	2.44G	0%
private-kernel-tail	208.77M	0%
private-kernel-reset	861.32M	0%
private-kernel-inner	307.52M	0%
parity-root	174.42M	0%

[github-actions]

Compilation Sample

Program	Compilation Time	%
sha256_regression	0m1.429s	-4%
regression_4709	0m1.528s	0%
ram_blowup_regression	0m16.630s	-3%
private-kernel-tail	0m1.102s	-14%
private-kernel-reset	0m9.585s	7%
private-kernel-inner	0m2.388s	2%
parity-root	0m0.919s	-20%
noir-contracts	2m46.807s	4%

[github-actions]

Changes to circuit sizes

Generated at commit: 5f8cb04e452c26fe2fcdeed400fe9ab58ce0515a, compared to commit: 1b0dd4149d9249f0ea4fb5e2228c688e0135618f

🧾 Summary (10% most significant diffs)

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
regression_struct_array_conditional	-16 ✅	-18.39%	-24 ✅	-0.74%
nested_array_in_slice	-70 ✅	-6.48%	-151 ✅	-2.65%

---

Full diff report 👇

Program	ACIR opcodes (+/-)	%	Circuit size (+/-)	%
array_dynamic_nested_blackbox_input	253 (-12)	-4.53%	7,326 (-35)	-0.48%
regression_struct_array_conditional	71 (-16)	-18.39%	3,203 (-24)	-0.74%
nested_array_in_slice	1,010 (-70)	-6.48%	5,550 (-151)	-2.65%