feat: inline simple functions #7160

asterite · 2025-01-23T13:27:11Z

Description

Problem

Part of #7148

Summary

Pending: add a few tests for this.

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.

This reverts commit f8ab025.

github-actions · 2025-01-23T13:30:46Z

Changes to number of Brillig opcodes executed

Generated at commit: 47cb8532cde5b4fef3f4fa607caef2cfba14a412, compared to commit: 8cab0e4578aada98402a95386e7c4258afd38a6c

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
trait_impl_base_type_inliner_min	-156 ✅	-37.23%
brillig_acir_as_brillig_inliner_min	-74 ✅	-38.74%
wrapping_operations_inliner_min	-84 ✅	-39.25%
derive_inliner_min	-412 ✅	-40.47%
fold_after_inlined_calls_inliner_min	-24 ✅	-40.68%
regression_6451_inliner_min	-38 ✅	-42.22%
regression_4124_inliner_min	-41 ✅	-42.71%
brillig_not_inliner_min	-30 ✅	-49.18%
binary_operator_overloading_inliner_min	-652 ✅	-50.54%
trait_as_return_type_inliner_zero	-22 ✅	-52.38%
array_to_slice_constant_length_inliner_min	-24 ✅	-55.81%
diamond_deps_0_inliner_min	-46 ✅	-63.89%
1327_concrete_in_generic_inliner_min	-83 ✅	-70.34%
brillig_identity_function_inliner_min	-85 ✅	-81.73%
brillig_loop_size_regression_inliner_min	-103 ✅	-83.74%
struct_inliner_min	-141 ✅	-84.43%
trait_as_return_type_inliner_min	-127 ✅	-86.39%
brillig_fns_as_values_inliner_zero	-160 ✅	-88.89%
brillig_fns_as_values_inliner_min	-250 ✅	-92.59%

Full diff report 👇

Program	Brillig opcodes (+/-)	%
keccak256_inliner_zero	31,641 (+870)	+2.83%
6_inliner_zero	7,832 (+93)	+1.20%
conditional_regression_short_circuit_inliner_zero	7,912 (+93)	+1.19%
strings_inliner_max	1,722 (+6)	+0.35%
slice_regex_inliner_max	3,103 (+6)	+0.19%
strings_inliner_zero	2,077 (+3)	+0.14%
ram_blowup_regression_inliner_max	683,267 (-1)	-0.00%
brillig_cow_regression_inliner_zero	540,398 (-1)	-0.00%
brillig_cow_regression_inliner_max	454,762 (-1)	-0.00%
fold_2_to_17_inliner_max	1,056,623 (-4)	-0.00%
sha256_regression_inliner_zero	155,323 (-1)	-0.00%
poseidonsponge_x5_254_inliner_max	171,515 (-2)	-0.00%
regression_5252_inliner_max	854,142 (-10)	-0.00%
poseidon_bn254_hash_inliner_max	151,734 (-2)	-0.00%
poseidon_bn254_hash_width_3_inliner_max	151,734 (-2)	-0.00%
sha256_var_padding_regression_inliner_max	196,666 (-6)	-0.00%
sha256_regression_inliner_max	111,644 (-4)	-0.00%
regression_5252_inliner_min	1,021,038 (-56)	-0.01%
regression_4709_inliner_min	1,064,045 (-61)	-0.01%
poseidonsponge_x5_254_inliner_min	204,273 (-13)	-0.01%
sha256_var_size_regression_inliner_max	15,504 (-1)	-0.01%
sha256_inliner_max	13,198 (-1)	-0.01%
u128_inliner_zero	44,448 (-5)	-0.01%
array_dynamic_blackbox_input_inliner_max	17,473 (-2)	-0.01%
poseidon_bn254_hash_inliner_min	183,702 (-25)	-0.01%
poseidon_bn254_hash_width_3_inliner_min	183,702 (-25)	-0.01%
sha256_var_witness_const_regression_inliner_max	6,268 (-1)	-0.02%
hashmap_inliner_max	48,267 (-10)	-0.02%
uhashmap_inliner_max	136,213 (-56)	-0.04%
slice_regex_inliner_zero	3,862 (-2)	-0.05%
u128_inliner_max	24,249 (-27)	-0.11%
uhashmap_inliner_zero	174,599 (-312)	-0.18%
array_to_slice_inliner_zero	1,958 (-4)	-0.20%
array_to_slice_inliner_max	1,621 (-4)	-0.25%
hashmap_inliner_zero	74,878 (-205)	-0.27%
verify_honk_proof_inliner_min	4,218 (-16)	-0.38%
double_verify_honk_proof_inliner_min	7,477 (-32)	-0.43%
ram_blowup_regression_inliner_min	817,990 (-3,989)	-0.49%
brillig_cow_regression_inliner_min	543,441 (-2,782)	-0.51%
fold_numeric_generic_poseidon_inliner_min	5,786 (-31)	-0.53%
regression_6674_3_inliner_zero	1,364 (-8)	-0.58%
derive_inliner_zero	331 (-2)	-0.60%
to_be_bytes_inliner_min	2,125 (-15)	-0.70%
derive_inliner_max	136 (-1)	-0.73%
single_verify_proof_inliner_min	1,530 (-15)	-0.97%
regression_6674_3_inliner_max	1,358 (-14)	-1.02%
brillig_rc_regression_6123_inliner_max	285 (-3)	-1.04%
brillig_rc_regression_6123_inliner_zero	285 (-3)	-1.04%
sha256_brillig_performance_regression_inliner_min	26,800 (-321)	-1.18%
to_le_bytes_inliner_min	1,122 (-15)	-1.32%
sha256_var_padding_regression_inliner_min	267,094 (-3,690)	-1.36%
double_verify_proof_inliner_min	2,199 (-31)	-1.39%
hash_to_field_inliner_min	965 (-15)	-1.53%
to_bytes_integration_inliner_min	2,389 (-45)	-1.85%
sha256_regression_inliner_min	157,332 (-3,690)	-2.29%
conditional_1_inliner_min	5,954 (-149)	-2.44%
regression_6674_2_inliner_max	593 (-16)	-2.63%
regression_6674_1_inliner_max	590 (-16)	-2.64%
slice_coercion_inliner_min	469 (-13)	-2.70%
array_dynamic_blackbox_input_inliner_min	22,308 (-779)	-3.37%
simple_shield_inliner_zero	2,718 (-101)	-3.58%
regression_4449_inliner_min	266,453 (-10,979)	-3.96%
nested_array_dynamic_inliner_max	3,008 (-136)	-4.33%
to_bytes_consistent_inliner_min	659 (-30)	-4.35%
array_dedup_regression_inliner_min	1,232 (-64)	-4.94%
slice_regex_inliner_min	8,490 (-464)	-5.18%
nested_array_in_slice_inliner_max	1,386 (-76)	-5.20%
sha256_var_size_regression_inliner_min	22,086 (-1,260)	-5.40%
keccak256_inliner_min	32,022 (-1,876)	-5.53%
debug_logs_inliner_min	10,317 (-636)	-5.81%
regression_7128_inliner_min	1,216 (-76)	-5.88%
embedded_curve_ops_inliner_min	666 (-43)	-6.06%
slices_inliner_min	4,269 (-282)	-6.20%
array_to_slice_inliner_min	2,238 (-150)	-6.28%
regression_6674_3_inliner_min	1,976 (-137)	-6.48%
array_sort_inliner_min	678 (-50)	-6.87%
7_function_inliner_max	2,059 (-152)	-6.87%
ecdsa_secp256k1_inliner_min	8,165 (-629)	-7.15%
reference_counts_inliner_min	970 (-86)	-8.14%
conditional_regression_short_circuit_inliner_min	8,701 (-779)	-8.22%
6_inliner_min	8,595 (-779)	-8.31%
regression_3394_inliner_min	127 (-12)	-8.63%
7_function_inliner_min	2,605 (-247)	-8.66%
slice_dynamic_index_inliner_max	4,248 (-408)	-8.76%
slice_dynamic_index_inliner_zero	4,248 (-408)	-8.76%
references_inliner_min	642 (-65)	-9.19%
sha256_var_witness_const_regression_inliner_min	7,704 (-780)	-9.19%
simple_print_inliner_min	240 (-25)	-9.43%
loop_keyword_inliner_min	102 (-11)	-9.73%
loop_keyword_inliner_zero	102 (-11)	-9.73%
fmtstr_with_global_inliner_min	124 (-14)	-10.14%
pedersen_check_inliner_zero	852 (-97)	-10.22%
sha256_inliner_min	16,448 (-1,995)	-10.82%
array_dynamic_nested_blackbox_input_inliner_min	4,921 (-629)	-11.33%
aes128_encrypt_inliner_min	3,482 (-480)	-12.12%
inline_decompose_hint_brillig_call_inliner_min	274 (-39)	-12.46%
slice_loop_inliner_min	1,284 (-184)	-12.53%
brillig_pedersen_inliner_min	1,035 (-149)	-12.58%
u16_support_inliner_min	82 (-12)	-12.77%
higher_order_functions_inliner_min	2,291 (-336)	-12.79%
reference_only_used_as_alias_inliner_min	475 (-70)	-12.84%
pedersen_commitment_inliner_min	319 (-50)	-13.55%
uhashmap_inliner_min	196,618 (-32,524)	-14.19%
databus_inliner_min	86 (-15)	-14.85%
simple_2d_array_inliner_min	171 (-30)	-14.93%
regression_7062_inliner_min	79 (-14)	-15.05%
pedersen_hash_inliner_min	548 (-99)	-15.30%
brillig_calls_conditionals_inliner_min	198 (-36)	-15.38%
fold_complex_outputs_inliner_min	1,085 (-210)	-16.22%
simple_shield_inliner_min	2,785 (-541)	-16.27%
merkle_insert_inliner_min	3,388 (-676)	-16.63%
tuple_inputs_inliner_min	647 (-135)	-17.26%
nested_array_dynamic_inliner_min	3,535 (-765)	-17.79%
u128_inliner_min	52,143 (-11,625)	-18.23%
struct_inputs_inliner_min	662 (-150)	-18.47%
regression_4088_inliner_min	60 (-14)	-18.92%
brillig_rc_regression_6123_inliner_min	359 (-84)	-18.96%
brillig_uninitialized_arrays_inliner_min	64 (-15)	-18.99%
hashmap_inliner_min	90,205 (-22,608)	-20.04%
pedersen_check_inliner_min	1,035 (-264)	-20.32%
sha2_byte_inliner_min	89,381 (-23,004)	-20.47%
nested_array_in_slice_inliner_min	1,721 (-450)	-20.73%
strings_inliner_min	2,896 (-759)	-20.77%
conditional_2_inliner_min	158 (-43)	-21.39%
regression_inliner_min	3,057 (-878)	-22.31%
prelude_inliner_min	235 (-68)	-22.44%
submodules_inliner_min	38 (-11)	-22.45%
regression_3051_inliner_min	201 (-59)	-22.69%
generics_inliner_min	328 (-101)	-23.54%
bit_shifts_comptime_inliner_min	35 (-11)	-23.91%
bit_shifts_comptime_inliner_zero	35 (-11)	-23.91%
global_consts_inliner_min	1,697 (-536)	-24.00%
brillig_cow_inliner_min	1,415 (-450)	-24.13%
side_effects_constrain_array_inliner_min	136 (-45)	-24.86%
signed_division_inliner_min	247 (-84)	-25.38%
closures_mut_ref_inliner_zero	34 (-13)	-27.66%
brillig_calls_inliner_min	189 (-74)	-28.14%
array_neq_inliner_min	1,204 (-480)	-28.50%
array_eq_inliner_min	1,203 (-480)	-28.52%
regression_method_cannot_be_found_inliner_min	85 (-37)	-30.33%
5_over_inliner_min	64 (-28)	-30.43%
submodules_inliner_zero	25 (-11)	-30.56%
no_predicates_brillig_inliner_min	29 (-13)	-30.95%
7_inliner_min	1,062 (-480)	-31.13%
blake3_inliner_min	1,062 (-480)	-31.13%
4_sub_inliner_min	61 (-28)	-31.46%
brillig_blake2s_inliner_min	1,059 (-494)	-31.81%
6_array_inliner_min	2,890 (-1,400)	-32.63%
array_len_inliner_min	203 (-101)	-33.22%
field_attribute_inliner_min	24 (-12)	-33.33%
import_inliner_min	140 (-70)	-33.33%
unary_operator_overloading_inliner_min	70 (-37)	-34.58%
modules_inliner_min	24 (-13)	-35.14%
modules_more_inliner_min	24 (-13)	-35.14%
missing_closure_env_inliner_zero	22 (-12)	-35.29%
global_var_regression_simple_inliner_min	214 (-122)	-36.31%
trait_impl_base_type_inliner_min	263 (-156)	-37.23%
brillig_acir_as_brillig_inliner_min	117 (-74)	-38.74%
wrapping_operations_inliner_min	130 (-84)	-39.25%
derive_inliner_min	606 (-412)	-40.47%
fold_after_inlined_calls_inliner_min	35 (-24)	-40.68%
regression_6451_inliner_min	52 (-38)	-42.22%
regression_4124_inliner_min	55 (-41)	-42.71%
brillig_not_inliner_min	31 (-30)	-49.18%
binary_operator_overloading_inliner_min	638 (-652)	-50.54%
trait_as_return_type_inliner_zero	20 (-22)	-52.38%
array_to_slice_constant_length_inliner_min	19 (-24)	-55.81%
diamond_deps_0_inliner_min	26 (-46)	-63.89%
1327_concrete_in_generic_inliner_min	35 (-83)	-70.34%
brillig_identity_function_inliner_min	19 (-85)	-81.73%
brillig_loop_size_regression_inliner_min	20 (-103)	-83.74%
struct_inliner_min	26 (-141)	-84.43%
trait_as_return_type_inliner_min	20 (-127)	-86.39%
brillig_fns_as_values_inliner_zero	20 (-160)	-88.89%
brillig_fns_as_values_inliner_min	20 (-250)	-92.59%

github-actions · 2025-01-23T13:30:50Z

Changes to Brillig bytecode sizes

Generated at commit: 47cb8532cde5b4fef3f4fa607caef2cfba14a412, compared to commit: 8cab0e4578aada98402a95386e7c4258afd38a6c

🧾 Summary (10% most significant diffs)

Program	Brillig opcodes (+/-)	%
brillig_not_inliner_min	-18 ✅	-32.73%
closures_mut_ref_inliner_zero	-17 ✅	-33.33%
regression_7128_inliner_min	-78 ✅	-34.82%
brillig_acir_as_brillig_inliner_min	-61 ✅	-35.06%
trait_impl_base_type_inliner_min	-146 ✅	-36.87%
regression_4124_inliner_min	-33 ✅	-37.50%
missing_closure_env_inliner_zero	-16 ✅	-38.10%
trait_as_return_type_inliner_zero	-14 ✅	-38.89%
array_len_inliner_min	-81 ✅	-42.86%
fold_after_inlined_calls_inliner_min	-27 ✅	-43.55%
array_to_slice_constant_length_inliner_min	-20 ✅	-48.78%
diamond_deps_0_inliner_min	-30 ✅	-51.72%
1327_concrete_in_generic_inliner_min	-59 ✅	-64.13%
brillig_loop_size_regression_inliner_min	-68 ✅	-75.56%
struct_inliner_min	-125 ✅	-79.62%
brillig_identity_function_inliner_min	-85 ✅	-80.19%
trait_as_return_type_inliner_min	-94 ✅	-81.03%
brillig_fns_as_values_inliner_zero	-148 ✅	-87.06%
brillig_fns_as_values_inliner_min	-199 ✅	-90.05%

Full diff report 👇

Program	Brillig opcodes (+/-)	%
keccak256_inliner_zero	1,967 (+283)	+16.81%
6_inliner_zero	843 (+113)	+15.48%
conditional_regression_short_circuit_inliner_zero	925 (+113)	+13.92%
keccak256_inliner_min	1,665 (+157)	+10.41%
simple_shield_inliner_zero	812 (+28)	+3.57%
reference_counts_inliner_zero	663 (+5)	+0.76%
strings_inliner_max	928 (+6)	+0.65%
slice_regex_inliner_max	2,076 (+6)	+0.29%
sha256_regression_inliner_zero	4,848 (-1)	-0.02%
poseidon_bn254_hash_inliner_max	5,268 (-2)	-0.04%
poseidon_bn254_hash_width_3_inliner_max	5,268 (-2)	-0.04%
regression_5252_inliner_max	4,433 (-2)	-0.05%
brillig_cow_regression_inliner_zero	2,089 (-1)	-0.05%
poseidonsponge_x5_254_inliner_max	4,105 (-2)	-0.05%
brillig_cow_regression_inliner_max	2,051 (-1)	-0.05%
regression_4709_inliner_min	133,741 (-67)	-0.05%
hashmap_inliner_max	19,605 (-10)	-0.05%
sha256_regression_inliner_max	6,687 (-4)	-0.06%
sha256_var_size_regression_inliner_max	1,664 (-1)	-0.06%
sha256_inliner_max	2,288 (-2)	-0.09%
array_dynamic_blackbox_input_inliner_max	1,006 (-1)	-0.10%
ram_blowup_regression_inliner_max	928 (-1)	-0.11%
sha256_var_padding_regression_inliner_max	4,906 (-6)	-0.12%
sha256_var_witness_const_regression_inliner_zero	803 (-1)	-0.12%
slice_regex_inliner_zero	1,600 (-2)	-0.12%
uhashmap_inliner_max	12,121 (-17)	-0.14%
sha256_inliner_zero	1,253 (-2)	-0.16%
sha256_var_witness_const_regression_inliner_max	1,248 (-2)	-0.16%
hashmap_inliner_zero	7,889 (-19)	-0.24%
u128_inliner_zero	1,986 (-5)	-0.25%
poseidon_bn254_hash_width_3_inliner_min	4,985 (-14)	-0.28%
poseidon_bn254_hash_inliner_min	4,985 (-14)	-0.28%
poseidonsponge_x5_254_inliner_min	3,129 (-9)	-0.29%
regression_5252_inliner_min	3,542 (-19)	-0.53%
array_to_slice_inliner_zero	743 (-4)	-0.54%
array_to_slice_inliner_max	686 (-4)	-0.58%
nested_array_dynamic_inliner_min	1,856 (-11)	-0.59%
derive_inliner_zero	333 (-2)	-0.60%
sha256_regression_inliner_min	5,281 (-33)	-0.62%
derive_inliner_max	139 (-1)	-0.71%
fold_2_to_17_inliner_max	552 (-4)	-0.72%
uhashmap_inliner_zero	7,417 (-67)	-0.90%
sha256_var_padding_regression_inliner_min	3,232 (-33)	-1.01%
array_to_slice_inliner_min	995 (-11)	-1.09%
nested_array_in_slice_inliner_min	985 (-11)	-1.10%
u128_inliner_max	2,658 (-31)	-1.15%
regression_6674_3_inliner_zero	467 (-8)	-1.68%
brillig_rc_regression_6123_inliner_max	166 (-3)	-1.78%
brillig_rc_regression_6123_inliner_zero	166 (-3)	-1.78%
slice_coercion_inliner_min	486 (-9)	-1.82%
reference_counts_inliner_min	839 (-17)	-1.99%
conditional_1_inliner_min	1,418 (-32)	-2.21%
simple_shield_inliner_min	939 (-23)	-2.39%
aes128_encrypt_inliner_min	434 (-11)	-2.47%
sha2_byte_inliner_min	2,986 (-88)	-2.86%
array_dedup_regression_inliner_min	265 (-8)	-2.93%
regression_6674_3_inliner_max	461 (-14)	-2.95%
sha256_inliner_min	1,408 (-44)	-3.03%
array_dynamic_blackbox_input_inliner_min	1,275 (-43)	-3.26%
brillig_cow_inliner_min	321 (-11)	-3.31%
fold_complex_outputs_inliner_min	707 (-25)	-3.42%
ecdsa_secp256k1_inliner_min	1,213 (-43)	-3.42%
tuple_inputs_inliner_min	310 (-11)	-3.43%
sha256_var_size_regression_inliner_min	1,239 (-44)	-3.43%
array_dynamic_nested_blackbox_input_inliner_min	1,200 (-43)	-3.46%
fold_numeric_generic_poseidon_inliner_min	632 (-23)	-3.51%
struct_inputs_inliner_min	292 (-11)	-3.63%
ram_blowup_regression_inliner_min	1,130 (-43)	-3.67%
bigint_inliner_min	1,979 (-76)	-3.70%
slices_inliner_min	2,517 (-100)	-3.82%
regression_4449_inliner_min	1,066 (-43)	-3.88%
sha256_brillig_performance_regression_inliner_min	1,838 (-75)	-3.92%
pedersen_check_inliner_zero	536 (-22)	-3.94%
6_inliner_min	1,018 (-42)	-3.96%
sha256_var_witness_const_regression_inliner_min	984 (-44)	-4.28%
brillig_cow_regression_inliner_min	2,317 (-104)	-4.30%
debug_logs_inliner_min	9,751 (-439)	-4.31%
conditional_regression_short_circuit_inliner_min	1,119 (-54)	-4.60%
strings_inliner_zero	925 (-45)	-4.64%
to_be_bytes_inliner_min	225 (-11)	-4.66%
hash_to_field_inliner_min	213 (-11)	-4.91%
slice_regex_inliner_min	2,055 (-112)	-5.17%
u128_inliner_min	2,279 (-136)	-5.63%
regression_3394_inliner_min	124 (-8)	-6.06%
higher_order_functions_inliner_min	1,349 (-89)	-6.19%
embedded_curve_ops_inliner_min	515 (-35)	-6.36%
brillig_pedersen_inliner_min	594 (-41)	-6.46%
signed_division_inliner_min	230 (-16)	-6.50%
7_inliner_min	154 (-11)	-6.67%
blake3_inliner_min	154 (-11)	-6.67%
regression_5045_inliner_min	224 (-16)	-6.67%
simple_print_inliner_min	232 (-17)	-6.83%
fmtstr_with_global_inliner_min	121 (-10)	-7.63%
6_array_inliner_min	462 (-40)	-7.97%
nested_array_dynamic_inliner_max	1,973 (-172)	-8.02%
side_effects_constrain_array_inliner_min	124 (-11)	-8.15%
regression_7062_inliner_min	109 (-10)	-8.40%
regression_6674_3_inliner_min	837 (-80)	-8.72%
array_sort_inliner_min	355 (-34)	-8.74%
slice_loop_inliner_min	327 (-32)	-8.91%
7_function_inliner_min	635 (-65)	-9.29%
array_neq_inliner_min	107 (-11)	-9.32%
array_eq_inliner_min	106 (-11)	-9.40%
regression_6674_2_inliner_max	186 (-20)	-9.71%
regression_6674_1_inliner_max	183 (-20)	-9.85%
pedersen_commitment_inliner_min	236 (-26)	-9.92%
strings_inliner_min	1,138 (-126)	-9.97%
references_inliner_min	421 (-47)	-10.04%
merkle_insert_inliner_min	481 (-54)	-10.09%
loop_keyword_inliner_min	62 (-7)	-10.14%
loop_keyword_inliner_zero	62 (-7)	-10.14%
nested_array_in_slice_inliner_max	1,014 (-118)	-10.42%
u16_support_inliner_min	65 (-8)	-10.96%
simple_2d_array_inliner_min	146 (-18)	-10.98%
inline_decompose_hint_brillig_call_inliner_min	243 (-31)	-11.31%
to_le_bytes_inliner_min	161 (-22)	-12.02%
brillig_blake2s_inliner_min	151 (-21)	-12.21%
to_bytes_consistent_inliner_min	78 (-11)	-12.36%
pedersen_hash_inliner_min	358 (-51)	-12.47%
databus_inliner_min	74 (-11)	-12.94%
7_function_inliner_max	477 (-74)	-13.43%
pedersen_check_inliner_min	594 (-93)	-13.54%
single_verify_proof_inliner_min	69 (-11)	-13.75%
bit_shifts_comptime_inliner_min	43 (-7)	-14.00%
bit_shifts_comptime_inliner_zero	43 (-7)	-14.00%
brillig_uninitialized_arrays_inliner_min	67 (-11)	-14.10%
uhashmap_inliner_min	8,338 (-1,446)	-14.78%
verify_honk_proof_inliner_min	69 (-12)	-14.81%
slice_dynamic_index_inliner_max	2,123 (-377)	-15.08%
slice_dynamic_index_inliner_zero	2,123 (-377)	-15.08%
regression_4088_inliner_min	56 (-10)	-15.15%
brillig_calls_conditionals_inliner_min	134 (-24)	-15.19%
reference_only_used_as_alias_inliner_min	315 (-58)	-15.55%
submodules_inliner_min	38 (-7)	-15.56%
to_bytes_integration_inliner_min	178 (-33)	-15.64%
regression_inliner_min	1,110 (-215)	-16.23%
regression_5435_inliner_min	41 (-8)	-16.33%
conditional_2_inliner_min	164 (-33)	-16.75%
brillig_rc_regression_6123_inliner_min	230 (-52)	-18.44%
prelude_inliner_min	227 (-52)	-18.64%
no_predicates_brillig_inliner_min	39 (-9)	-18.75%
generics_inliner_min	244 (-58)	-19.21%
global_consts_inliner_min	253 (-61)	-19.43%
submodules_inliner_zero	29 (-7)	-19.44%
regression_3051_inliner_min	193 (-47)	-19.58%
double_verify_proof_inliner_min	81 (-20)	-19.80%
5_over_inliner_min	62 (-16)	-20.51%
double_verify_honk_proof_inliner_min	81 (-22)	-21.36%
4_sub_inliner_min	57 (-16)	-21.92%
hashmap_inliner_min	9,771 (-2,773)	-22.11%
field_attribute_inliner_min	28 (-8)	-22.22%
unary_operator_overloading_inliner_min	72 (-21)	-22.58%
regression_method_cannot_be_found_inliner_min	82 (-25)	-23.36%
brillig_calls_inliner_min	193 (-61)	-24.02%
modules_inliner_min	28 (-9)	-24.32%
modules_more_inliner_min	28 (-9)	-24.32%
wrapping_operations_inliner_min	104 (-40)	-27.78%
binary_operator_overloading_inliner_min	448 (-173)	-27.86%
import_inliner_min	134 (-52)	-27.96%
global_var_regression_simple_inliner_min	167 (-67)	-28.63%
regression_6451_inliner_min	58 (-26)	-30.95%
derive_inliner_min	584 (-270)	-31.62%
brillig_not_inliner_min	37 (-18)	-32.73%
closures_mut_ref_inliner_zero	34 (-17)	-33.33%
regression_7128_inliner_min	146 (-78)	-34.82%
brillig_acir_as_brillig_inliner_min	113 (-61)	-35.06%
trait_impl_base_type_inliner_min	250 (-146)	-36.87%
regression_4124_inliner_min	55 (-33)	-37.50%
missing_closure_env_inliner_zero	26 (-16)	-38.10%
trait_as_return_type_inliner_zero	22 (-14)	-38.89%
array_len_inliner_min	108 (-81)	-42.86%
fold_after_inlined_calls_inliner_min	35 (-27)	-43.55%
array_to_slice_constant_length_inliner_min	21 (-20)	-48.78%
diamond_deps_0_inliner_min	28 (-30)	-51.72%
1327_concrete_in_generic_inliner_min	33 (-59)	-64.13%
brillig_loop_size_regression_inliner_min	22 (-68)	-75.56%
struct_inliner_min	32 (-125)	-79.62%
brillig_identity_function_inliner_min	21 (-85)	-80.19%
trait_as_return_type_inliner_min	22 (-94)	-81.03%
brillig_fns_as_values_inliner_zero	22 (-148)	-87.06%
brillig_fns_as_values_inliner_min	22 (-199)	-90.05%

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Compilation Time'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: `36ff6ad`	Previous: `c8d5ce5`	Ratio
`rollup-root`	`4.438` s	`3.556` s	`1.25`
`rollup-merge`	`2.738` s	`2.004` s	`1.37`
`rollup-block-merge`	`4.384` s	`3.626` s	`1.21`
`rollup-base-private`	`12.78` s	`10.32` s	`1.24`

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

CC: @TomAFrench

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Compilation Memory'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: `8c96690`	Previous: `c8d5ce5`	Ratio
`ram_blowup_regression`	`2200` MB	`1480` MB	`1.49`
`rollup-block-root-single-tx`	`16210` MB	`6250` MB	`2.59`

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

CC: @TomAFrench

github-actions

⚠️ 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: `01ceaa3`	Previous: `c8d5ce5`	Ratio
`regression_4709`	`0.002` s	`0.001` s	`2`

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

CC: @TomAFrench

TomAFrench · 2025-01-23T14:41:50Z

Very nice! There's some regressions on keccak256 when using a zero aggressiveness, can you take a look (non blocker but it would need an issue)

asterite · 2025-01-23T15:31:22Z

Sure! I'll take a look at it before merging this.

asterite · 2025-01-23T16:35:37Z

@TomAFrench I was wondering if the --inliner-aggressiveness argument should be taken into account by this new pass or not. It's maybe strange that --inliner-aggressiveness=0 would still inline these, but I'm not sure.

(so far the only difference I found was that a to_le_radix call is inlined now, and that somehow makes things work with zero inlining)

asterite · 2025-01-23T17:53:04Z

I think I'll merge this PR and then look at that issue because it seems to need an extra fix.

asterite · 2025-01-23T18:02:51Z

Regarding keccak256, what I found was that std::hash::keccak256 calls the one in another module:

noir/noir_stdlib/src/hash/mod.nr

Lines 118 to 122 in 2da0a4f

    
           pub fn keccak256<let N: u32>(input: [u8; N], message_size: u32) -> [u8; 32] 
        
           // docs:end:keccak256 
        
           { 
        
               crate::hash::keccak::keccak256(input, message_size) 
        
           }

And that function is no_predicates. So what's end up happening is that the call to the no_predicates function ends up in the caller... but maybe that's fine? After all the wrapper function doesn't have no_predicates. From there I still couldn't understand why it changes the final SSA.

vezenovm · 2025-01-23T18:03:49Z

rollup-block-root-single-tx 16210 MB 6250 MB 2.59

Looks like we have a really big regression here as well reversing the gains made by #7072.

TomAFrench · 2025-01-23T18:15:10Z

Looks like we have a really big regression here as well reversing the gains made by #7072.

Interesting. I think this would likely be fixed by running some of the additional passes done as part of the preprocessing pass on the functions which we've just inlined into. Once we've inlined these simple functions then we will be able to remove a lot of allocations whereas I'm guessing that this PR is resulting in these optimizations being deferred again.

TomAFrench · 2025-01-23T18:17:22Z

@asterite Chucking a mem2reg pass afterwards should help

asterite · 2025-01-23T18:20:39Z

@asterite Chucking a mem2reg pass afterwards should help

Done.

Regarding keccak256, I think now that there are some inlined functions before the preoprocessing, the weight of functions changes and so that pass behaves differently... in this case maybe it makes it into a regression. It's a bit hard to debug or understand 😊

TomAFrench · 2025-01-23T18:36:07Z

Looks like the regression is fixed now but there's a test failure 🤔

TomAFrench · 2025-01-23T18:38:30Z

Yeah, inlining with weight scores is a little bit of a pain for that reason. Let's just record it and once we've done all of these optimizations which definitely should reduce bytecode sizes then we can see if there's ways to improve these cases.

vezenovm · 2025-01-23T18:44:24Z

Looks like the regression is fixed now but there's a test failure 🤔

I ran into the same thing in #7165 and fixed it there. I'll extract it.

vezenovm · 2025-01-23T18:49:10Z

Merging in #7169 should fix those panics in https://github.com/noir-lang/noir/actions/runs/12935521653/job/36079229284?pr=7160

vezenovm · 2025-01-23T19:15:10Z

I guess linking the comment we want to resolve will close the PR 😅

asterite · 2025-01-23T19:39:47Z

Looks like the regression is fixed

Where can this be seen?

vezenovm · 2025-01-23T19:42:47Z

Where can this be seen?

https://github.com/noir-lang/noir/actions/runs/12936395030/job/36082771664?pr=7160 under the merge reports job we print out the report to the shell. Looks like there was even an improvement.

fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…r-lang/noir#7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…r-lang/noir#7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…r-lang/noir#7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

…7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095)

Automated pull of development from the [noir](https://github.com/noir-lang/noir) programming language, a dependency of Aztec. BEGIN_COMMIT_OVERRIDE chore: let `add_definition_location` take a Location (noir-lang/noir#7185) fix(LSP): correct signature for assert and assert_eq (noir-lang/noir#7184) chore(experimental): Prevent enum panics by returning Options where possible instead of panicking (noir-lang/noir#7180) feat(experimental): Construct enum variants in expressions (noir-lang/noir#7174) feat: add `noir-inspector` (noir-lang/noir#7136) fix: ensure canonical bits decomposition (noir-lang/noir#7168) fix: Keep `inc_rc` for array inputs during preprocessing (noir-lang/noir#7163) fix(docs): Update broken links to EC lib (noir-lang/noir#7141) feat: inline simple functions (noir-lang/noir#7160) feat(ssa): Expand feature set of the Brillig constraint check (noir-lang/noir#7060) fix(ssa): Resolve value before fetching from DFG in a couple cases (noir-lang/noir#7169) fix: `Function::is_no_predicates` always returned false for brillig f… (noir-lang/noir#7167) chore(refactor): Remove globals field on Ssa object and use only the shared globals graph (noir-lang/noir#7156) chore: let `Function::inlined` take a `should_inline_call` function (noir-lang/noir#7149) chore: add compile-time assertions on generic arguments of stdlib functions (noir-lang/noir#6981) fix: LSP hover over function with `&mut self` (noir-lang/noir#7155) feat(brillig): Set global memory size at program compile time (noir-lang/noir#7151) feat: LSP autocomplete module declaration (noir-lang/noir#7154) feat(ssa): Reuse constants from the globals graph when making constants in a function DFG (noir-lang/noir#7153) feat: LSP chain inlay hints (noir-lang/noir#7152) chore: turn on overflow checks in CI rust tests (noir-lang/noir#7145) fix(ssa): Use post order when mapping instructions in loop invariant pass (noir-lang/noir#7140) fix: preserve types when reading from calldata arrays (noir-lang/noir#7144) feat: Resolve enums & prepare type system (noir-lang/noir#7115) feat: `loop` must have at least one `break` (noir-lang/noir#7126) feat: parse globals in SSA parser (noir-lang/noir#7112) fix: allow calling trait impl method from struct if multiple impls exist (noir-lang/noir#7124) fix: avoid creating unnecessary memory blocks (noir-lang/noir#7114) chore: relax threshold for reporting regressions (noir-lang/noir#7130) fix: proper cleanup when breaking from comptime loop on error (noir-lang/noir#7125) fix: Prevent overlapping associated types impls (noir-lang/noir#7047) feat: unconstrained optimizations for BoundedVec (noir-lang/noir#7119) chore: mark libs good (noir-lang/noir#7123) chore: remove comments for time/memory benchmarks (noir-lang/noir#7121) fix: don't always use an exclusive lock in `nargo check` (noir-lang/noir#7120) feat(ssa): Pass to preprocess functions (noir-lang/noir#7072) chore: Formatting issues / minor errors in the docs (noir-lang/noir#7105) fix: defunctionalize pass on the caller runtime to apply (noir-lang/noir#7100) feat: Parser and formatter support for `enum`s (noir-lang/noir#7110) feat(brillig): SSA globals code gen (noir-lang/noir#7021) feat: `loop` keyword in runtime and comptime code (noir-lang/noir#7096) chore: Add benchmarking dashboard (noir-lang/noir#7068) feat(experimental): try to infer lambda argument types inside calls (noir-lang/noir#7088) feat(ssa): Add flag to DIE pass to be able to keep `store` instructions (noir-lang/noir#7106) chore: Cookbook Onboard integration (noir-lang/noir#7044) chore: lock to ubuntu 22.04 (noir-lang/noir#7098) fix: Remove unused brillig functions (noir-lang/noir#7102) chore(ssa): Use correct prefix when printing array values in global space (noir-lang/noir#7095) END_COMMIT_OVERRIDE --------- Co-authored-by: Tom French <[email protected]> Co-authored-by: Tom French <[email protected]> Co-authored-by: aakoshh <[email protected]> Co-authored-by: Akosh Farkash <[email protected]> Co-authored-by: Maxim Vezenov <[email protected]>

asterite added 9 commits January 22, 2025 12:57

chore: let Function::inlined take a should_inline_call function

beb1d34

Remove PerFunctionContext from callback

0309ea9

Always check some conditions

891bc45

Let the callback receive &Function

abaa595

Don't confuse entry_function with source_function

27818df

Check that runtimes match

f8ab025

Revert "Check that runtimes match"

b8f4aad

This reverts commit f8ab025.

Move should_inline_call up

5f6ec4b

feat: inline simple functions

8c96690

Try to first see how it changes without the "called only once" condition

edde1a2

github-actions bot reviewed Jan 23, 2025

View reviewed changes

asterite added 2 commits January 23, 2025 10:45

Add a couple of tests

36ff6ad

Add the usual check we do when inlining

01ceaa3

github-actions bot reviewed Jan 23, 2025

View reviewed changes

Base automatically changed from ab/inlined-should_inline_cal to master January 23, 2025 16:05

Merge branch 'master' into ab/simple-inline

0256bdf

asterite enabled auto-merge January 23, 2025 17:53

asterite mentioned this pull request Jan 23, 2025

The "inline simple functions" pass caused a regression in keccak256 when inliner is zero #7166

Open

asterite disabled auto-merge January 23, 2025 18:16

vezenovm mentioned this pull request Jan 23, 2025

fix(ssa): Resolve value before fetching from DFG in a couple cases #7169

Merged

5 tasks

vezenovm closed this in #7169 Jan 23, 2025

vezenovm reopened this Jan 23, 2025

Merge branch 'master' into ab/simple-inline

2bc5b59

TomAFrench approved these changes Jan 23, 2025

View reviewed changes

TomAFrench added this pull request to the merge queue Jan 23, 2025

Merged via the queue into master with commit c17e228 Jan 23, 2025
99 checks passed

TomAFrench deleted the ab/simple-inline branch January 23, 2025 19:59

noirwhal mentioned this pull request Jan 23, 2025

chore: Release Noir(1.0.0-beta.2) #6914

Open

AztecBot mentioned this pull request Jan 24, 2025

feat: Sync from noir AztecProtocol/aztec-packages#11294

Merged

AztecBot mentioned this pull request Jan 30, 2025

chore(master): Release 0.73.0 AztecProtocol/aztec-packages#11496

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: inline simple functions #7160

feat: inline simple functions #7160

asterite commented Jan 23, 2025 •

edited

Loading

github-actions bot commented Jan 23, 2025 •

edited

Loading

github-actions bot commented Jan 23, 2025 •

edited

Loading

github-actions bot left a comment •

edited

Loading

github-actions bot left a comment •

edited

Loading

github-actions bot left a comment

TomAFrench commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

vezenovm commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

asterite commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

vezenovm commented Jan 23, 2025

vezenovm commented Jan 23, 2025

vezenovm commented Jan 23, 2025

asterite commented Jan 23, 2025

vezenovm commented Jan 23, 2025 •

edited

Loading

feat: inline simple functions #7160

feat: inline simple functions #7160

Conversation

asterite commented Jan 23, 2025 • edited Loading

Description

Problem

Summary

Additional Context

Documentation

PR Checklist

github-actions bot commented Jan 23, 2025 • edited Loading

Changes to number of Brillig opcodes executed

🧾 Summary (10% most significant diffs)

github-actions bot commented Jan 23, 2025 • edited Loading

Changes to Brillig bytecode sizes

🧾 Summary (10% most significant diffs)

github-actions bot left a comment • edited Loading

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

github-actions bot left a comment • edited Loading

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

github-actions bot left a comment

Choose a reason for hiding this comment

⚠️ Performance Alert ⚠️

TomAFrench commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

asterite commented Jan 23, 2025

vezenovm commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

asterite commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

TomAFrench commented Jan 23, 2025

vezenovm commented Jan 23, 2025

vezenovm commented Jan 23, 2025

vezenovm commented Jan 23, 2025

asterite commented Jan 23, 2025

vezenovm commented Jan 23, 2025 • edited Loading

asterite commented Jan 23, 2025 •

edited

Loading

github-actions bot commented Jan 23, 2025 •

edited

Loading

github-actions bot commented Jan 23, 2025 •

edited

Loading

github-actions bot left a comment •

edited

Loading

github-actions bot left a comment •

edited

Loading

vezenovm commented Jan 23, 2025 •

edited

Loading