feat: measure plonk rounds #3065

ludamad · 2023-10-26T14:37:45Z

Features:

Adds round benchmarking executables
Adds benchmarking circuits of size 2**13 to 2**18
Refactoring

ultrahonk flame graph:
Reference with 'Ultra Honk Round Breakdown' below

ultraplonk flame graph:
Reference with 'Ultra Plonk Round Breakdown' below
To read this one, you have to count plonk_round from round 1 through 6.

Current stats:
ultrahonk ~43-49% faster than ultraplonk on 10x ECDSA, ~25% faster on 2**18 circuit basic arithmetic circuit

Example output of scripts/benchmarks.sh:

MACBOOK STATS 10 cores

Standard Plonk

Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
2023-10-26T19:40:15-04:00
Running ./bin/standard_plonk_bench
Run on (10 X 24.1218 MHz CPU s)
CPU Caches:
  L1 Data 64 KiB
  L1 Instruction 128 KiB
  L2 Unified 4096 KiB (x10)
Load Average: 63.73, 26.17, 14.93
---------------------------------------------------------------------------------
Benchmark                                       Time             CPU   Iterations
---------------------------------------------------------------------------------
construct_proof_standard_power_of_2/13       82.0 ms         70.9 ms            9
construct_proof_standard_power_of_2/14        132 ms          117 ms            6
construct_proof_standard_power_of_2/15        223 ms          202 ms            3
construct_proof_standard_power_of_2/16        413 ms          370 ms            2
construct_proof_standard_power_of_2/17        746 ms          716 ms            1
construct_proof_standard_power_of_2/18       1418 ms         1364 ms            1

Ultra Honk Round Breakdown

Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
2023-10-26T19:40:24-04:00
Running ./bin/ultra_honk_rounds_bench
Run on (10 X 24.121 MHz CPU s)
CPU Caches:
  L1 Data 64 KiB
  L1 Instruction 128 KiB
  L2 Unified 4096 KiB (x10)
Load Average: 59.43, 25.90, 14.90
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                        16.7 ms         16.6 ms            1
ROUND_WIRE_COMMITMENTS/iterations:1                 205 ms          169 ms            1
ROUND_SORTED_LIST_ACCUMULATOR/iterations:1          189 ms          169 ms            1
ROUND_GRAND_PRODUCT_COMPUTATION/iterations:1        452 ms          419 ms            1
ROUND_RELATION_CHECK                               1426 ms         1407 ms            1
ROUND_ZEROMORPH                                    1204 ms         1158 ms            1

Ultra Plonk Round Breakdown

Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
2023-10-26T19:40:56-04:00
Running ./bin/ultra_plonk_rounds_bench
Run on (10 X 24.1204 MHz CPU s)
CPU Caches:
  L1 Data 64 KiB
  L1 Instruction 128 KiB
  L2 Unified 4096 KiB (x10)
Load Average: 33.61, 23.16, 14.35
------------------------------------------------------------------------------------
Benchmark                                          Time             CPU   Iterations
------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                      121 ms          112 ms            1
ROUND_FIRST_WIRE_COMMITMENTS                     679 ms          657 ms            1
ROUND_SECOND_FIAT_SHAMIR_ETA                     636 ms          561 ms            1
ROUND_THIRD_FIAT_SHAMIR_BETA_GAMMA              1255 ms          972 ms            1
ROUND_FOURTH_FIAT_SHAMIR_ALPHA_AND_COMMIT       1944 ms         1893 ms            1
ROUND_FIFTH_COMPUTE_QUOTIENT_EVALUTION           245 ms          187 ms            4
ROUND_SIXTH_BATCH_OPEN                           569 ms          558 ms            1

Ultra Honk

Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
2023-10-26T19:42:43-04:00
Running ./bin/ultra_honk_bench
Run on (10 X 24.1212 MHz CPU s)
CPU Caches:
  L1 Data 64 KiB
  L1 Instruction 128 KiB
  L2 Unified 4096 KiB (x10)
Load Average: 10.61, 17.83, 13.27
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
construct_proof_ultrahonk/sha256                    510 ms          479 ms            2
construct_proof_ultrahonk/keccak                   1833 ms         1749 ms            1
construct_proof_ultrahonk/ecdsa_verification       3396 ms         3193 ms            1
construct_proof_ultrahonk/merkle_membership         297 ms          279 ms            3
construct_proof_ultrahonk_power_of_2/13             121 ms          112 ms            6
construct_proof_ultrahonk_power_of_2/14             187 ms          177 ms            4
construct_proof_ultrahonk_power_of_2/15             320 ms          294 ms            2
construct_proof_ultrahonk_power_of_2/16             547 ms          524 ms            1
construct_proof_ultrahonk_power_of_2/17            1029 ms          991 ms            1
construct_proof_ultrahonk_power_of_2/18            1936 ms         1875 ms            1

Ultra Plonk

Unable to determine clock rate from sysctl: hw.cpufrequency: No such file or directory
This does not affect benchmark measurements, only the metadata output.
2023-10-26T19:43:02-04:00
Running ./bin/ultra_plonk_bench
Run on (10 X 24.1209 MHz CPU s)
CPU Caches:
  L1 Data 64 KiB
  L1 Instruction 128 KiB
  L2 Unified 4096 KiB (x10)
Load Average: 9.93, 17.21, 13.15
----------------------------------------------------------------------------------------
Benchmark                                              Time             CPU   Iterations
----------------------------------------------------------------------------------------
construct_proof_ultraplonk/sha256                    672 ms          628 ms            1
construct_proof_ultraplonk/keccak                   2477 ms         2374 ms            1
construct_proof_ultraplonk/ecdsa_verification       5086 ms         4582 ms            1
construct_proof_ultraplonk/merkle_membership         459 ms          346 ms            2
construct_proof_ultraplonk_power_of_2/13             184 ms          126 ms            5
construct_proof_ultraplonk_power_of_2/14             293 ms          202 ms            3
construct_proof_ultraplonk_power_of_2/15             463 ms          349 ms            2
construct_proof_ultraplonk_power_of_2/16             772 ms          601 ms            1
construct_proof_ultraplonk_power_of_2/17            1410 ms         1229 ms            1
construct_proof_ultraplonk_power_of_2/18            2468 ms         2394 ms            1

MAINFRAME STATS, ~128 cores

Standard Plonk

2023-10-26T23:39:37+00:00
Running ./bin/standard_plonk_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 15.48, 13.93, 12.70
---------------------------------------------------------------------------------
Benchmark                                       Time             CPU   Iterations
---------------------------------------------------------------------------------
construct_proof_standard_power_of_2/13        128 ms          112 ms            6
construct_proof_standard_power_of_2/14        150 ms          131 ms            5
construct_proof_standard_power_of_2/15        182 ms          155 ms            4
construct_proof_standard_power_of_2/16        261 ms          203 ms            3
construct_proof_standard_power_of_2/17        410 ms          324 ms            2
construct_proof_standard_power_of_2/18        914 ms          710 ms            1

Ultra Honk Round Breakdown

2023-10-26T23:39:46+00:00
Running ./bin/ultra_honk_rounds_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 16.06, 14.13, 12.78
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                        29.8 ms         29.7 ms            1
ROUND_WIRE_COMMITMENTS/iterations:1                 100 ms         53.4 ms            1
ROUND_SORTED_LIST_ACCUMULATOR/iterations:1          110 ms         87.6 ms            1
ROUND_GRAND_PRODUCT_COMPUTATION/iterations:1        190 ms          116 ms            1
ROUND_RELATION_CHECK                                467 ms          300 ms            2
ROUND_ZEROMORPH                                     921 ms          807 ms            1

Ultra Plonk Round Breakdown

2023-10-26T23:40:21+00:00
Running ./bin/ultra_plonk_rounds_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 16.26, 14.76, 13.06
------------------------------------------------------------------------------------
Benchmark                                          Time             CPU   Iterations
------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                      238 ms          229 ms            1
ROUND_FIRST_WIRE_COMMITMENTS                     375 ms          270 ms            2
ROUND_SECOND_FIAT_SHAMIR_ETA                     282 ms          212 ms            3
ROUND_THIRD_FIAT_SHAMIR_BETA_GAMMA               675 ms          513 ms            1
ROUND_FOURTH_FIAT_SHAMIR_ALPHA_AND_COMMIT        775 ms          596 ms            1
ROUND_FIFTH_COMPUTE_QUOTIENT_EVALUTION          76.0 ms         56.9 ms           13
ROUND_SIXTH_BATCH_OPEN                           357 ms          291 ms            2

Ultra Honk

2023-10-26T23:43:17+00:00
Running ./bin/ultra_honk_bench
Run on (128 X 3444.11 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 36.58, 26.06, 17.88
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
construct_proof_ultrahonk/sha256                    344 ms          277 ms            3
construct_proof_ultrahonk/keccak                    913 ms          706 ms            1
construct_proof_ultrahonk/ecdsa_verification       1645 ms         1270 ms            1
construct_proof_ultrahonk/merkle_membership         231 ms          191 ms            4
construct_proof_ultrahonk_power_of_2/13             149 ms          127 ms            5
construct_proof_ultrahonk_power_of_2/14             184 ms          152 ms            5
construct_proof_ultrahonk_power_of_2/15             243 ms          193 ms            4
construct_proof_ultrahonk_power_of_2/16             365 ms          278 ms            3
construct_proof_ultrahonk_power_of_2/17             545 ms          423 ms            2
construct_proof_ultrahonk_power_of_2/18            1053 ms          750 ms            1

Ultra Plonk

2023-10-26T23:43:37+00:00
Running ./bin/ultra_plonk_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 28.74, 24.91, 17.67
----------------------------------------------------------------------------------------
Benchmark                                              Time             CPU   Iterations
----------------------------------------------------------------------------------------
construct_proof_ultraplonk/sha256                    473 ms          353 ms            2
construct_proof_ultraplonk/keccak                   1379 ms         1044 ms            1
construct_proof_ultraplonk/ecdsa_verification       2363 ms         1785 ms            1
construct_proof_ultraplonk/merkle_membership         275 ms          240 ms            3
construct_proof_ultraplonk_power_of_2/13             196 ms          181 ms            4
construct_proof_ultraplonk_power_of_2/14             226 ms          203 ms            3
construct_proof_ultraplonk_power_of_2/15             287 ms          238 ms            3
construct_proof_ultraplonk_power_of_2/16             363 ms          300 ms            2
construct_proof_ultraplonk_power_of_2/17             656 ms          532 ms            1
construct_proof_ultraplonk_power_of_2/18            1326 ms         1019 ms            1

…ore-profiling

…to ad/feat/measure_plonk_rounds

AztecBot · 2023-10-26T22:29:26Z

Benchmark results

No metrics with a significant change found.

Detailed results

All benchmarks are run on txs on the Benchmarking contract on the repository. Each tx consists of a batch call to create_note and increment_balance, which guarantees that each tx has a private call, a nested private call, a public call, and a nested public call, as well as an emitted private note, an unencrypted log, and public storage read and write.

This benchmark source data is available in JSON format on S3 here.

Values are compared against data from master at commit 524cecf4 and shown if the difference exceeds 1%.

L2 block published to L1

Each column represents the number of txs on an L2 block published to L1.

Metric	8 txs	32 txs	128 txs
l1_rollup_calldata_size_in_bytes	45,444	179,588	716,132
l1_rollup_calldata_gas	222,972	868,016	3,449,636
l1_rollup_execution_gas	842,059	3,595,124	22,205,005
l2_block_processing_time_in_ms	1,979 (-2%)	7,471 (-1%)	29,472 (-1%)
note_successful_decrypting_time_in_ms	293 (-1%)	862 (-1%)	3,152
note_trial_decrypting_time_in_ms	94.0 (-4%)	70.0 (-12%)	133 (-1%)
l2_block_building_time_in_ms	13,135 (-1%)	52,012 (-1%)	208,006 (-1%)
l2_block_rollup_simulation_time_in_ms	11,826 (-1%)	46,796 (-1%)	186,989 (-1%)
l2_block_public_tx_process_time_in_ms	1,268 (-1%)	5,087	20,528

L2 chain processing

Each column represents the number of blocks on the L2 chain where each block has 16 txs.

Metric	5 blocks	10 blocks
node_history_sync_time_in_ms	21,463 (+1%)	41,937
note_history_successful_decrypting_time_in_ms	1,991	3,913
note_history_trial_decrypting_time_in_ms	121	143 (-1%)
node_database_size_in_bytes	1,630,289	1,098,855
pxe_database_size_in_bytes	27,188	54,187

Circuits stats

Stats on running time and I/O sizes collected for every circuit run across all benchmarks.

Circuit	circuit_simulation_time_in_ms	circuit_input_size_in_bytes	circuit_output_size_in_bytes
private-kernel-init	72.1	61,697	18,841
private-kernel-ordering	43.6	24,233	8,089
base-rollup	1,767	656,309	811
root-rollup	80.5 (-1%)	4,072	1,097
private-kernel-inner	52.9	81,504	18,841
public-kernel-private-input	41.4	41,455	18,841
public-kernel-non-first-iteration	26.4	41,497	18,841
merge-rollup	0.891 (-2%)	2,592	873

Miscellaneous

Transaction sizes based on how many contracts are deployed in the tx.

Metric	0 deployed contracts	1 deployed contracts
tx_size_in_bytes	8,723	27,240

…easure_plonk_rounds

…ecProtocol/aztec3-packages into ad/feat/measure_plonk_rounds

ludamad · 2023-10-26T23:51:40Z

barretenberg/cpp/src/barretenberg/benchmark/honk_bench/benchmark_utilities.hpp

@@ -23,35 +24,25 @@ using namespace benchmark;

 namespace bench_utils {

-struct BenchParams {


Now local to the files

ludamad · 2023-10-26T23:53:53Z

barretenberg/cpp/src/barretenberg/plonk/proof_system/prover/c_bind.cpp

@@ -1,134 +0,0 @@
-#include "prover.hpp"
-


…nk_rounds

codygunton

This is great, thanks for this. I double checked that your purported circuit sizes are correct (they are) and I looked at ECDSA circuit size (it's $2^{19}$). I also merged master, which includes an optimization of ZeroMorph to see how performance changes, and I found we get a nice speedup (numbers from run on the mainframe).

Results after merging master, which includes Tohru's optimization.

Standard Plonk

2023-10-27T19:15:19+00:00
Running ./bin/standard_plonk_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 25.41, 20.94, 17.61
---------------------------------------------------------------------------------
Benchmark                                       Time             CPU   Iterations
---------------------------------------------------------------------------------
construct_proof_standard_power_of_2/13        135 ms          116 ms            6
construct_proof_standard_power_of_2/14        153 ms          130 ms            5
construct_proof_standard_power_of_2/15        185 ms          153 ms            4
construct_proof_standard_power_of_2/16        251 ms          194 ms            3
construct_proof_standard_power_of_2/17        443 ms          340 ms            2
construct_proof_standard_power_of_2/18        848 ms          659 ms            1

Ultra Honk Round Breakdown

2023-10-27T19:15:28+00:00
Running ./bin/ultra_honk_rounds_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 33.51, 22.75, 18.24
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                        28.7 ms         28.7 ms            1
ROUND_WIRE_COMMITMENTS/iterations:1                94.3 ms         45.0 ms            1
ROUND_SORTED_LIST_ACCUMULATOR/iterations:1          130 ms         89.6 ms            1
ROUND_GRAND_PRODUCT_COMPUTATION/iterations:1        172 ms          107 ms            1
ROUND_RELATION_CHECK                                489 ms          348 ms            2
ROUND_ZEROMORPH                                     375 ms          285 ms            3

Ultra Plonk Round Breakdown

2023-10-27T19:16:10+00:00
Running ./bin/ultra_plonk_rounds_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 33.74, 23.77, 18.75
------------------------------------------------------------------------------------
Benchmark                                          Time             CPU   Iterations
------------------------------------------------------------------------------------
ROUND_PREAMBLE/iterations:1                      222 ms          215 ms            1
ROUND_FIRST_WIRE_COMMITMENTS                     300 ms          216 ms            3
ROUND_SECOND_FIAT_SHAMIR_ETA                     323 ms          237 ms            3
ROUND_THIRD_FIAT_SHAMIR_BETA_GAMMA               693 ms          517 ms            2
ROUND_FOURTH_FIAT_SHAMIR_ALPHA_AND_COMMIT        841 ms          650 ms            1
ROUND_FIFTH_COMPUTE_QUOTIENT_EVALUTION          78.5 ms         62.3 ms           10
ROUND_SIXTH_BATCH_OPEN                           333 ms          269 ms            3

Ultra Honk

2023-10-27T19:19:22+00:00
Running ./bin/ultra_honk_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 33.09, 32.11, 23.24
---------------------------------------------------------------------------------------
Benchmark                                             Time             CPU   Iterations
---------------------------------------------------------------------------------------
construct_proof_ultrahonk/sha256                    291 ms          222 ms            3
construct_proof_ultrahonk/keccak                    648 ms          444 ms            2
construct_proof_ultrahonk/ecdsa_verification       1117 ms          725 ms            1
construct_proof_ultrahonk/merkle_membership         206 ms          162 ms            4
construct_proof_ultrahonk_power_of_2/13             149 ms          118 ms            6
construct_proof_ultrahonk_power_of_2/14             168 ms          135 ms            5
construct_proof_ultrahonk_power_of_2/15             217 ms          163 ms            4
construct_proof_ultrahonk_power_of_2/16             293 ms          213 ms            3
construct_proof_ultrahonk_power_of_2/17             424 ms          296 ms            2
construct_proof_ultrahonk_power_of_2/18             748 ms          443 ms            2

Ultra Plonk

2023-10-27T19:19:45+00:00
Running ./bin/ultra_plonk_bench
Run on (128 X 2649.99 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x64)
  L1 Instruction 32 KiB (x64)
  L2 Unified 512 KiB (x64)
  L3 Unified 32768 KiB (x8)
Load Average: 25.41, 30.41, 22.87
----------------------------------------------------------------------------------------
Benchmark                                              Time             CPU   Iterations
----------------------------------------------------------------------------------------
construct_proof_ultraplonk/sha256                    423 ms          329 ms            2
construct_proof_ultraplonk/keccak                   1555 ms         1041 ms            1
construct_proof_ultraplonk/ecdsa_verification       2670 ms         1938 ms            1
construct_proof_ultraplonk/merkle_membership         298 ms          246 ms            3
construct_proof_ultraplonk_power_of_2/13             208 ms          183 ms            4
construct_proof_ultraplonk_power_of_2/14             249 ms          209 ms            3
construct_proof_ultraplonk_power_of_2/15             308 ms          251 ms            3
construct_proof_ultraplonk_power_of_2/16             397 ms          310 ms            2
construct_proof_ultraplonk_power_of_2/17             660 ms          523 ms            1
construct_proof_ultraplonk_power_of_2/18            1336 ms         1030 ms            1

barretenberg/cpp/src/barretenberg/benchmark/honk_bench/ultra_honk_rounds.bench.cpp

barretenberg/cpp/src/barretenberg/benchmark/honk_bench/ultra_plonk_rounds.bench.cpp

🤖 I have created a release *beep* *boop* --- <details><summary>aztec-packages: 0.13.0</summary> ## [0.13.0](aztec-packages-v0.12.0...aztec-packages-v0.13.0) (2023-10-31) ### ⚠ BREAKING CHANGES * PXE.getNotes(...) + refactor of note types ([#3051](#3051)) ### Features * `FieldNote` ([#3037](#3037)) ([3d1ffd0](3d1ffd0)) * Add Aztec Boxes page to docs ([#2569](#2569)) ([997c15c](997c15c)) * Adding structure to Transcript ([#2937](#2937)) ([db67aa1](db67aa1)) * Compile noir contracts with noir_wasm ([#2737](#2737)) ([524cecf](524cecf)) * Dockerize aztec-cli ([#3031](#3031)) ([ec2e3c2](ec2e3c2)) * Efficient ZM quotient computation ([#3016](#3016)) ([ebda5fc](ebda5fc)) * **feature_branch:** Private Kernel Circuit ([#2740](#2740)) ([f800a36](f800a36)) * Measure plonk rounds ([#3065](#3065)) ([c8e1d8b](c8e1d8b)) * Migrate the init kernel CPP tests to noir ([#3091](#3091)) ([906429f](906429f)) * New script to output table of benchmarks for README pasting. ([#2780](#2780)) ([6c20b45](6c20b45)) * Pedersen in typescript. ([#3111](#3111)) ([933f1b2](933f1b2)) * Protogalaxy folding of challenges ([#2935](#2935)) ([7ed30e8](7ed30e8)) * PXE.getNotes(...) + refactor of note types ([#3051](#3051)) ([16abb5a](16abb5a)) * Zeromorph with concatenation (Goblin Translator part 10) ([#3006](#3006)) ([70b0f17](70b0f17)) ### Bug Fixes * Bad contract txs publishing contract data ([#2673](#2673)) ([ccd4611](ccd4611)) * Better error message for compute_note_hash_and_nullifier. ([#3097](#3097)) ([57bec53](57bec53)) * Broken `FieldNote` test ([#3135](#3135)) ([fe78ecf](fe78ecf)) * Docker-compose up, rather than run. ([#3081](#3081)) ([242f780](242f780)) * Formatting ([#3070](#3070)) ([e1633d3](e1633d3)) * Minor stale naming fix ([#3117](#3117)) ([a6786ae](a6786ae)) * Push cli docker image to docker hub ([#3120](#3120)) ([ccad50f](ccad50f)) * Remove duplicate terraform resource definition ([#3066](#3066)) ([d5abadb](d5abadb)) * Retry request spot ([#3116](#3116)) ([82de5f1](82de5f1)) ### Miscellaneous * Add stdlib tests for pedersen commitment ([#3075](#3075)) ([87fa621](87fa621)) * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](#3072)) ([2e289a5](2e289a5)) * Cleanup remaining mentions of `compress` with pedersen in cpp and ts ([#3074](#3074)) ([52cf383](52cf383)) * E2e on spots [ci rebuild] ([#3068](#3068)) ([15db6bf](15db6bf)) * Fix dapp_testing e2e race condition ([#3094](#3094)) ([89e7c21](89e7c21)) * Remove docs mirror ([#3122](#3122)) ([3fa51e2](3fa51e2)) * Remove endomorphism coefficient from ecc_add_gate ([#3115](#3115)) ([d294987](d294987)) * Remove unecessary calls to `pedersen__init` ([#3079](#3079)) ([84f8db2](84f8db2)) * Remove unused pedersen c_binds ([#3058](#3058)) ([e71e5f9](e71e5f9)) * Removes pedersen commit native pairs method ([#3073](#3073)) ([69a34c7](69a34c7)) * Rename private-kernel subpackage to protocol-circuits ([#3134](#3134)) ([3e07104](3e07104)) ### Documentation * Initial keys spec ([#3035](#3035)) ([4b24c58](4b24c58)) </details> <details><summary>barretenberg.js: 0.13.0</summary> ## [0.13.0](barretenberg.js-v0.12.0...barretenberg.js-v0.13.0) (2023-10-31) ### Features * New script to output table of benchmarks for README pasting. ([#2780](#2780)) ([6c20b45](6c20b45)) ### Miscellaneous * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](#3072)) ([2e289a5](2e289a5)) * Remove unecessary calls to `pedersen__init` ([#3079](#3079)) ([84f8db2](84f8db2)) * Remove unused pedersen c_binds ([#3058](#3058)) ([e71e5f9](e71e5f9)) </details> <details><summary>barretenberg: 0.13.0</summary> ## [0.13.0](barretenberg-v0.12.0...barretenberg-v0.13.0) (2023-10-31) ### Features * Adding structure to Transcript ([#2937](#2937)) ([db67aa1](db67aa1)) * Efficient ZM quotient computation ([#3016](#3016)) ([ebda5fc](ebda5fc)) * Measure plonk rounds ([#3065](#3065)) ([c8e1d8b](c8e1d8b)) * New script to output table of benchmarks for README pasting. ([#2780](#2780)) ([6c20b45](6c20b45)) * Pedersen in typescript. ([#3111](#3111)) ([933f1b2](933f1b2)) * Protogalaxy folding of challenges ([#2935](#2935)) ([7ed30e8](7ed30e8)) * Zeromorph with concatenation (Goblin Translator part 10) ([#3006](#3006)) ([70b0f17](70b0f17)) ### Miscellaneous * Add stdlib tests for pedersen commitment ([#3075](#3075)) ([87fa621](87fa621)) * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](#3072)) ([2e289a5](2e289a5)) * Cleanup remaining mentions of `compress` with pedersen in cpp and ts ([#3074](#3074)) ([52cf383](52cf383)) * Remove endomorphism coefficient from ecc_add_gate ([#3115](#3115)) ([d294987](d294987)) * Remove unecessary calls to `pedersen__init` ([#3079](#3079)) ([84f8db2](84f8db2)) * Remove unused pedersen c_binds ([#3058](#3058)) ([e71e5f9](e71e5f9)) * Removes pedersen commit native pairs method ([#3073](#3073)) ([69a34c7](69a34c7)) </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).

🤖 I have created a release *beep* *boop* --- <details><summary>aztec-packages: 0.13.0</summary> ## [0.13.0](AztecProtocol/aztec-packages@aztec-packages-v0.12.0...aztec-packages-v0.13.0) (2023-10-31) ### ⚠ BREAKING CHANGES * PXE.getNotes(...) + refactor of note types ([#3051](AztecProtocol/aztec-packages#3051)) ### Features * `FieldNote` ([#3037](AztecProtocol/aztec-packages#3037)) ([3d1ffd0](AztecProtocol/aztec-packages@3d1ffd0)) * Add Aztec Boxes page to docs ([#2569](AztecProtocol/aztec-packages#2569)) ([997c15c](AztecProtocol/aztec-packages@997c15c)) * Adding structure to Transcript ([#2937](AztecProtocol/aztec-packages#2937)) ([db67aa1](AztecProtocol/aztec-packages@db67aa1)) * Compile noir contracts with noir_wasm ([#2737](AztecProtocol/aztec-packages#2737)) ([524cecf](AztecProtocol/aztec-packages@524cecf)) * Dockerize aztec-cli ([#3031](AztecProtocol/aztec-packages#3031)) ([ec2e3c2](AztecProtocol/aztec-packages@ec2e3c2)) * Efficient ZM quotient computation ([#3016](AztecProtocol/aztec-packages#3016)) ([ebda5fc](AztecProtocol/aztec-packages@ebda5fc)) * **feature_branch:** Private Kernel Circuit ([#2740](AztecProtocol/aztec-packages#2740)) ([f800a36](AztecProtocol/aztec-packages@f800a36)) * Measure plonk rounds ([#3065](AztecProtocol/aztec-packages#3065)) ([c8e1d8b](AztecProtocol/aztec-packages@c8e1d8b)) * Migrate the init kernel CPP tests to noir ([#3091](AztecProtocol/aztec-packages#3091)) ([906429f](AztecProtocol/aztec-packages@906429f)) * New script to output table of benchmarks for README pasting. ([#2780](AztecProtocol/aztec-packages#2780)) ([6c20b45](AztecProtocol/aztec-packages@6c20b45)) * Pedersen in typescript. ([#3111](AztecProtocol/aztec-packages#3111)) ([933f1b2](AztecProtocol/aztec-packages@933f1b2)) * Protogalaxy folding of challenges ([#2935](AztecProtocol/aztec-packages#2935)) ([7ed30e8](AztecProtocol/aztec-packages@7ed30e8)) * PXE.getNotes(...) + refactor of note types ([#3051](AztecProtocol/aztec-packages#3051)) ([16abb5a](AztecProtocol/aztec-packages@16abb5a)) * Zeromorph with concatenation (Goblin Translator part 10) ([#3006](AztecProtocol/aztec-packages#3006)) ([70b0f17](AztecProtocol/aztec-packages@70b0f17)) ### Bug Fixes * Bad contract txs publishing contract data ([#2673](AztecProtocol/aztec-packages#2673)) ([ccd4611](AztecProtocol/aztec-packages@ccd4611)) * Better error message for compute_note_hash_and_nullifier. ([#3097](AztecProtocol/aztec-packages#3097)) ([57bec53](AztecProtocol/aztec-packages@57bec53)) * Broken `FieldNote` test ([#3135](AztecProtocol/aztec-packages#3135)) ([fe78ecf](AztecProtocol/aztec-packages@fe78ecf)) * Docker-compose up, rather than run. ([#3081](AztecProtocol/aztec-packages#3081)) ([242f780](AztecProtocol/aztec-packages@242f780)) * Formatting ([#3070](AztecProtocol/aztec-packages#3070)) ([e1633d3](AztecProtocol/aztec-packages@e1633d3)) * Minor stale naming fix ([#3117](AztecProtocol/aztec-packages#3117)) ([a6786ae](AztecProtocol/aztec-packages@a6786ae)) * Push cli docker image to docker hub ([#3120](AztecProtocol/aztec-packages#3120)) ([ccad50f](AztecProtocol/aztec-packages@ccad50f)) * Remove duplicate terraform resource definition ([#3066](AztecProtocol/aztec-packages#3066)) ([d5abadb](AztecProtocol/aztec-packages@d5abadb)) * Retry request spot ([#3116](AztecProtocol/aztec-packages#3116)) ([82de5f1](AztecProtocol/aztec-packages@82de5f1)) ### Miscellaneous * Add stdlib tests for pedersen commitment ([#3075](AztecProtocol/aztec-packages#3075)) ([87fa621](AztecProtocol/aztec-packages@87fa621)) * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](AztecProtocol/aztec-packages#3072)) ([2e289a5](AztecProtocol/aztec-packages@2e289a5)) * Cleanup remaining mentions of `compress` with pedersen in cpp and ts ([#3074](AztecProtocol/aztec-packages#3074)) ([52cf383](AztecProtocol/aztec-packages@52cf383)) * E2e on spots [ci rebuild] ([#3068](AztecProtocol/aztec-packages#3068)) ([15db6bf](AztecProtocol/aztec-packages@15db6bf)) * Fix dapp_testing e2e race condition ([#3094](AztecProtocol/aztec-packages#3094)) ([89e7c21](AztecProtocol/aztec-packages@89e7c21)) * Remove docs mirror ([#3122](AztecProtocol/aztec-packages#3122)) ([3fa51e2](AztecProtocol/aztec-packages@3fa51e2)) * Remove endomorphism coefficient from ecc_add_gate ([#3115](AztecProtocol/aztec-packages#3115)) ([d294987](AztecProtocol/aztec-packages@d294987)) * Remove unecessary calls to `pedersen__init` ([#3079](AztecProtocol/aztec-packages#3079)) ([84f8db2](AztecProtocol/aztec-packages@84f8db2)) * Remove unused pedersen c_binds ([#3058](AztecProtocol/aztec-packages#3058)) ([e71e5f9](AztecProtocol/aztec-packages@e71e5f9)) * Removes pedersen commit native pairs method ([#3073](AztecProtocol/aztec-packages#3073)) ([69a34c7](AztecProtocol/aztec-packages@69a34c7)) * Rename private-kernel subpackage to protocol-circuits ([#3134](AztecProtocol/aztec-packages#3134)) ([3e07104](AztecProtocol/aztec-packages@3e07104)) ### Documentation * Initial keys spec ([#3035](AztecProtocol/aztec-packages#3035)) ([4b24c58](AztecProtocol/aztec-packages@4b24c58)) </details> <details><summary>barretenberg.js: 0.13.0</summary> ## [0.13.0](AztecProtocol/aztec-packages@barretenberg.js-v0.12.0...barretenberg.js-v0.13.0) (2023-10-31) ### Features * New script to output table of benchmarks for README pasting. ([#2780](AztecProtocol/aztec-packages#2780)) ([6c20b45](AztecProtocol/aztec-packages@6c20b45)) ### Miscellaneous * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](AztecProtocol/aztec-packages#3072)) ([2e289a5](AztecProtocol/aztec-packages@2e289a5)) * Remove unecessary calls to `pedersen__init` ([#3079](AztecProtocol/aztec-packages#3079)) ([84f8db2](AztecProtocol/aztec-packages@84f8db2)) * Remove unused pedersen c_binds ([#3058](AztecProtocol/aztec-packages#3058)) ([e71e5f9](AztecProtocol/aztec-packages@e71e5f9)) </details> <details><summary>barretenberg: 0.13.0</summary> ## [0.13.0](AztecProtocol/aztec-packages@barretenberg-v0.12.0...barretenberg-v0.13.0) (2023-10-31) ### Features * Adding structure to Transcript ([#2937](AztecProtocol/aztec-packages#2937)) ([db67aa1](AztecProtocol/aztec-packages@db67aa1)) * Efficient ZM quotient computation ([#3016](AztecProtocol/aztec-packages#3016)) ([ebda5fc](AztecProtocol/aztec-packages@ebda5fc)) * Measure plonk rounds ([#3065](AztecProtocol/aztec-packages#3065)) ([c8e1d8b](AztecProtocol/aztec-packages@c8e1d8b)) * New script to output table of benchmarks for README pasting. ([#2780](AztecProtocol/aztec-packages#2780)) ([6c20b45](AztecProtocol/aztec-packages@6c20b45)) * Pedersen in typescript. ([#3111](AztecProtocol/aztec-packages#3111)) ([933f1b2](AztecProtocol/aztec-packages@933f1b2)) * Protogalaxy folding of challenges ([#2935](AztecProtocol/aztec-packages#2935)) ([7ed30e8](AztecProtocol/aztec-packages@7ed30e8)) * Zeromorph with concatenation (Goblin Translator part 10) ([#3006](AztecProtocol/aztec-packages#3006)) ([70b0f17](AztecProtocol/aztec-packages@70b0f17)) ### Miscellaneous * Add stdlib tests for pedersen commitment ([#3075](AztecProtocol/aztec-packages#3075)) ([87fa621](AztecProtocol/aztec-packages@87fa621)) * Automatic c_binds for commit should return a point instead of an Fr element ([#3072](AztecProtocol/aztec-packages#3072)) ([2e289a5](AztecProtocol/aztec-packages@2e289a5)) * Cleanup remaining mentions of `compress` with pedersen in cpp and ts ([#3074](AztecProtocol/aztec-packages#3074)) ([52cf383](AztecProtocol/aztec-packages@52cf383)) * Remove endomorphism coefficient from ecc_add_gate ([#3115](AztecProtocol/aztec-packages#3115)) ([d294987](AztecProtocol/aztec-packages@d294987)) * Remove unecessary calls to `pedersen__init` ([#3079](AztecProtocol/aztec-packages#3079)) ([84f8db2](AztecProtocol/aztec-packages@84f8db2)) * Remove unused pedersen c_binds ([#3058](AztecProtocol/aztec-packages#3058)) ([e71e5f9](AztecProtocol/aztec-packages@e71e5f9)) * Removes pedersen commit native pairs method ([#3073](AztecProtocol/aztec-packages#3073)) ([69a34c7](AztecProtocol/aztec-packages@69a34c7)) </details> --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please).

ludamad0 and others added 30 commits October 20, 2023 17:26

Tinkering

9c3045b

mesing

6915582

add parallel_for_batched

bc2981b

tinkering

887b346

Zero morph cleanup

889dc1f

batch 1 loop

e930a31

batch 1 loop

fbd0ef4

Update

6f1a26d

Benchmark zeromorph

68f7bd2

Benchmark zeromorph

bedd0b3

Benchmark zeromorph

cc32991

Benchmark zeromorph

f7375cf

stash

1c222d6

stash

dc72e3f

stash

26ef756

Get compiling again

e37b673

Merge remote-tracking branch 'origin/master' into feat/more-profiling

42cadb4

Surprising results

b8840d7

Surprising results

5bdc5da

Undo

a740a24

Simplify batched parallel_for

2fbe761

chore: cleanup old profiling entrypoint, default to honk_passes

95ab25a

Merge branch 'master' into feat/more-profiling

3b801a3

Fix profile script

cf19371

Force fast passes to be one iteration

1e65db3

Merge remote-tracking branch 'origin/feat/more-profiling' into feat/m…

5e02316

…ore-profiling

Lighter weight stuffs

799203d

review feedback

cd42d16

ultra plonk benching

4040fb8

ultra plonk benching

d7d775b

ludamad0 added 3 commits October 26, 2023 21:58

feat: power of 2 bench functions

527d0d0

Merge remote-tracking branch 'origin/ad/feat/measure_plonk_rounds' in…

85f7200

…to ad/feat/measure_plonk_rounds

feat: benchmarks.sh

bc6401d

ludamad0 and others added 9 commits October 26, 2023 23:02

chore: benchmark alignment

812e897

chore: simplification

99670ec

Time to be bold

bdacc65

Try update

a7a77a2

Merge https://github.com/AztecProtocol/aztec3-packages into ad/feat/m…

fd1c368

…easure_plonk_rounds

Merge branch 'ad/feat/measure_plonk_rounds' of https://github.com/Azt…

c870b73

…ecProtocol/aztec3-packages into ad/feat/measure_plonk_rounds

fix: benchmarks.sh

cb51790

chore: clarify

ff29f0d

chore: simplify

f24fe81

ludamad marked this pull request as ready for review October 26, 2023 23:51

ludamad commented Oct 26, 2023

View reviewed changes

barretenberg/cpp/src/barretenberg/plonk/proof_system/prover/c_bind.cpp

@@ -1,134 +0,0 @@

#include "prover.hpp"

Copy link

Collaborator Author

ludamad Oct 26, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Unused

ludamad requested a review from codygunton October 26, 2023 23:54

codygunton assigned ludamad Oct 27, 2023

ludamad and others added 3 commits October 27, 2023 14:44

Update ultra_honk_rounds.bench.cpp

5de8fda

Update benchmarks.sh

c9a7d61

Merge remote-tracking branch 'origin/master' into ad/feat/measure_plo…

097541b

…nk_rounds

codygunton approved these changes Oct 27, 2023

View reviewed changes

barretenberg/cpp/src/barretenberg/benchmark/honk_bench/ultra_honk_rounds.bench.cpp Outdated Show resolved Hide resolved

barretenberg/cpp/src/barretenberg/benchmark/honk_bench/ultra_plonk_rounds.bench.cpp Outdated Show resolved Hide resolved

ludamad added 2 commits October 27, 2023 15:52

Update ultra_honk_rounds.bench.cpp

0c7482b

Update ultra_plonk_rounds.bench.cpp

796a708

ludamad enabled auto-merge (squash) October 27, 2023 19:53

ludamad merged commit c8e1d8b into master Oct 27, 2023
2 checks passed

ludamad deleted the ad/feat/measure_plonk_rounds branch October 27, 2023 20:19

AztecBot mentioned this pull request Oct 27, 2023

chore(master): Release 0.13.0 #3067

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: measure plonk rounds #3065

feat: measure plonk rounds #3065

ludamad commented Oct 26, 2023 •

edited

Loading

AztecBot commented Oct 26, 2023 •

edited

Loading

L2 block published to L1

L2 chain processing

Circuits stats

Miscellaneous

ludamad Oct 26, 2023

ludamad Oct 26, 2023

codygunton left a comment •

edited

Loading

		@@ -23,35 +24,25 @@ using namespace benchmark;

		namespace bench_utils {

		struct BenchParams {

feat: measure plonk rounds #3065

feat: measure plonk rounds #3065

Conversation

ludamad commented Oct 26, 2023 • edited Loading

AztecBot commented Oct 26, 2023 • edited Loading

Benchmark results

L2 block published to L1

L2 chain processing

Circuits stats

Miscellaneous

ludamad Oct 26, 2023

Choose a reason for hiding this comment

ludamad Oct 26, 2023

Choose a reason for hiding this comment

codygunton left a comment • edited Loading

Choose a reason for hiding this comment

Results after merging master, which includes Tohru's optimization.

ludamad commented Oct 26, 2023 •

edited

Loading

AztecBot commented Oct 26, 2023 •

edited

Loading

codygunton left a comment •

edited

Loading