status-im
/
constantine
mirror of https://github.com/status-im/constantine.git
2
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
constantine/constantine.nimble

838 lines
35 KiB
Plaintext
Raw Normal View History

Rebrand to Constantine. Bigints representation should stay opaque. Exporting just the word_types would make a super small library.
2018-12-01 19:12:05 +00:00
packageName = "constantine"
initial commit
2018-07-24 14:52:18 +00:00
version = "0.0.1"
author = "Status Research & Development GmbH"
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
description = "This library provides thoroughly tested and highly-optimized implementations of cryptography protocols."
initial commit
2018-07-24 14:52:18 +00:00
license = "MIT or Apache License 2.0"
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Dependencies
# ----------------------------------------------------------------
Skeleton of modular exponentiation
2020-02-22 15:37:31 +00:00
requires "nim >= 1.1.0"
initial commit
2018-07-24 14:52:18 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Test config
# ----------------------------------------------------------------
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
const buildParallel = "test_parallel.txt"
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# Testing strategy: to reduce CI time we test leaf functionality
# and skip testing codepath that would be exercised by leaves.
# While debugging, relevant unit-test can be reactivated.
# New features should stay on.
# Code refactoring requires re-enabling the full suite.
# Basic primitives should stay on to catch compiler regressions.
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
const testDesc: seq[tuple[path: string, useGMP: bool]] = @[
# Primitives
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_primitives.nim", false),
("tests/math/t_primitives_extended_precision.nim", false),
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Big ints
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_io_bigints.nim", false),
("tests/math/t_io_unsaturated.nim", false),
("tests/math/t_bigints.nim", false),
("tests/math/t_bigints_multimod.nim", false),
("tests/math/t_bigints_mod_vs_gmp.nim", true),
("tests/math/t_bigints_mul_vs_gmp.nim", true),
("tests/math/t_bigints_mul_high_words_vs_gmp.nim", true),
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Field
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_io_fields", false),
("tests/math/t_finite_fields.nim", false),
("tests/math/t_finite_fields_conditional_arithmetic.nim", false),
("tests/math/t_finite_fields_mulsquare.nim", false),
("tests/math/t_finite_fields_sqrt.nim", false),
("tests/math/t_finite_fields_powinv.nim", false),
("tests/math/t_finite_fields_vs_gmp.nim", true),
("tests/math/t_fp_cubic_root.nim", false),
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
# Double-precision finite fields
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_finite_fields_double_precision.nim", false),
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Towers of extension fields
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_fp2.nim", false),
("tests/math/t_fp2_sqrt.nim", false),
("tests/math/t_fp4.nim", false),
("tests/math/t_fp6_bn254_snarks.nim", false),
("tests/math/t_fp6_bls12_377.nim", false),
("tests/math/t_fp6_bls12_381.nim", false),
("tests/math/t_fp6_bw6_761.nim", false),
("tests/math/t_fp12_bn254_snarks.nim", false),
("tests/math/t_fp12_bls12_377.nim", false),
("tests/math/t_fp12_bls12_381.nim", false),
("tests/math/t_fp12_exponentiation.nim", false),
("tests/math/t_fp12_anti_regression.nim", false),
("tests/math/t_fp4_frobenius.nim", false),
("tests/math/t_fp6_frobenius.nim", false),
("tests/math/t_fp12_frobenius.nim", false),
Fast modular inversion (#172) * split modular inversion in its own file * Stash fast GCD inversion https://eprint.iacr.org/2020/972.pdf * Stash Pornin's bingcd -> issue with inner modular reduction * Implement Bernstein-Yang inversion * Avoid Nim checks on signed integers (32-bit runtime issue) * cleanup: remove old inversion impls * cleanup: static moduli, move div2 * small comments (skip ci) * comment cleanup (skip ci) * fix total iterations on 32-bit * Add batch conversion to affine coordinates using simultaneous inversion trick * fix conditional setZero and batchAffine conversion * cleanup unneeded branches following affine conversion unification * Fix batchAffine with zero inputs and add fuzz failure to test suite
2022-02-10 13:05:07 +00:00
# Elliptic curve arithmetic
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_ec_conversion.nim", false),
Fast modular inversion (#172) * split modular inversion in its own file * Stash fast GCD inversion https://eprint.iacr.org/2020/972.pdf * Stash Pornin's bingcd -> issue with inner modular reduction * Implement Bernstein-Yang inversion * Avoid Nim checks on signed integers (32-bit runtime issue) * cleanup: remove old inversion impls * cleanup: static moduli, move div2 * small comments (skip ci) * comment cleanup (skip ci) * fix total iterations on 32-bit * Add batch conversion to affine coordinates using simultaneous inversion trick * fix conditional setZero and batchAffine conversion * cleanup unneeded branches following affine conversion unification * Fix batchAffine with zero inputs and add fuzz failure to test suite
2022-02-10 13:05:07 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Elliptic curve arithmetic G1
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
# ("tests/math/t_ec_shortw_prj_g1_add_double.nim", false),
# ("tests/math/t_ec_shortw_prj_g1_mul_sanity.nim", false),
# ("tests/math/t_ec_shortw_prj_g1_mul_distri.nim", false),
("tests/math/t_ec_shortw_prj_g1_mul_vs_ref.nim", false),
("tests/math/t_ec_shortw_prj_g1_mixed_add.nim", false),
# ("tests/math/t_ec_shortw_jac_g1_add_double.nim", false),
# ("tests/math/t_ec_shortw_jac_g1_mul_sanity.nim", false),
# ("tests/math/t_ec_shortw_jac_g1_mul_distri.nim", false),
("tests/math/t_ec_shortw_jac_g1_mul_vs_ref.nim", false),
("tests/math/t_ec_shortw_jac_g1_mixed_add.nim", false),
("tests/math/t_ec_twedwards_prj_add_double", false),
("tests/math/t_ec_twedwards_prj_mul_sanity", false),
("tests/math/t_ec_twedwards_prj_mul_distri", false),
Initial support for Twisted Edwards curves (#167) * Point decoding: optimized sqrt for p ≡ 5 (mod 8) (Curve25519) * Implement fused sqrt(u/v) for twisted edwards point deserialization * Introduce twisted edwards affine * Allow declaration of curve field elements (and fight against recursive dependencies * Twisted edwards group law + tests * Add support for jubjub and bandersnatch #162 * test twisted edwards scalar mul
2021-12-29 00:54:17 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Elliptic curve arithmetic G2
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
# ("tests/math/t_ec_shortw_prj_g2_add_double_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_sanity_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_distri_bn254_snarks.nim", false),
("tests/math/t_ec_shortw_prj_g2_mul_vs_ref_bn254_snarks.nim", false),
("tests/math/t_ec_shortw_prj_g2_mixed_add_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_add_double_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_sanity_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_distri_bls12_381.nim", false),
("tests/math/t_ec_shortw_prj_g2_mul_vs_ref_bls12_381.nim", false),
("tests/math/t_ec_shortw_prj_g2_mixed_add_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_add_double_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_sanity_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_distri_bls12_377.nim", false),
("tests/math/t_ec_shortw_prj_g2_mul_vs_ref_bls12_377.nim", false),
("tests/math/t_ec_shortw_prj_g2_mixed_add_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_add_double_bw6_761.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_sanity_bw6_761.nim", false),
# ("tests/math/t_ec_shortw_prj_g2_mul_distri_bw6_761.nim", false),
("tests/math/t_ec_shortw_prj_g2_mul_vs_ref_bw6_761.nim", false),
("tests/math/t_ec_shortw_prj_g2_mixed_add_bw6_761.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_add_double_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_sanity_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_distri_bn254_snarks.nim", false),
("tests/math/t_ec_shortw_jac_g2_mul_vs_ref_bn254_snarks.nim", false),
("tests/math/t_ec_shortw_jac_g2_mixed_add_bn254_snarks.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_add_double_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_sanity_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_distri_bls12_381.nim", false),
("tests/math/t_ec_shortw_jac_g2_mul_vs_ref_bls12_381.nim", false),
("tests/math/t_ec_shortw_jac_g2_mixed_add_bls12_381.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_add_double_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_sanity_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_distri_bls12_377.nim", false),
("tests/math/t_ec_shortw_jac_g2_mul_vs_ref_bls12_377.nim", false),
("tests/math/t_ec_shortw_jac_g2_mixed_add_bls12_377.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_add_double_bw6_761.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_sanity_bw6_761.nim", false),
# ("tests/math/t_ec_shortw_jac_g2_mul_distri_bw6_761.nim", false),
("tests/math/t_ec_shortw_jac_g2_mul_vs_ref_bw6_761.nim", false),
("tests/math/t_ec_shortw_jac_g2_mixed_add_bw6_761.nim", false),
BW6-761 part 1 (#100) * Add Fp, Fp2, Fp6 support for BW6-761 * Add G1 for BW6-761 * Prepare to support G2 twists on the same field as G1 * Remove a useless dependent type for lines * Implement G2 for BW6-761 * Fix Line leftover
2020-10-09 05:51:47 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Elliptic curve arithmetic vs Sagemath
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_ec_frobenius.nim", false),
("tests/math/t_ec_sage_bn254_nogami.nim", false),
("tests/math/t_ec_sage_bn254_snarks.nim", false),
("tests/math/t_ec_sage_bls12_377.nim", false),
("tests/math/t_ec_sage_bls12_381.nim", false),
Pasta curves (#191) * Pasta curves field arithmetic * implement elliptic curve arith for the Pasta curves
2022-04-26 22:58:48 +00:00
("tests/math/t_ec_sage_pallas.nim", false),
("tests/math/t_ec_sage_vesta.nim", false),
Fuzzing campaign fixes (#58) * Add test case for #30 - Euler's criterion doesn't return 1 for a square * Detect #42 in the test suite * Detect #43 in the test suite * comment in sqrt tests * Add #67 to the anti-regression suite * Add #61 to the anti-regression suite * Add #62 to anti-regression suite * Add #60 to the anti-regression suite * Add #64 to the test suite * Add #65 - case 1 * Add #65 case 2 * Add #65 case 3 * Add debug check to isSquare/Euler's Criterion/Legendre Symbol * Make sure our primitives are correct * For now deactivate montySquare CIOS fix #61 #62 * Narrow down #42 and #43 to powinv on 32-bit * Detect #42 #43 at the fast squaring level * More #42, #43 tests, Use multiplication instead of squaring as a temporary workaround, see https://github.com/mratsim/constantine/issues/68 * Prevent regression of #67 now that squaring is "fixed"
2020-06-22 23:27:40 +00:00
# Edge cases highlighted by past bugs
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_ec_shortw_prj_edge_cases.nim", false),
Optimized subgroup checks and cofactor clearing (#169) * Move cofactor clearing to dedicated per-curve subgroups file * Add BLS12-381 fast subgroup checks * Implement fast cofactor clearing for BN254_snarks * Add fast subgroup check to BN254Snarks * add BLS12_377 optimized cofactor and subgroup functions * Add BN254_Nogami * Add GT-subgroup tests * Use the new subgroup checks for Eth1 EVM precompiles
2022-01-03 13:12:58 +00:00
# Subgroups and cofactors
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_ec_subgroups_bn254_nogami.nim", false),
("tests/math/t_ec_subgroups_bn254_snarks.nim", false),
("tests/math/t_ec_subgroups_bls12_377.nim", false),
("tests/math/t_ec_subgroups_bls12_381.nim", false),
Optimized subgroup checks and cofactor clearing (#169) * Move cofactor clearing to dedicated per-curve subgroups file * Add BLS12-381 fast subgroup checks * Implement fast cofactor clearing for BN254_snarks * Add fast subgroup check to BN254Snarks * add BLS12_377 optimized cofactor and subgroup functions * Add BN254_Nogami * Add GT-subgroup tests * Use the new subgroup checks for Eth1 EVM precompiles
2022-01-03 13:12:58 +00:00
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_pairing_bn254_nogami_gt_subgroup.nim", false),
("tests/math/t_pairing_bn254_snarks_gt_subgroup.nim", false),
("tests/math/t_pairing_bls12_377_gt_subgroup.nim", false),
("tests/math/t_pairing_bls12_381_gt_subgroup.nim", false),
("tests/math/t_pairing_bw6_761_gt_subgroup.nim", false),
Optimized subgroup checks and cofactor clearing (#169) * Move cofactor clearing to dedicated per-curve subgroups file * Add BLS12-381 fast subgroup checks * Implement fast cofactor clearing for BN254_snarks * Add fast subgroup check to BN254Snarks * add BLS12_377 optimized cofactor and subgroup functions * Add BN254_Nogami * Add GT-subgroup tests * Use the new subgroup checks for Eth1 EVM precompiles
2022-01-03 13:12:58 +00:00
Naive pairings + Naive cofactor clearing (#82) * Pairing - initial commit - line functions - sparse Fp12 functions * Small fixes: - Line parametrized by twist for generic algorithm - Add a conjugate operator for quadratic extensions - Have frobenius use it - Create an Affine coordinate type for elliptic curve * Implement (failing) pairing test * Stash pairing debug session, temp switch Fp12 over Fp4 * Proper naive pairing on BLS12-381 * Frobenius map * Implement naive pairing for BN curves * Add pairing tests to CI + reduce time spent on lower-level tests * Test without assembler in Github Actions + less base layers test iterations
2020-09-21 21:24:00 +00:00
# Pairing
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
# ("tests/math/t_pairing_bls12_377_line_functions.nim", false),
# ("tests/math/t_pairing_bls12_381_line_functions.nim", false),
("tests/math/t_pairing_mul_fp12_by_lines.nim", false),
("tests/math/t_pairing_cyclotomic_subgroup.nim", false),
("tests/math/t_pairing_bn254_nogami_optate.nim", false),
("tests/math/t_pairing_bn254_snarks_optate.nim", false),
("tests/math/t_pairing_bls12_377_optate.nim", false),
("tests/math/t_pairing_bls12_381_optate.nim", false),
Add multipairing for BN curves (#194)
2022-05-08 17:01:23 +00:00
# Multi-Pairing
# ----------------------------------------------------------
("tests/math/t_pairing_bn254_nogami_multi.nim", false),
("tests/math/t_pairing_bn254_snarks_multi.nim", false),
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_pairing_bls12_381_multi.nim", false),
Fr: Finite Field parametrized by the curve order (#115) * Introduce Fr type: finite field over curve order. Need workaround for https://github.com/nim-lang/Nim/issues/16774 * Split curve properties into core and derived * Attach field properties to an instantiated field instead of the curve enum * Workaround https://github.com/nim-lang/Nim/issues/14021, yet another "working with types in macros" is difficult https://github.com/nim-lang/RFCs/issues/44 * Implement finite field over prime order of a curve subgroup * skip OpenSSL tests on windows
2021-01-21 23:09:52 +00:00
# Prime order fields
# ----------------------------------------------------------
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
("tests/math/t_fr.nim", false),
Productionize: move protocols top-level vs backend (#179) * Productionize: move protocols top-level vs backend * fix path * import fix * the last one * benches as well
2022-02-21 00:04:53 +00:00
# Hashing vs OpenSSL
# ----------------------------------------------------------
("tests/t_hash_sha256_vs_openssl.nim", true), # skip OpenSSL tests on Windows
Tests for the eth2 BLS signature protocol (BLS12-381, pubkeys G1, signatures G2) using low-level primitives (#164)
2021-08-15 09:41:46 +00:00
BN254 - Hash-to-Curve (SVDW method) (#190) * Hash to BN254-Snarks * Test SVDW code path with old v7 vectors for BLS12-381 * add benches
2022-04-26 19:24:07 +00:00
# Hashing to elliptic curves
# ----------------------------------------------------------
("tests/t_hash_to_field.nim", false),
("tests/t_hash_to_curve_random.nim", false),
("tests/t_hash_to_curve.nim", false),
Add ChaCha20 stream cipher
2022-03-02 00:18:47 +00:00
# Ciphers
# ----------------------------------------------------------
("tests/t_cipher_chacha20.nim", false),
Poly1305 Message Authentication Code (#186) * Groundwork for Poly1305 MAC * Implement fast reduction for Poly1305 * don't import assembly files when compiling without assembly
2022-03-05 22:39:24 +00:00
# Message Authentication Code
# ----------------------------------------------------------
("tests/t_mac_poly1305.nim", false),
Tests for the eth2 BLS signature protocol (BLS12-381, pubkeys G1, signatures G2) using low-level primitives (#164)
2021-08-15 09:41:46 +00:00
# Protocols
# ----------------------------------------------------------
Productionize: move protocols top-level vs backend (#179) * Productionize: move protocols top-level vs backend * fix path * import fix * the last one * benches as well
2022-02-21 00:04:53 +00:00
("tests/t_ethereum_evm_precompiles.nim", false),
BLS signatures for Ethereum (BLS sig on BLS12-381 G2 with SHA256) (#183) * Finally add the (Ethereum) bls signatures (on BLS12-381 G2) * fix test path and remove old low-level signature test
2022-02-26 20:22:34 +00:00
("tests/t_blssig_pop_on_bls12381_g2.nim", false),
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
]
Github Action CI (without GMP) (#29) * Github Action CI (without GMP) * Deactivate MacOS, spurious failures: https://github.com/actions/virtual-environments/issues/841 * force install with nimble * Add badge * Don"t include Nim 1.2.x https://github.com/mratsim/constantine/pull/20#issuecomment-646327952 * Action branch mistake * Add back OSX? https://github.com/actions/virtual-environments/issues/841, https://github.com/actions/virtual-environments/issues/969 * fix MacOS target * comment out RDTSC on i386 * Add initialization canaries * Add more verbose output to debug windows failures * spurious windows i386 test * For now only activate Linux and mac * missed include
2020-06-19 20:08:15 +00:00
# For temporary (hopefully) investigation that can only be reproduced in CI
const useDebug = [
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
"tests/math/t_bigints.nim",
"tests/math/t_hash_sha256_vs_openssl.nim",
SHA256 Hash function
2020-12-15 18:18:36 +00:00
]
# Tests that uses sequences require Nim GC, stack scanning and nil pointer passed to openarray
# In particular the tests that uses the json test vectors, don't sanitize them.
# we do use gc:none to help
const skipSanitizers = [
Don't dump all in "backend" (#184) * backend -> math * towers -> extension fields * move ISA and compiler specific code out of math/ * fix export
2022-02-27 00:49:08 +00:00
"tests/math/t_ec_sage_bn254_nogami.nim",
"tests/math/t_ec_sage_bn254_snarks.nim",
"tests/math/t_ec_sage_bls12_377.nim",
"tests/math/t_ec_sage_bls12_381.nim",
BN254 - Hash-to-Curve (SVDW method) (#190) * Hash to BN254-Snarks * Test SVDW code path with old v7 vectors for BLS12-381 * add benches
2022-04-26 19:24:07 +00:00
"tests/t_hash_to_field.nim",
"tests/t_hash_to_curve.nim",
"tests/t_hash_to_curve_random.nim"
Github Action CI (without GMP) (#29) * Github Action CI (without GMP) * Deactivate MacOS, spurious failures: https://github.com/actions/virtual-environments/issues/841 * force install with nimble * Add badge * Don"t include Nim 1.2.x https://github.com/mratsim/constantine/pull/20#issuecomment-646327952 * Action branch mistake * Add back OSX? https://github.com/actions/virtual-environments/issues/841, https://github.com/actions/virtual-environments/issues/969 * fix MacOS target * comment out RDTSC on i386 * Add initialization canaries * Add more verbose output to debug windows failures * spurious windows i386 test * For now only activate Linux and mac * missed include
2020-06-19 20:08:15 +00:00
]
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
when defined(windows):
# UBSAN is not available on mingw
const sanitizers = ""
else:
const sanitizers =
" --passC:-fsanitize=undefined --passL:-fsanitize=undefined" &
" --passC:-fno-sanitize-recover" & # Enforce crash on undefined behaviour
" --gc:none" # The conservative stack scanning of Nim default GC triggers, alignment UB and stack-buffer-overflow check.
# " --passC:-fsanitize=address --passL:-fsanitize=address" & # Requires too much stack for the inline assembly
SHA256 Hash function
2020-12-15 18:18:36 +00:00
Github Action CI (without GMP) (#29) * Github Action CI (without GMP) * Deactivate MacOS, spurious failures: https://github.com/actions/virtual-environments/issues/841 * force install with nimble * Add badge * Don"t include Nim 1.2.x https://github.com/mratsim/constantine/pull/20#issuecomment-646327952 * Action branch mistake * Add back OSX? https://github.com/actions/virtual-environments/issues/841, https://github.com/actions/virtual-environments/issues/969 * fix MacOS target * comment out RDTSC on i386 * Add initialization canaries * Add more verbose output to debug windows failures * spurious windows i386 test * For now only activate Linux and mac * missed include
2020-06-19 20:08:15 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Helper functions
# ----------------------------------------------------------------
SHA256 Hash function
2020-12-15 18:18:36 +00:00
proc clearParallelBuild() =
exec "> " & buildParallel
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
proc test(flags, path: string, commandFile = false) =
# commandFile should be a "file" but Nimscript doesn't support IO
initial commit
2018-07-24 14:52:18 +00:00
if not dirExists "build":
mkDir "build"
Set up CI with Azure Pipelines (#13) * Set up CI with Azure Pipelines [skip ci] * Add task for testing without GMP * Add C++ testing + no GMP on windows * Add the Nim wrapper for GMP to Azure build * Add Azure badge * Fix nimble test tasks * Workaround windows path in Azure * Fix nim binaries path and mingw on 32-bit * add stew test dependency * Fix nim/nimble path * disable GMP tests on windows
2020-02-23 17:27:26 +00:00
# Compilation language is controlled by WEAVE_TEST_LANG
var lang = "c"
if existsEnv"TEST_LANG":
lang = getEnv"TEST_LANG"
Internals refactor + renewed focus on perf (#17) * Lay out the refactoring objectives and tradeoffs * Refactor the 32 and 64-bit primitives [skip ci] * BigInts and Modular BigInts compile * Make the bigints test compile * Fix modular reduction * Fix reduction tests vs GMP * Implement montegomery mul, pow, inverse, WIP finite field compilation * Make FiniteField compile * Fix exponentiation compilation * Fix Montgomery magic constant computation for 2^64 words * Fix typo in non-optimized CIOS - passing finite fields IO tests * Add limbs comparisons [skip ci] * Fix on precomputation of the Montgomery magic constant * Passing all tests including 𝔽p2 * modular addition, the test for mersenne prime was wrong * update benches * Fix "nimble test" + typo on out-of-place field addition * bigint division, normalization is needed: https://travis-ci.com/github/mratsim/constantine/jobs/298359743 * missing conversion in subborrow non-x86 fallback - https://travis-ci.com/github/mratsim/constantine/jobs/298359744 * Fix little-endian serialization * Constantine32 flag to run 32-bit constantine on 64-bit machines * IO Field test, ensure that BaseType is used instead of uint64 when the prime can field in uint32 * Implement proper addcarry and subborrow fallback for the compile-time VM * Fix export issue when the logical wordbitwidth == physical wordbitwidth - passes all tests (32-bit and 64-bit) * Fix uint128 on ARM * Fix C++ conditional copy and ARM addcarry/subborrow * Add investigation for SIGFPE in Travis * Fix debug display for unsafeDiv2n1n * multiplexer typo * moveMem bug in glibc of Ubuntu 16.04? * Was probably missing an early clobbered register annotation on conditional mov * Note on Montgomery-friendly moduli * Strongly suspect a GCC before GCC 7 codegen bug (https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87139) * hex conversion was (for debugging) not taking requested order into account + inlining comment * Use 32-bit limbs on ARM64, uint128 builtin __udivti4 bug? * Revert "Use 32-bit limbs on ARM64, uint128 builtin __udivti4 bug?" This reverts commit 087f9aa7fb40bbd058d05cbd8eec7fc082911f49. * Fix subborrow fallback for non-x86 (need to maks the borrow)
2020-03-16 15:33:51 +00:00
var cc = ""
if existsEnv"CC":
cc = " --cc:" & getEnv"CC"
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
var flags = flags
when not defined(windows):
# Not available in MinGW https://github.com/libressl-portable/portable/issues/54
flags &= " --passC:-fstack-protector-all"
Consolidate output folders of bench and testsuite
2020-08-22 21:00:05 +00:00
let command = "nim " & lang & cc & " " & flags &
" --verbosity:0 --outdir:build/testsuite -r --hints:off --warnings:off " &
" --nimcache:nimcache/" & path & " " &
path
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
if not commandFile:
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
echo "\n=============================================================================================="
SHA256 Hash function
2020-12-15 18:18:36 +00:00
echo "Running [flags:", flags, "] ", path
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
echo "=============================================================================================="
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
exec command
else:
exec "echo \'" & command & "\' >> " & buildParallel
Square Root & Inversion addition chains - 20% perf increase (#132) * Addition chain for sqrt BLS12-381: 20% perf improvement * sqrt addchain for BN254_Snarks - 20% perf improvement as well * Fix operation count [skip ci] * BLS12-377 sqrt - 10% perf improvement * sqrt addition chain for BW6-761 - 6% speedup * BN254_Nogami inversion addchain * sqrt addchain for BN254_Nogami * Inversion addchain for BLS12-377 * inversion ddition chain for BW6-761
2021-01-23 19:55:40 +00:00
exec "echo \"------------------------------------------------------\""
initial commit
2018-07-24 14:52:18 +00:00
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
proc buildBench(benchName: string, compiler = "", useAsm = true, run = false) =
Add benches to the test suite to prevent forgetting about updating them
2020-04-15 17:46:25 +00:00
if not dirExists "build":
mkDir "build"
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
let runFlag = if run: " -r "
else: " "
Add benches to the test suite to prevent forgetting about updating them
2020-04-15 17:46:25 +00:00
var cc = ""
if compiler != "":
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
cc = "--cc:" & compiler
if not useAsm:
FpDbl revisited (#144) - 7% perf improvement everywhere, up to 30% in double-width primitives * reorg mul -> limbs_double_width, ConstantineASM CttASM * Implement squaring specialized scalar path (22% faster than mul) * Implement "portable" assembly for squaring * stash part of the changes * Reorg montgomery reduction - prepare to introduce Comba optimization * Implement comba Montgomery reduce (but it's slower!) * rename t -> a * 30% performance improvement by avoiding toOpenArray! * variable renaming * Fix 32-bit imports * slightly better assembly for sub2x * There is an annoying bottleneck * use out-of-place Fp assembly instead of in-place * diffAlias is unneeded now * cosmetic * speedup fpDbl sub by 20% * Fix Fp2 -> Fp6 -> Fp12 towering. It seems 5% faster * Stash ADCX/ADOX squaring
2021-02-01 02:52:27 +00:00
cc &= " -d:CttASM=false"
Add benches to the test suite to prevent forgetting about updating them
2020-04-15 17:46:25 +00:00
exec "nim c " & cc &
Consolidate output folders of bench and testsuite
2020-08-22 21:00:05 +00:00
" -d:danger --verbosity:0 -o:build/bench/" & benchName & "_" & compiler & "_" & (if useAsm: "useASM" else: "noASM") &
" --nimcache:nimcache/" & benchName & "_" & compiler & "_" & (if useAsm: "useASM" else: "noASM") &
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
runFlag & "--hints:off --warnings:off benchmarks/" & benchName & ".nim"
proc runBench(benchName: string, compiler = "", useAsm = true) =
buildBench(benchName, compiler, useAsm, run = true)
Add benches to the test suite to prevent forgetting about updating them
2020-04-15 17:46:25 +00:00
SHA256 Hash function
2020-12-15 18:18:36 +00:00
proc runTests(requireGMP: bool, dumpCmdFile = false, test32bit = false, testASM = true) =
for td in testDesc:
if not(td.useGMP and not requireGMP):
var flags = ""
if not testASM:
FpDbl revisited (#144) - 7% perf improvement everywhere, up to 30% in double-width primitives * reorg mul -> limbs_double_width, ConstantineASM CttASM * Implement squaring specialized scalar path (22% faster than mul) * Implement "portable" assembly for squaring * stash part of the changes * Reorg montgomery reduction - prepare to introduce Comba optimization * Implement comba Montgomery reduce (but it's slower!) * rename t -> a * 30% performance improvement by avoiding toOpenArray! * variable renaming * Fix 32-bit imports * slightly better assembly for sub2x * There is an annoying bottleneck * use out-of-place Fp assembly instead of in-place * diffAlias is unneeded now * cosmetic * speedup fpDbl sub by 20% * Fix Fp2 -> Fp6 -> Fp12 towering. It seems 5% faster * Stash ADCX/ADOX squaring
2021-02-01 02:52:27 +00:00
flags &= " -d:CttASM=false"
SHA256 Hash function
2020-12-15 18:18:36 +00:00
if test32bit:
flags &= " -d:Constantine32"
if td.path in useDebug:
flags &= " -d:debugConstantine"
if td.path notin skipSanitizers:
flags &= sanitizers
test flags, td.path, dumpCmdFile
hash-to-curve BLS12-381 perf (#163) * fp square noasm split from non-4 non-6 limbs fallback (40% speedup) * optimized cofactor clearing for BLS12-381 G2 * Support jacobian isogenies and point_add on isogenies * fuse addition and isogeny map * {.noInit.} and sparseMul * poly_eval_horner init * dedicated invsqrt + cleanup square root file * hash to field: reduce copy overhead and don't return arrays * h2c isogeny jacobian reuse pow 3 precomputed value * Fix sqrt bench
2021-08-14 19:01:50 +00:00
proc buildAllBenches(useAsm = true) =
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
echo "\n\n------------------------------------------------------\n"
echo "Building benchmarks to ensure they stay relevant ..."
hash-to-curve BLS12-381 perf (#163) * fp square noasm split from non-4 non-6 limbs fallback (40% speedup) * optimized cofactor clearing for BLS12-381 G2 * Support jacobian isogenies and point_add on isogenies * fuse addition and isogeny map * {.noInit.} and sparseMul * poly_eval_horner init * dedicated invsqrt + cleanup square root file * hash to field: reduce copy overhead and don't return arrays * h2c isogeny jacobian reuse pow 3 precomputed value * Fix sqrt bench
2021-08-14 19:01:50 +00:00
buildBench("bench_fp", useAsm = useAsm)
buildBench("bench_fp_double_precision", useAsm = useAsm)
buildBench("bench_fp2", useAsm = useAsm)
buildBench("bench_fp6", useAsm = useAsm)
buildBench("bench_fp12", useAsm = useAsm)
buildBench("bench_ec_g1", useAsm = useAsm)
buildBench("bench_ec_g2", useAsm = useAsm)
buildBench("bench_pairing_bls12_377", useAsm = useAsm)
buildBench("bench_pairing_bls12_381", useAsm = useAsm)
buildBench("bench_pairing_bn254_nogami", useAsm = useAsm)
buildBench("bench_pairing_bn254_snarks", useAsm = useAsm)
buildBench("bench_summary_bls12_377", useAsm = useAsm)
buildBench("bench_summary_bls12_381", useAsm = useAsm)
buildBench("bench_summary_bn254_nogami", useAsm = useAsm)
buildBench("bench_summary_bn254_snarks", useAsm = useAsm)
buildBench("bench_sha256", useAsm = useAsm)
buildBench("bench_hash_to_curve", useAsm = useAsm)
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
echo "All benchmarks compile successfully."
Fr: Finite Field parametrized by the curve order (#115) * Introduce Fr type: finite field over curve order. Need workaround for https://github.com/nim-lang/Nim/issues/16774 * Split curve properties into core and derived * Attach field properties to an instantiated field instead of the curve enum * Workaround https://github.com/nim-lang/Nim/issues/14021, yet another "working with types in macros" is difficult https://github.com/nim-lang/RFCs/issues/44 * Implement finite field over prime order of a curve subgroup * skip OpenSSL tests on windows
2021-01-21 23:09:52 +00:00
Self-contained bindings generation (#196) * First draft at bindings generation * finite field bindings PoC * support openarray, export NimMain * PoC extension fields and elliptic curve bindings * Pasta * expose more bindings, remove nimZeroMem, remove tracer when unused, codegen name_mangling`gensym issue * workaround bad C gensym codegen with {.inline.} pragma in non-dirty template nested in generic proc instantiated by template
2022-08-06 17:05:54 +00:00
proc genBindings(bindingsName, prefixNimMain: string) =
proc compile(libName: string, flags = "") =
# -d:danger to avoid boundsCheck, overflowChecks that would trigger exceptions or allocations in a crypto library.
# Those are internally guaranteed at compile-time by fixed-sized array
# and checked at runtime with an appropriate error code if any for user-input.
# -gc:arc Constantine stack allocates everything. Inputs are through unmanaged ptr+len.
# In the future, Constantine might use:
# - heap-allocated sequences and objects manually managed or managed by destructors for multithreading.
# - heap-allocated strings for hex-string or decimal strings
exec "nim c -f " & flags & " --noMain -d:danger --app:lib --gc:arc " &
" --nimMainPrefix:" & prefixNimMain &
" --out:" & libName & " --outdir:bindings/generated " &
" --nimcache:nimcache/bindings/" & bindingsName &
" bindings/" & bindingsName & ".nim"
when defined(windows):
compile bindingsName & ".dll"
elif defined(macosx):
compile "lib" & bindingsName & ".dylib.arm", "--cpu:arm64 -l:'-target arm64-apple-macos11' -t:'-target arm64-apple-macos11'"
compile "lib" & bindingsName & ".dylib.x64", "--cpu:amd64 -l:'-target x86_64-apple-macos10.12' -t:'-target x86_64-apple-macos10.12'"
exec "lipo bindings/generated/lib" & bindingsName & ".dylib.arm " &
" bindings/generated/lib" & bindingsName & ".dylib.x64 " &
" -output bindings/generated/lib" & bindingsName & ".dylib -create"
else:
compile "lib" & bindingsName & ".so"
proc genHeaders(bindingsName: string) =
exec "nim c -r -d:release -d:CttGenerateHeaders " &
" --out:" & bindingsName & "_gen_header.exe --outdir:bindings/generated " &
" --nimcache:nimcache/bindings/" & bindingsName & "_header" &
" bindings/" & bindingsName & ".nim"
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
# Tasks
# ----------------------------------------------------------------
Self-contained bindings generation (#196) * First draft at bindings generation * finite field bindings PoC * support openarray, export NimMain * PoC extension fields and elliptic curve bindings * Pasta * expose more bindings, remove nimZeroMem, remove tracer when unused, codegen name_mangling`gensym issue * workaround bad C gensym codegen with {.inline.} pragma in non-dirty template nested in generic proc instantiated by template
2022-08-06 17:05:54 +00:00
task bindings, "Generate Constantine bindings":
genBindings("constantine_bls12_381", "ctt_bls12381_")
genHeaders("constantine_bls12_381")
genBindings("constantine_pasta", "ctt_pasta_")
genHeaders("constantine_pasta")
initial commit
2018-07-24 14:52:18 +00:00
task test, "Run all tests":
Set up CI with Azure Pipelines (#13) * Set up CI with Azure Pipelines [skip ci] * Add task for testing without GMP * Add C++ testing + no GMP on windows * Add the Nim wrapper for GMP to Azure build * Add Azure badge * Fix nimble test tasks * Workaround windows path in Azure * Fix nim binaries path and mingw on 32-bit * add stew test dependency * Fix nim/nimble path * disable GMP tests on windows
2020-02-23 17:27:26 +00:00
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
runTests(requireGMP = true)
Initial support for Elliptic Curve (#24) * Elliptic curve and Twisted curve templates - initial commit * Support EC Add on G2 (Sextic Twisted curve for BN and BLS12 families) * Refactor the config parser to prepare for elliptic coefficient support * Add elliptic curve parameter for BN254 (Snarks), BLS12-381 and Zexe curve BLS12-377 * Add accessors to curve parameters * Allow computing the right-hand-side of of Weierstrass equation "y² = x³ + a x + b" * Randomized test infrastructure for elliptic curves * Start a testing suite on ellptic curve addition (failing) * detail projective addition * Fix EC addition test (forgot initializing Z=1 and that there ar emultiple infinity points) * Test with random Z coordinate + add elliptic curve test to test suite * fix reference to the (deactivated) addchain inversion for BN curves [skip ci] * .nims file leftover [skip ci]
2020-04-13 17:25:59 +00:00
Skip 32-bit tests on 64-bit machines (too long)
2020-10-31 13:51:17 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
SHA256 Hash function
2020-12-15 18:18:36 +00:00
# runTests(requireGMP = true, test32bit = true)
Initial support for Elliptic Curve (#24) * Elliptic curve and Twisted curve templates - initial commit * Support EC Add on G2 (Sextic Twisted curve for BN and BLS12 families) * Refactor the config parser to prepare for elliptic coefficient support * Add elliptic curve parameter for BN254 (Snarks), BLS12-381 and Zexe curve BLS12-377 * Add accessors to curve parameters * Allow computing the right-hand-side of of Weierstrass equation "y² = x³ + a x + b" * Randomized test infrastructure for elliptic curves * Start a testing suite on ellptic curve addition (failing) * detail projective addition * Fix EC addition test (forgot initializing Z=1 and that there ar emultiple infinity points) * Test with random Z coordinate + add elliptic curve test to test suite * fix reference to the (deactivated) addchain inversion for BN curves [skip ci] * .nims file leftover [skip ci]
2020-04-13 17:25:59 +00:00
Fix ARM bench and ignore Windows 32-bit bench
2020-04-15 19:28:37 +00:00
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
buildAllBenches()
Add benches to the test suite to prevent forgetting about updating them
2020-04-15 17:46:25 +00:00
Fuzz fix - non-unique modular representation after Assembly negate (#137) * Fix #114 - Negating 0 left the prime modulus, which is working most of the time for everything except for comparison. (also somehow triggers and workaround weird compiler bug where exceptions tracking is activated in macros and all the curve enums were stringified as their ordinal value) * https://github.com/mratsim/constantine/issues/136 was also fixed, add to anti-regression * add comment in test * Fix the pure Nim fallback as well
2021-01-24 11:35:27 +00:00
task test_no_assembler, "Run all tests":
# -d:testingCurves is configured in a *.nim.cfg for convenience
runTests(requireGMP = true, testASM = false)
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# runTests(requireGMP = true, test32bit = true)
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
hash-to-curve BLS12-381 perf (#163) * fp square noasm split from non-4 non-6 limbs fallback (40% speedup) * optimized cofactor clearing for BLS12-381 G2 * Support jacobian isogenies and point_add on isogenies * fuse addition and isogeny map * {.noInit.} and sparseMul * poly_eval_horner init * dedicated invsqrt + cleanup square root file * hash to field: reduce copy overhead and don't return arrays * h2c isogeny jacobian reuse pow 3 precomputed value * Fix sqrt bench
2021-08-14 19:01:50 +00:00
buildAllBenches(useASM = false)
Fuzz fix - non-unique modular representation after Assembly negate (#137) * Fix #114 - Negating 0 left the prime modulus, which is working most of the time for everything except for comparison. (also somehow triggers and workaround weird compiler bug where exceptions tracking is activated in macros and all the curve enums were stringified as their ordinal value) * https://github.com/mratsim/constantine/issues/136 was also fixed, add to anti-regression * add comment in test * Fix the pure Nim fallback as well
2021-01-24 11:35:27 +00:00
Set up CI with Azure Pipelines (#13) * Set up CI with Azure Pipelines [skip ci] * Add task for testing without GMP * Add C++ testing + no GMP on windows * Add the Nim wrapper for GMP to Azure build * Add Azure badge * Fix nimble test tasks * Workaround windows path in Azure * Fix nim binaries path and mingw on 32-bit * add stew test dependency * Fix nim/nimble path * disable GMP tests on windows
2020-02-23 17:27:26 +00:00
task test_no_gmp, "Run tests that don't require GMP":
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
runTests(requireGMP = false)
Add optimized squaring (~15% speedup) (#18) * Add optimized squaring (~15% speedup) * avoid repetitions in tests
2020-03-17 21:04:37 +00:00
SHA256 Hash function
2020-12-15 18:18:36 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# runTests(requireGMP = true, test32bit = true)
Github Action CI (without GMP) (#29) * Github Action CI (without GMP) * Deactivate MacOS, spurious failures: https://github.com/actions/virtual-environments/issues/841 * force install with nimble * Add badge * Don"t include Nim 1.2.x https://github.com/mratsim/constantine/pull/20#issuecomment-646327952 * Action branch mistake * Add back OSX? https://github.com/actions/virtual-environments/issues/841, https://github.com/actions/virtual-environments/issues/969 * fix MacOS target * comment out RDTSC on i386 * Add initialization canaries * Add more verbose output to debug windows failures * spurious windows i386 test * For now only activate Linux and mac * missed include
2020-06-19 20:08:15 +00:00
Fix ARM bench and ignore Windows 32-bit bench
2020-04-15 19:28:37 +00:00
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
buildAllBenches()
30% faster constant-time inversion
2020-03-20 22:03:52 +00:00
Nim 1.6 in CI (#170) * try 1.6 CI * Try CI with 1.6 and windows. * Bend the knee * have fun debugging CI * have fun debugging CI * more CI spam * branch -> nim_version * fight or flight * properly detect windows * Fix galore * :snake: :snake: snake: * meh give up on parallelizing windows and dealing with windows PATH issues * ¯\_ (ツ)_/¯
2022-02-20 22:44:00 +00:00
task test_no_gmp_no_assembler, "Run tests that don't require GMP using a pure Nim backend":
# -d:testingCurves is configured in a *.nim.cfg for convenience
runTests(requireGMP = false, testASM = false)
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# runTests(requireGMP = true, test32bit = true)
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
buildAllBenches()
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
task test_parallel, "Run all tests in parallel (via GNU parallel)":
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
clearParallelBuild()
runTests(requireGMP = true, dumpCmdFile = true)
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
exec "parallel --keep-order --group < " & buildParallel
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# clearParallelBuild()
# runTests(requireGMP = true, dumpCmdFile = true, test32bit = true)
# exec "parallel --keep-order --group < " & buildParallel
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
# Now run the benchmarks
#
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# Benchmarks compile
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
# ignore Windows 32-bit for the moment
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
buildAllBenches()
Use GMP and GNU parallel in GIthub Actions (#63) * Use GMP and GNU parallel in GIthub Actions * try gmp:i386 * Don't do GMP tests on i386
2020-06-20 17:46:30 +00:00
Assembly backend (#69) * Proof-of-Concept Assembly code generator * Tag inline per procedure so we can easily track the tradeoff on tower fields * Implement Assembly for modular addition (but very curious off-by-one) * Fix off-by one for moduli with non msb set * Stash (super fast) alternative but still off by carry * Fix GCC optimizing ASM away * Save 1 register to allow compiling for BLS12-381 (in the GMP test) * The compiler cannot find enough registers if the ASM file is not compiled with -O3 * Add modsub * Add field negation * Implement no-carry Assembly optimized field multiplication * Expose UseX86ASM to the EC benchmark * omit frame pointer to save registers instead of hardcoding -O3. Also ensure early clobber constraints for Clang * Prepare for assembly fallback * Implement fallback for CPU that don't support ADX and BMI2 * Add CPU runtime detection * Update README closes #66 * Remove commented out code
2020-07-24 20:02:30 +00:00
task test_parallel_no_assembler, "Run all tests (without macro assembler) in parallel (via GNU parallel)":
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
clearParallelBuild()
runTests(requireGMP = true, dumpCmdFile = true, testASM = false)
Assembly backend (#69) * Proof-of-Concept Assembly code generator * Tag inline per procedure so we can easily track the tradeoff on tower fields * Implement Assembly for modular addition (but very curious off-by-one) * Fix off-by one for moduli with non msb set * Stash (super fast) alternative but still off by carry * Fix GCC optimizing ASM away * Save 1 register to allow compiling for BLS12-381 (in the GMP test) * The compiler cannot find enough registers if the ASM file is not compiled with -O3 * Add modsub * Add field negation * Implement no-carry Assembly optimized field multiplication * Expose UseX86ASM to the EC benchmark * omit frame pointer to save registers instead of hardcoding -O3. Also ensure early clobber constraints for Clang * Prepare for assembly fallback * Implement fallback for CPU that don't support ADX and BMI2 * Add CPU runtime detection * Update README closes #66 * Remove commented out code
2020-07-24 20:02:30 +00:00
exec "parallel --keep-order --group < " & buildParallel
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# clearParallelBuild()
# runTests(requireGMP = true, dumpCmdFile = true, test32bit = true, testASM = false)
# exec "parallel --keep-order --group < " & buildParallel
Assembly backend (#69) * Proof-of-Concept Assembly code generator * Tag inline per procedure so we can easily track the tradeoff on tower fields * Implement Assembly for modular addition (but very curious off-by-one) * Fix off-by one for moduli with non msb set * Stash (super fast) alternative but still off by carry * Fix GCC optimizing ASM away * Save 1 register to allow compiling for BLS12-381 (in the GMP test) * The compiler cannot find enough registers if the ASM file is not compiled with -O3 * Add modsub * Add field negation * Implement no-carry Assembly optimized field multiplication * Expose UseX86ASM to the EC benchmark * omit frame pointer to save registers instead of hardcoding -O3. Also ensure early clobber constraints for Clang * Prepare for assembly fallback * Implement fallback for CPU that don't support ADX and BMI2 * Add CPU runtime detection * Update README closes #66 * Remove commented out code
2020-07-24 20:02:30 +00:00
# Now run the benchmarks
#
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# Benchmarks compile
Assembly backend (#69) * Proof-of-Concept Assembly code generator * Tag inline per procedure so we can easily track the tradeoff on tower fields * Implement Assembly for modular addition (but very curious off-by-one) * Fix off-by one for moduli with non msb set * Stash (super fast) alternative but still off by carry * Fix GCC optimizing ASM away * Save 1 register to allow compiling for BLS12-381 (in the GMP test) * The compiler cannot find enough registers if the ASM file is not compiled with -O3 * Add modsub * Add field negation * Implement no-carry Assembly optimized field multiplication * Expose UseX86ASM to the EC benchmark * omit frame pointer to save registers instead of hardcoding -O3. Also ensure early clobber constraints for Clang * Prepare for assembly fallback * Implement fallback for CPU that don't support ADX and BMI2 * Add CPU runtime detection * Update README closes #66 * Remove commented out code
2020-07-24 20:02:30 +00:00
# ignore Windows 32-bit for the moment
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
hash-to-curve BLS12-381 perf (#163) * fp square noasm split from non-4 non-6 limbs fallback (40% speedup) * optimized cofactor clearing for BLS12-381 G2 * Support jacobian isogenies and point_add on isogenies * fuse addition and isogeny map * {.noInit.} and sparseMul * poly_eval_horner init * dedicated invsqrt + cleanup square root file * hash to field: reduce copy overhead and don't return arrays * h2c isogeny jacobian reuse pow 3 precomputed value * Fix sqrt bench
2021-08-14 19:01:50 +00:00
buildAllBenches(useASM = false)
Assembly backend (#69) * Proof-of-Concept Assembly code generator * Tag inline per procedure so we can easily track the tradeoff on tower fields * Implement Assembly for modular addition (but very curious off-by-one) * Fix off-by one for moduli with non msb set * Stash (super fast) alternative but still off by carry * Fix GCC optimizing ASM away * Save 1 register to allow compiling for BLS12-381 (in the GMP test) * The compiler cannot find enough registers if the ASM file is not compiled with -O3 * Add modsub * Add field negation * Implement no-carry Assembly optimized field multiplication * Expose UseX86ASM to the EC benchmark * omit frame pointer to save registers instead of hardcoding -O3. Also ensure early clobber constraints for Clang * Prepare for assembly fallback * Implement fallback for CPU that don't support ADX and BMI2 * Add CPU runtime detection * Update README closes #66 * Remove commented out code
2020-07-24 20:02:30 +00:00
Use GMP and GNU parallel in GIthub Actions (#63) * Use GMP and GNU parallel in GIthub Actions * try gmp:i386 * Don't do GMP tests on i386
2020-06-20 17:46:30 +00:00
task test_parallel_no_gmp, "Run all tests in parallel (via GNU parallel)":
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
clearParallelBuild()
runTests(requireGMP = false, dumpCmdFile = true)
Use GMP and GNU parallel in GIthub Actions (#63) * Use GMP and GNU parallel in GIthub Actions * try gmp:i386 * Don't do GMP tests on i386
2020-06-20 17:46:30 +00:00
exec "parallel --keep-order --group < " & buildParallel
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# clearParallelBuild()
# runTests(requireGMP = false, dumpCmdFile = true, test32bit = true)
# exec "parallel --keep-order --group < " & buildParallel
Naive pairings + Naive cofactor clearing (#82) * Pairing - initial commit - line functions - sparse Fp12 functions * Small fixes: - Line parametrized by twist for generic algorithm - Add a conjugate operator for quadratic extensions - Have frobenius use it - Create an Affine coordinate type for elliptic curve * Implement (failing) pairing test * Stash pairing debug session, temp switch Fp12 over Fp4 * Proper naive pairing on BLS12-381 * Frobenius map * Implement naive pairing for BN curves * Add pairing tests to CI + reduce time spent on lower-level tests * Test without assembler in Github Actions + less base layers test iterations
2020-09-21 21:24:00 +00:00
# Now run the benchmarks
#
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# Benchmarks compile
Naive pairings + Naive cofactor clearing (#82) * Pairing - initial commit - line functions - sparse Fp12 functions * Small fixes: - Line parametrized by twist for generic algorithm - Add a conjugate operator for quadratic extensions - Have frobenius use it - Create an Affine coordinate type for elliptic curve * Implement (failing) pairing test * Stash pairing debug session, temp switch Fp12 over Fp4 * Proper naive pairing on BLS12-381 * Frobenius map * Implement naive pairing for BN curves * Add pairing tests to CI + reduce time spent on lower-level tests * Test without assembler in Github Actions + less base layers test iterations
2020-09-21 21:24:00 +00:00
# ignore Windows 32-bit for the moment
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
buildAllBenches()
Naive pairings + Naive cofactor clearing (#82) * Pairing - initial commit - line functions - sparse Fp12 functions * Small fixes: - Line parametrized by twist for generic algorithm - Add a conjugate operator for quadratic extensions - Have frobenius use it - Create an Affine coordinate type for elliptic curve * Implement (failing) pairing test * Stash pairing debug session, temp switch Fp12 over Fp4 * Proper naive pairing on BLS12-381 * Frobenius map * Implement naive pairing for BN curves * Add pairing tests to CI + reduce time spent on lower-level tests * Test without assembler in Github Actions + less base layers test iterations
2020-09-21 21:24:00 +00:00
task test_parallel_no_gmp_no_assembler, "Run all tests in parallel (via GNU parallel)":
# -d:testingCurves is configured in a *.nim.cfg for convenience
SHA256 Hash function
2020-12-15 18:18:36 +00:00
clearParallelBuild()
runTests(requireGMP = false, dumpCmdFile = true, testASM = false)
Naive pairings + Naive cofactor clearing (#82) * Pairing - initial commit - line functions - sparse Fp12 functions * Small fixes: - Line parametrized by twist for generic algorithm - Add a conjugate operator for quadratic extensions - Have frobenius use it - Create an Affine coordinate type for elliptic curve * Implement (failing) pairing test * Stash pairing debug session, temp switch Fp12 over Fp4 * Proper naive pairing on BLS12-381 * Frobenius map * Implement naive pairing for BN curves * Add pairing tests to CI + reduce time spent on lower-level tests * Test without assembler in Github Actions + less base layers test iterations
2020-09-21 21:24:00 +00:00
exec "parallel --keep-order --group < " & buildParallel
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# if sizeof(int) == 8: # 32-bit tests on 64-bit arch
# clearParallelBuild()
# runTests(requireGMP = false, dumpCmdFile = true, test32bit = true, testASM = false)
# exec "parallel --keep-order --group < " & buildParallel
Use GMP and GNU parallel in GIthub Actions (#63) * Use GMP and GNU parallel in GIthub Actions * try gmp:i386 * Don't do GMP tests on i386
2020-06-20 17:46:30 +00:00
# Now run the benchmarks
#
Fix test suite (#116) * Pin nim-serialization. Workaround #113 and https://github.com/status-im/nim-serialization/issues/33 * Need to workaround nimble installing dependency multiple times * non-interactive * UB sanitizer missing on mingw * Fix OpenSSL benchmark on non-Linux platforms * Accelerate CI: - Skip 32-bit on 64-bit tests - Only test leaf functionality. * Don't define -fstack-protector-all with MinGW * skip line functions and cyclotomic tests (already tested in pairing) + only compile the benches don't run them.
2021-01-21 20:25:42 +00:00
# Benchmarks compile
Use GMP and GNU parallel in GIthub Actions (#63) * Use GMP and GNU parallel in GIthub Actions * try gmp:i386 * Don't do GMP tests on i386
2020-06-20 17:46:30 +00:00
# ignore Windows 32-bit for the moment
# Ensure benchmarks stay relevant. Ignore Windows 32-bit at the moment
if not defined(windows) or not (existsEnv"UCPU" or getEnv"UCPU" == "i686"):
hash-to-curve BLS12-381 perf (#163) * fp square noasm split from non-4 non-6 limbs fallback (40% speedup) * optimized cofactor clearing for BLS12-381 G2 * Support jacobian isogenies and point_add on isogenies * fuse addition and isogeny map * {.noInit.} and sparseMul * poly_eval_horner init * dedicated invsqrt + cleanup square root file * hash to field: reduce copy overhead and don't return arrays * h2c isogeny jacobian reuse pow 3 precomputed value * Fix sqrt bench
2021-08-14 19:01:50 +00:00
buildAllBenches(useASM = false)
Build and run tests in parallel (#41) * Build and run tests in parallel * Workaround travis OSX Homebrew bug * semicolons ... * Don't auto-update before installing homebrew packages
2020-06-07 17:39:34 +00:00
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Finite field 𝔽p
# ------------------------------------------
Benchmark template for 𝔽p, 𝔽p2, 𝔽p6
2020-03-21 01:31:31 +00:00
task bench_fp, "Run benchmark 𝔽p with your default compiler":
runBench("bench_fp")
task bench_fp_gcc, "Run benchmark 𝔽p with gcc":
runBench("bench_fp", "gcc")
task bench_fp_clang, "Run benchmark 𝔽p with clang":
runBench("bench_fp", "clang")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fp_gcc_noasm, "Run benchmark 𝔽p with gcc - no Assembly":
runBench("bench_fp", "gcc", useAsm = false)
task bench_fp_clang_noasm, "Run benchmark 𝔽p with clang - no Assembly":
runBench("bench_fp", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Double-precision field 𝔽pDbl
# ------------------------------------------
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fpdbl, "Run benchmark 𝔽pDbl with your default compiler":
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
runBench("bench_fp_double_precision")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fpdbl_gcc, "Run benchmark 𝔽p with gcc":
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
runBench("bench_fp_double_precision", "gcc")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fpdbl_clang, "Run benchmark 𝔽p with clang":
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
runBench("bench_fp_double_precision", "clang")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fpdbl_gcc_noasm, "Run benchmark 𝔽p with gcc - no Assembly":
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
runBench("bench_fp_double_precision", "gcc", useAsm = false)
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fpdbl_clang_noasm, "Run benchmark 𝔽p with clang - no Assembly":
Double-Precision towering (#155) * consistent naming for dbl-width * Isolate double-width Fp2 mul * Implement double-width complex multiplication * Lay out Fp4 double-width mul * Off by p in square Fp4 as well :/ * less copies and stack space in addition chains * Address https://github.com/mratsim/constantine/issues/154 partly * Fix #154, faster Fp4 square: less non-residue, no Mul, only square (bit more ops total) * Fix typo * better assembly scheduling for add/sub * Double-width -> Double-precision * Unred -> Unr * double-precision modular addition * Replace canUseNoCarryMontyMul and canUseNoCarryMontySquare by getSpareBits * Complete the double-precision implementation * Use double-precision path for Fp4 squaring and mul * remove mixin annotations * Lazy reduction in Fp4 prod * Fix assembly for sum2xMod * Assembly for double-precision negation * reduce white spaces in pairing benchmarks * ADX implies BMI2
2021-02-09 21:57:45 +00:00
runBench("bench_fp_double_precision", "clang", useAsm = false)
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Extension field 𝔽p2
# ------------------------------------------
Benchmark template for 𝔽p, 𝔽p2, 𝔽p6
2020-03-21 01:31:31 +00:00
task bench_fp2, "Run benchmark with 𝔽p2 your default compiler":
runBench("bench_fp2")
task bench_fp2_gcc, "Run benchmark 𝔽p2 with gcc":
runBench("bench_fp2", "gcc")
task bench_fp2_clang, "Run benchmark 𝔽p2 with clang":
runBench("bench_fp2", "clang")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fp2_gcc_noasm, "Run benchmark 𝔽p2 with gcc - no Assembly":
runBench("bench_fp2", "gcc", useAsm = false)
task bench_fp2_clang_noasm, "Run benchmark 𝔽p2 with clang - no Assembly":
runBench("bench_fp2", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Extension field 𝔽p4
# ------------------------------------------
Jacobian coordinates (#95) * Add projective-> affine bench * Add conditional copy and div2 benches * Fp4 benchmarks * Constant-time Jacobian addition * Jacobian doubling * Use a simpler Add+Dbl complete formula * Update tests * Fix conditional negate * Rollaback complete addition, we were only handling curve coef a == 0
2020-10-01 22:01:09 +00:00
task bench_fp4, "Run benchmark with 𝔽p4 your default compiler":
runBench("bench_fp4")
task bench_fp4_gcc, "Run benchmark 𝔽p4 with gcc":
runBench("bench_fp4", "gcc")
task bench_fp4_clang, "Run benchmark 𝔽p4 with clang":
runBench("bench_fp4", "clang")
task bench_fp4_gcc_noasm, "Run benchmark 𝔽p4 with gcc - no Assembly":
runBench("bench_fp4", "gcc", useAsm = false)
task bench_fp4_clang_noasm, "Run benchmark 𝔽p4 with clang - no Assembly":
runBench("bench_fp4", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Extension field 𝔽p6
# ------------------------------------------
Benchmark template for 𝔽p, 𝔽p2, 𝔽p6
2020-03-21 01:31:31 +00:00
task bench_fp6, "Run benchmark with 𝔽p6 your default compiler":
runBench("bench_fp6")
task bench_fp6_gcc, "Run benchmark 𝔽p6 with gcc":
runBench("bench_fp6", "gcc")
task bench_fp6_clang, "Run benchmark 𝔽p6 with clang":
runBench("bench_fp6", "clang")
Implement 𝔽p12 inversion, enable 𝔽p12 tests and bench
2020-04-09 12:28:01 +00:00
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fp6_gcc_noasm, "Run benchmark 𝔽p6 with gcc - no Assembly":
runBench("bench_fp6", "gcc", useAsm = false)
task bench_fp6_clang_noasm, "Run benchmark 𝔽p6 with clang - no Assembly":
runBench("bench_fp6", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Extension field 𝔽p12
# ------------------------------------------
Implement 𝔽p12 inversion, enable 𝔽p12 tests and bench
2020-04-09 12:28:01 +00:00
task bench_fp12, "Run benchmark with 𝔽p12 your default compiler":
runBench("bench_fp12")
task bench_fp12_gcc, "Run benchmark 𝔽p12 with gcc":
runBench("bench_fp12", "gcc")
task bench_fp12_clang, "Run benchmark 𝔽p12 with clang":
runBench("bench_fp12", "clang")
add EC bench to nimble tasks
2020-04-15 17:43:31 +00:00
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_fp12_gcc_noasm, "Run benchmark 𝔽p12 with gcc - no Assembly":
runBench("bench_fp12", "gcc", useAsm = false)
task bench_fp12_clang_noasm, "Run benchmark 𝔽p12 with clang - no Assembly":
runBench("bench_fp12", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Elliptic curve G1
# ------------------------------------------
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
task bench_ec_g1, "Run benchmark on Elliptic Curve group 𝔾1 - Short Weierstrass with Projective Coordinates - Default compiler":
Endomorphism acceleration for Scalar Multiplication (#44) * Add MultiScalar recoding from "Efficient and Secure Algorithms for GLV-Based Scalar Multiplication" by Faz et al * precompute cube root of unity - Add VM precomputation of Fp - workaround upstream bug https://github.com/nim-lang/Nim/issues/14585 * Add the φ-accelerated lookup table builder * Add a dedicated bithacks file * cosmetic import consistency * Build the φ precompute table with n-1 EC additions instead of 2^(n-1) additions * remove binary * Add the GLV precomputations to the sage scripts * You can't avoid it, bigint multiplication is needed at one point * Add bigint multiplication discarding some low words * Implement the lattice decomposition in sage * Proper decomposition for BN254 * Prepare the code for a new scalar mul * We compile, and now debugging hunt * More helpers to debug GLV scalar Mul * Fix conditional negation * Endomorphism accelerated scalar mul working for BN254 curve * Implement endomorphism acceleration for BLS12-381 (needed cofactor clearing of the point) * fix nimble test script after bench rename
2020-06-14 13:39:06 +00:00
runBench("bench_ec_g1")
add EC bench to nimble tasks
2020-04-15 17:43:31 +00:00
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
task bench_ec_g1_gcc, "Run benchmark on Elliptic Curve group 𝔾1 - Short Weierstrass with Projective Coordinates - GCC":
Endomorphism acceleration for Scalar Multiplication (#44) * Add MultiScalar recoding from "Efficient and Secure Algorithms for GLV-Based Scalar Multiplication" by Faz et al * precompute cube root of unity - Add VM precomputation of Fp - workaround upstream bug https://github.com/nim-lang/Nim/issues/14585 * Add the φ-accelerated lookup table builder * Add a dedicated bithacks file * cosmetic import consistency * Build the φ precompute table with n-1 EC additions instead of 2^(n-1) additions * remove binary * Add the GLV precomputations to the sage scripts * You can't avoid it, bigint multiplication is needed at one point * Add bigint multiplication discarding some low words * Implement the lattice decomposition in sage * Proper decomposition for BN254 * Prepare the code for a new scalar mul * We compile, and now debugging hunt * More helpers to debug GLV scalar Mul * Fix conditional negation * Endomorphism accelerated scalar mul working for BN254 curve * Implement endomorphism acceleration for BLS12-381 (needed cofactor clearing of the point) * fix nimble test script after bench rename
2020-06-14 13:39:06 +00:00
runBench("bench_ec_g1", "gcc")
add EC bench to nimble tasks
2020-04-15 17:43:31 +00:00
Endomorphism acceleration for Scalar Multiplication (#44) * Add MultiScalar recoding from "Efficient and Secure Algorithms for GLV-Based Scalar Multiplication" by Faz et al * precompute cube root of unity - Add VM precomputation of Fp - workaround upstream bug https://github.com/nim-lang/Nim/issues/14585 * Add the φ-accelerated lookup table builder * Add a dedicated bithacks file * cosmetic import consistency * Build the φ precompute table with n-1 EC additions instead of 2^(n-1) additions * remove binary * Add the GLV precomputations to the sage scripts * You can't avoid it, bigint multiplication is needed at one point * Add bigint multiplication discarding some low words * Implement the lattice decomposition in sage * Proper decomposition for BN254 * Prepare the code for a new scalar mul * We compile, and now debugging hunt * More helpers to debug GLV scalar Mul * Fix conditional negation * Endomorphism accelerated scalar mul working for BN254 curve * Implement endomorphism acceleration for BLS12-381 (needed cofactor clearing of the point) * fix nimble test script after bench rename
2020-06-14 13:39:06 +00:00
task bench_ec_g1_clang, "Run benchmark on Elliptic Curve group 𝔾1 - Short Weierstrass with Projective Coordinates - Clang":
runBench("bench_ec_g1", "clang")
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_ec_g1_gcc_noasm, "Run benchmark on Elliptic Curve group 𝔾1 - Short Weierstrass with Projective Coordinates - GCC no Assembly":
runBench("bench_ec_g1", "gcc", useAsm = false)
task bench_ec_g1_clang_noasm, "Run benchmark on Elliptic Curve group 𝔾1 - Short Weierstrass with Projective Coordinates - Clang no Assembly":
runBench("bench_ec_g1", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Elliptic curve G2
# ------------------------------------------
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
task bench_ec_g2, "Run benchmark on Elliptic Curve group 𝔾2 - Short Weierstrass with Projective Coordinates - Default compiler":
G2 / Operations on the twisted curve E'(Fp2) (#51) * Split elliptic curve tests to better use parallel testing * Add support for printing points on G2 * Implement multiplication and division by optimal sextic non-residue (BLS12-381) * Implement modular square root in 𝔽p2 * Support EC add and EC double on G2 (for BLS12-381) * Support G2 divisive twists with non-unit sextic-non-residue like BN254 snarks * Add EC G2 bench * cleanup some unused warnings * Reorg the tests for parallelization and to avoid instantiating huge files
2020-06-15 20:58:56 +00:00
runBench("bench_ec_g2")
task bench_ec_g2_gcc, "Run benchmark on Elliptic Curve group 𝔾2 - Short Weierstrass with Projective Coordinates - GCC":
runBench("bench_ec_g2", "gcc")
task bench_ec_g2_clang, "Run benchmark on Elliptic Curve group 𝔾2 - Short Weierstrass with Projective Coordinates - Clang":
runBench("bench_ec_g2", "clang")
Double-width tower extension part 1 (#72) * Implement double-width field multiplication for double-width towering * Fp2 mul acceleration via double-width lazy reduction (pure Nim) * Inline assembly for basic add and sub * Use 2 registers instead of 12+ for ASM conditional copy * Prepare assembly for extended multiprecision multiplication support * Add assembly for mul * initial implementation of assembly reduction * stash current progress of assembly reduction * Fix clobbering issue, only P256 comparison remain buggy * Fix asm montgomery reduction for NIST P256 as well * MULX/ADCX/ADOX multi-precision multiplication * MULX/ADCX/ADOX reduction v1 * Add (deactivated) assembly for double-width substraction + rework benches * Add bench to nimble and deactivate double-width for now. slower than classic * Fix x86-32 running out of registers for mul * Clang needs to be at v9 to support flag output constraints (Xcode 11.4.2 / OSX Catalina) * 32-bit doesn't have enough registers for ASM mul * Fix again Travis Clang 9 issues * LLVM 9 is not whitelisted in travis * deactivated assembler with travis clang * syntax error * another * ... * missing space, yeah ...
2020-08-20 08:21:39 +00:00
task bench_ec_g2_gcc_noasm, "Run benchmark on Elliptic Curve group 𝔾2 - Short Weierstrass with Projective Coordinates - GCC no Assembly":
runBench("bench_ec_g2", "gcc", useAsm = false)
task bench_ec_g2_clang_noasm, "Run benchmark on Elliptic Curve group 𝔾2 - Short Weierstrass with Projective Coordinates - Clang no Assembly":
runBench("bench_ec_g2", "clang", useAsm = false)
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Pairings
# ------------------------------------------
BLS12-377 (#91) * add Sage for constant time tonelli shanks * Fused sqrt and invsqrt via Tonelli Shanks * isolate sqrt in their own folder * Implement constant-time Tonelli Shanks for any prime * Implement Fp2 sqrt for any non-residue * Add tests for BLS12_377 * Lattice decomposition script for BLS12_377 G1 * BLS12-377 G1 GLV ok, G2 GLV issue * Proper endomorphism acceleration support for BLS12-377 * Add naive pairing support for BLS12-377 * Activate more bench for BLS12-377 * Fix MSB computation * Optimize final exponentiation + add benches
2020-09-27 07:15:14 +00:00
task bench_pairing_bls12_377, "Run pairings benchmarks for BLS12-377 - Default compiler":
runBench("bench_pairing_bls12_377")
task bench_pairing_bls12_377_gcc, "Run pairings benchmarks for BLS12-377 - GCC":
runBench("bench_pairing_bls12_377", "gcc")
task bench_pairing_bls12_377_clang, "Run pairings benchmarks for BLS12-377 - Clang":
runBench("bench_pairing_bls12_377", "clang")
task bench_pairing_bls12_377_gcc_noasm, "Run pairings benchmarks for BLS12-377 - GCC no Assembly":
runBench("bench_pairing_bls12_377", "gcc", useAsm = false)
task bench_pairing_bls12_377_clang_noasm, "Run pairings benchmarks for BLS12-377 - Clang no Assembly":
runBench("bench_pairing_bls12_377", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# --
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
task bench_pairing_bls12_381, "Run pairings benchmarks for BLS12-381 - Default compiler":
runBench("bench_pairing_bls12_381")
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bls12_381_gcc, "Run pairings benchmarks for BLS12-381 - GCC":
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
runBench("bench_pairing_bls12_381", "gcc")
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bls12_381_clang, "Run pairings benchmarks for BLS12-381 - Clang":
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
runBench("bench_pairing_bls12_381", "clang")
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bls12_381_gcc_noasm, "Run pairings benchmarks for BLS12-381 - GCC no Assembly":
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
runBench("bench_pairing_bls12_381", "gcc", useAsm = false)
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bls12_381_clang_noasm, "Run pairings benchmarks for BLS12-381 - Clang no Assembly":
Pairing optim (#85) * Fix fp12 Frobenius map * Implement cyclotomic subgroup acceleration * make cyclotomic squaring in-place * Add back out-place cycl squaring and add cyclotomic inverse * Implement state-of-the-art BLS12-381 final exponentiation * save a cyclotomic squaring * Accelerate sparse line multiplication in Miller loop * Add pairing bench * fix comments
2020-09-24 15:18:23 +00:00
runBench("bench_pairing_bls12_381", "clang", useAsm = false)
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# --
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bn254_nogami, "Run pairings benchmarks for BN254-Nogami - Default compiler":
runBench("bench_pairing_bn254_nogami")
task bench_pairing_bn254_nogami_gcc, "Run pairings benchmarks for BN254-Nogami - GCC":
runBench("bench_pairing_bn254_nogami", "gcc")
task bench_pairing_bn254_nogami_clang, "Run pairings benchmarks for BN254-Nogami - Clang":
runBench("bench_pairing_bn254_nogami", "clang")
task bench_pairing_bn254_nogami_gcc_noasm, "Run pairings benchmarks for BN254-Nogami - GCC no Assembly":
runBench("bench_pairing_bn254_nogami", "gcc", useAsm = false)
task bench_pairing_bn254_nogami_clang_noasm, "Run pairings benchmarks for BN254-Nogami - Clang no Assembly":
runBench("bench_pairing_bn254_nogami", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# --
Pairings for BN254-Nogami and BN254-Snarks (#86) * Implement optimized final exponentiation for BN254-Nogami * And BN254 Snarks support * Optimize D-Twist sparse Fp12 x line multiplication * Move quadruple/octuple and add to Github issues: https://github.com/mratsim/constantine/issues/88 [skip ci]
2020-09-25 19:58:20 +00:00
task bench_pairing_bn254_snarks, "Run pairings benchmarks for BN254-Snarks - Default compiler":
runBench("bench_pairing_bn254_snarks")
task bench_pairing_bn254_snarks_gcc, "Run pairings benchmarks for BN254-Snarks - GCC":
runBench("bench_pairing_bn254_snarks", "gcc")
task bench_pairing_bn254_snarks_clang, "Run pairings benchmarks for BN254-Snarks - Clang":
runBench("bench_pairing_bn254_snarks", "clang")
task bench_pairing_bn254_snarks_gcc_noasm, "Run pairings benchmarks for BN254-Snarks - GCC no Assembly":
runBench("bench_pairing_bn254_snarks", "gcc", useAsm = false)
task bench_pairing_bn254_snarks_clang_noasm, "Run pairings benchmarks for BN254-Snarks - Clang no Assembly":
runBench("bench_pairing_bn254_snarks", "clang", useAsm = false)
SHA256 Hash function
2020-12-15 18:18:36 +00:00
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Curve summaries
# ------------------------------------------
task bench_summary_bls12_377, "Run summary benchmarks for BLS12-377 - Default compiler":
runBench("bench_summary_bls12_377")
task bench_summary_bls12_377_gcc, "Run summary benchmarks for BLS12-377 - GCC":
runBench("bench_summary_bls12_377", "gcc")
task bench_summary_bls12_377_clang, "Run summary benchmarks for BLS12-377 - Clang":
runBench("bench_summary_bls12_377", "clang")
task bench_summary_bls12_377_gcc_noasm, "Run summary benchmarks for BLS12-377 - GCC no Assembly":
runBench("bench_summary_bls12_377", "gcc", useAsm = false)
task bench_summary_bls12_377_clang_noasm, "Run summary benchmarks for BLS12-377 - Clang no Assembly":
runBench("bench_summary_bls12_377", "clang", useAsm = false)
# --
task bench_summary_bls12_381, "Run summary benchmarks for BLS12-381 - Default compiler":
runBench("bench_summary_bls12_381")
task bench_summary_bls12_381_gcc, "Run summary benchmarks for BLS12-381 - GCC":
runBench("bench_summary_bls12_381", "gcc")
task bench_summary_bls12_381_clang, "Run summary benchmarks for BLS12-381 - Clang":
runBench("bench_summary_bls12_381", "clang")
task bench_summary_bls12_381_gcc_noasm, "Run summary benchmarks for BLS12-381 - GCC no Assembly":
runBench("bench_summary_bls12_381", "gcc", useAsm = false)
task bench_summary_bls12_381_clang_noasm, "Run summary benchmarks for BLS12-381 - Clang no Assembly":
runBench("bench_summary_bls12_381", "clang", useAsm = false)
# --
task bench_summary_bn254_nogami, "Run summary benchmarks for BN254-Nogami - Default compiler":
runBench("bench_summary_bn254_nogami")
task bench_summary_bn254_nogami_gcc, "Run summary benchmarks for BN254-Nogami - GCC":
runBench("bench_summary_bn254_nogami", "gcc")
task bench_summary_bn254_nogami_clang, "Run summary benchmarks for BN254-Nogami - Clang":
runBench("bench_summary_bn254_nogami", "clang")
task bench_summary_bn254_nogami_gcc_noasm, "Run summary benchmarks for BN254-Nogami - GCC no Assembly":
runBench("bench_summary_bn254_nogami", "gcc", useAsm = false)
task bench_summary_bn254_nogami_clang_noasm, "Run summary benchmarks for BN254-Nogami - Clang no Assembly":
runBench("bench_summary_bn254_nogami", "clang", useAsm = false)
# --
task bench_summary_bn254_snarks, "Run summary benchmarks for BN254-Snarks - Default compiler":
runBench("bench_summary_bn254_snarks")
task bench_summary_bn254_snarks_gcc, "Run summary benchmarks for BN254-Snarks - GCC":
runBench("bench_summary_bn254_snarks", "gcc")
task bench_summary_bn254_snarks_clang, "Run summary benchmarks for BN254-Snarks - Clang":
runBench("bench_summary_bn254_snarks", "clang")
task bench_summary_bn254_snarks_gcc_noasm, "Run summary benchmarks for BN254-Snarks - GCC no Assembly":
runBench("bench_summary_bn254_snarks", "gcc", useAsm = false)
task bench_summary_bn254_snarks_clang_noasm, "Run summary benchmarks for BN254-Snarks - Clang no Assembly":
runBench("bench_summary_bn254_snarks", "clang", useAsm = false)
Pasta curves (#191) * Pasta curves field arithmetic * implement elliptic curve arith for the Pasta curves
2022-04-26 22:58:48 +00:00
# --
task bench_summary_pasta, "Run summary benchmarks for the Pasta curves - Default compiler":
runBench("bench_summary_pasta")
task bench_summary_pasta_gcc, "Run summary benchmarks for the Pasta curves - GCC":
runBench("bench_summary_pasta", "gcc")
task bench_summary_pasta_clang, "Run summary benchmarks for the Pasta curves - Clang":
runBench("bench_summary_pasta", "clang")
task bench_summary_pasta_gcc_noasm, "Run summary benchmarks for the Pasta curves - GCC no Assembly":
runBench("bench_summary_pasta", "gcc", useAsm = false)
task bench_summary_pasta_clang_noasm, "Run summary benchmarks for the Pasta curves - Clang no Assembly":
runBench("bench_summary_pasta", "clang", useAsm = false)
add bench whole summary for curves
2021-02-14 13:24:48 +00:00
# Hashes
# ------------------------------------------
SHA256 Hash function
2020-12-15 18:18:36 +00:00
task bench_sha256, "Run SHA256 benchmarks":
runBench("bench_sha256")
Hash to curve - BLS12-381 (#110) * Hash to Curve: impl expand_message_xmd * Try to precompute part of hash to curve at compile-time * sha256 bench - use the new hashes module * [WIP] smoke test hash to field * Implement hash_to_field with expected output * unoptimized hash-to-curve G2 for BLS12-381 * Don't run sanitizer on hash to field as it uses GC-ed strings
2021-08-13 20:07:26 +00:00
# Hash-to-curve
# ------------------------------------------
task bench_hash_to_curve, "Run Hash-to-Curve benchmarks":
runBench("bench_hash_to_curve")
task bench_hash_to_curve_gcc, "Run Hash-to-Curve benchmarks":
runBench("bench_hash_to_curve", "gcc")
task bench_hash_to_curve_clang, "Run Hash-to-Curve benchmarks":
runBench("bench_hash_to_curve", "clang")
task bench_hash_to_curve_gcc_noasm, "Run Hash-to-Curve benchmarks":
runBench("bench_hash_to_curve", "gcc", useAsm = false)
task bench_hash_to_curve_clang_noasm, "Run Hash-to-Curve benchmarks":
runBench("bench_hash_to_curve", "clang", useAsm = false)
BLS signatures for Ethereum (BLS sig on BLS12-381 G2 with SHA256) (#183) * Finally add the (Ethereum) bls signatures (on BLS12-381 G2) * fix test path and remove old low-level signature test
2022-02-26 20:22:34 +00:00
# BLS signatures
# ------------------------------------------
task bench_blssig_on_bls12_381_g2, "Run Hash-to-Curve benchmarks":
runBench("bench_blssig_on_bls12_381_g2")
task bench_blssig_on_bls12_381_g2_gcc, "Run Hash-to-Curve benchmarks":
runBench("bench_blssig_on_bls12_381_g2", "gcc")
task bench_blssig_on_bls12_381_g2_clang, "Run Hash-to-Curve benchmarks":
runBench("bench_blssig_on_bls12_381_g2", "clang")
task bench_blssig_on_bls12_381_g2_gcc_noasm, "Run Hash-to-Curve benchmarks":
runBench("bench_blssig_on_bls12_381_g2", "gcc", useAsm = false)
task bench_blssig_on_bls12_381_g2_clang_noasm, "Run Hash-to-Curve benchmarks":
runBench("bench_blssig_on_bls12_381_g2", "clang", useAsm = false)