AArch64: apply SLOTHY to poly_decompose_32 and poly_decompose_88 by mkannwischer · Pull Request #1180 · pq-code-package/mldsa-native

mkannwischer · 2026-06-14T23:58:08Z

AArch64: apply SLOTHY to poly_decompose_32 and poly_decompose_88

Switch the dev/aarch64_opt Makefile rules from a plain copy to a
slothy-cli invocation and check in the resulting schedules. Measured
on Neoverse-N1: poly_decompose_88 323→298 cyc (~7.7%) and
poly_decompose_32 314→298 cyc (~5.1%). The HOL Light proofs are
unchanged apart from regenerated embedded bytecode.

Resolves SLOTHY: Apply to AArch64 poly_decompose #488

Switch the dev/aarch64_opt Makefile rules from a plain copy to a slothy-cli invocation and check in the resulting schedules. Measured on Neoverse-N1: poly_decompose_88 323→298 cyc (~7.7%) and poly_decompose_32 314→298 cyc (~5.1%). The HOL Light proofs are unchanged apart from regenerated embedded bytecode. - Resolves #488 Signed-off-by: Matthias J. Kannwischer <matthias@zerorisc.com>

github-actions

Mac Mini (M1, 2020) benchmarks (opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`46537` cycles	`46504` cycles	`1.00`
`ML-DSA-44 sign`	`131043` cycles	`131079` cycles	`1.00`
`ML-DSA-44 verify`	`47342` cycles	`47310` cycles	`1.00`
`ML-DSA-65 keypair`	`81690` cycles	`81680` cycles	`1.00`
`ML-DSA-65 sign`	`215294` cycles	`215311` cycles	`1.00`
`ML-DSA-65 verify`	`79299` cycles	`79299` cycles	`1`
`ML-DSA-87 keypair`	`132403` cycles	`132405` cycles	`1.00`
`ML-DSA-87 sign`	`277435` cycles	`277328` cycles	`1.00`
`ML-DSA-87 verify`	`134053` cycles	`134051` cycles	`1.00`

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

github-actions

Mac Mini (M1, 2020) benchmarks (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`112834` cycles	`112758` cycles	`1.00`
`ML-DSA-44 sign`	`401056` cycles	`400845` cycles	`1.00`
`ML-DSA-44 verify`	`119495` cycles	`119413` cycles	`1.00`
`ML-DSA-65 keypair`	`193001` cycles	`192933` cycles	`1.00`
`ML-DSA-65 sign`	`650124` cycles	`649924` cycles	`1.00`
`ML-DSA-65 verify`	`192920` cycles	`192850` cycles	`1.00`
`ML-DSA-87 keypair`	`318883` cycles	`318753` cycles	`1.00`
`ML-DSA-87 sign`	`828737` cycles	`828716` cycles	`1.00`
`ML-DSA-87 verify`	`326732` cycles	`326677` cycles	`1.00`

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

github-actions

SpacemiT K1 8 (Banana Pi F3) benchmarks (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`759745` cycles	`759567` cycles	`1.00`
`ML-DSA-44 sign`	`3138419` cycles	`3137937` cycles	`1.00`
`ML-DSA-44 verify`	`858706` cycles	`858870` cycles	`1.00`
`ML-DSA-65 keypair`	`1286602` cycles	`1285661` cycles	`1.00`
`ML-DSA-65 sign`	`5077186` cycles	`5075651` cycles	`1.00`
`ML-DSA-65 verify`	`1364475` cycles	`1364449` cycles	`1.00`
`ML-DSA-87 keypair`	`2111705` cycles	`2111749` cycles	`1.00`
`ML-DSA-87 sign`	`6349699` cycles	`6355811` cycles	`1.00`
`ML-DSA-87 verify`	`2227295` cycles	`2228276` cycles	`1.00`

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

oqs-bot

Intel Xeon 4th gen (c7i)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`43385` cycles	`43415` cycles	`1.00`
`ML-DSA-44 sign`	`130751` cycles	`130832` cycles	`1.00`
`ML-DSA-44 verify`	`45118` cycles	`45316` cycles	`1.00`
`ML-DSA-65 keypair`	`75780` cycles	`75430` cycles	`1.00`
`ML-DSA-65 sign`	`215460` cycles	`215011` cycles	`1.00`
`ML-DSA-65 verify`	`74415` cycles	`74350` cycles	`1.00`
`ML-DSA-87 keypair`	`123441` cycles	`123299` cycles	`1.00`
`ML-DSA-87 sign`	`272129` cycles	`271370` cycles	`1.00`
`ML-DSA-87 verify`	`120746` cycles	`120802` cycles	`1.00`

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

oqs-bot

Intel Xeon 4th gen (c7i) (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`91420` cycles	`91498` cycles	`1.00`
`ML-DSA-44 sign`	`351755` cycles	`352373` cycles	`1.00`
`ML-DSA-44 verify`	`99750` cycles	`99791` cycles	`1.00`
`ML-DSA-65 keypair`	`154006` cycles	`153828` cycles	`1.00`
`ML-DSA-65 sign`	`571622` cycles	`571106` cycles	`1.00`
`ML-DSA-65 verify`	`159790` cycles	`159761` cycles	`1.00`
`ML-DSA-87 keypair`	`255069` cycles	`255664` cycles	`1.00`
`ML-DSA-87 sign`	`723883` cycles	`726163` cycles	`1.00`
`ML-DSA-87 verify`	`264098` cycles	`263879` cycles	`1.00`

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

oqs-bot

AMD EPYC 3rd gen (c6a)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`55251` cycles	`55320` cycles	`1.00`
`ML-DSA-44 sign`	`159039` cycles	`159183` cycles	`1.00`
`ML-DSA-44 verify`	`57463` cycles	`57873` cycles	`0.99`
`ML-DSA-65 keypair`	`96210` cycles	`95830` cycles	`1.00`
`ML-DSA-65 sign`	`263757` cycles	`263504` cycles	`1.00`
`ML-DSA-65 verify`	`96156` cycles	`96058` cycles	`1.00`
`ML-DSA-87 keypair`	`154748` cycles	`154555` cycles	`1.00`
`ML-DSA-87 sign`	`322661` cycles	`322765` cycles	`1.00`
`ML-DSA-87 verify`	`150975` cycles	`150846` cycles	`1.00`

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

oqs-bot

AMD EPYC 3rd gen (c6a) (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`133164` cycles	`133213` cycles	`1.00`
`ML-DSA-44 sign`	`518907` cycles	`518708` cycles	`1.00`
`ML-DSA-44 verify`	`146469` cycles	`146419` cycles	`1.00`
`ML-DSA-65 keypair`	`223893` cycles	`225658` cycles	`0.99`
`ML-DSA-65 sign`	`842603` cycles	`848883` cycles	`0.99`
`ML-DSA-65 verify`	`233779` cycles	`236008` cycles	`0.99`
`ML-DSA-87 keypair`	`368482` cycles	`366727` cycles	`1.00`
`ML-DSA-87 sign`	`1066561` cycles	`1060028` cycles	`1.01`
`ML-DSA-87 verify`	`382702` cycles	`380316` cycles	`1.01`

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

oqs-bot

AMD EPYC 4th gen (c7a)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`46302` cycles	`46860` cycles	`0.99`
`ML-DSA-44 sign`	`138499` cycles	`139486` cycles	`0.99`
`ML-DSA-44 verify`	`49383` cycles	`49318` cycles	`1.00`
`ML-DSA-65 keypair`	`82453` cycles	`81920` cycles	`1.01`
`ML-DSA-65 sign`	`226079` cycles	`227729` cycles	`0.99`
`ML-DSA-65 verify`	`82028` cycles	`81877` cycles	`1.00`
`ML-DSA-87 keypair`	`129336` cycles	`131127` cycles	`0.99`
`ML-DSA-87 sign`	`277767` cycles	`281288` cycles	`0.99`
`ML-DSA-87 verify`	`127997` cycles	`129015` cycles	`0.99`

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

oqs-bot

Graviton4

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`67216` cycles	`67251` cycles	`1.00`
`ML-DSA-44 sign`	`198307` cycles	`198318` cycles	`1.00`
`ML-DSA-44 verify`	`70252` cycles	`70240` cycles	`1.00`
`ML-DSA-65 keypair`	`119288` cycles	`119362` cycles	`1.00`
`ML-DSA-65 sign`	`325990` cycles	`325982` cycles	`1.00`
`ML-DSA-65 verify`	`116826` cycles	`116934` cycles	`1.00`
`ML-DSA-87 keypair`	`196527` cycles	`196595` cycles	`1.00`
`ML-DSA-87 sign`	`421291` cycles	`421892` cycles	`1.00`
`ML-DSA-87 verify`	`193331` cycles	`193379` cycles	`1.00`

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

oqs-bot

Graviton2

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`112324` cycles	`112529` cycles	`1.00`
`ML-DSA-44 sign`	`353714` cycles	`354040` cycles	`1.00`
`ML-DSA-44 verify`	`117482` cycles	`117389` cycles	`1.00`
`ML-DSA-65 keypair`	`194624` cycles	`194738` cycles	`1.00`
`ML-DSA-65 sign`	`583294` cycles	`584565` cycles	`1.00`
`ML-DSA-65 verify`	`193425` cycles	`193297` cycles	`1.00`
`ML-DSA-87 keypair`	`320920` cycles	`321000` cycles	`1.00`
`ML-DSA-87 sign`	`747036` cycles	`746870` cycles	`1.00`
`ML-DSA-87 verify`	`318279` cycles	`318737` cycles	`1.00`

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

oqs-bot

AMD EPYC 4th gen (c7a) (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`117989` cycles	`118568` cycles	`1.00`
`ML-DSA-44 sign`	`457446` cycles	`458777` cycles	`1.00`
`ML-DSA-44 verify`	`130118` cycles	`131112` cycles	`0.99`
`ML-DSA-65 keypair`	`203668` cycles	`200686` cycles	`1.01`
`ML-DSA-65 sign`	`747194` cycles	`743736` cycles	`1.00`
`ML-DSA-65 verify`	`212113` cycles	`209264` cycles	`1.01`
`ML-DSA-87 keypair`	`329210` cycles	`330171` cycles	`1.00`
`ML-DSA-87 sign`	`936999` cycles	`935697` cycles	`1.00`
`ML-DSA-87 verify`	`342522` cycles	`343489` cycles	`1.00`

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

oqs-bot

Graviton4 (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`127588` cycles	`127647` cycles	`1.00`
`ML-DSA-44 sign`	`441200` cycles	`441052` cycles	`1.00`
`ML-DSA-44 verify`	`136388` cycles	`136340` cycles	`1.00`
`ML-DSA-65 keypair`	`220545` cycles	`220707` cycles	`1.00`
`ML-DSA-65 sign`	`714429` cycles	`713810` cycles	`1.00`
`ML-DSA-65 verify`	`220997` cycles	`220735` cycles	`1.00`
`ML-DSA-87 keypair`	`364582` cycles	`365112` cycles	`1.00`
`ML-DSA-87 sign`	`915573` cycles	`921310` cycles	`0.99`
`ML-DSA-87 verify`	`370884` cycles	`370798` cycles	`1.00`

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

oqs-bot

Intel Xeon 3rd gen (c6i)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`62013` cycles	`61685` cycles	`1.01`
`ML-DSA-44 sign`	`188813` cycles	`188995` cycles	`1.00`
`ML-DSA-44 verify`	`66307` cycles	`66273` cycles	`1.00`
`ML-DSA-65 keypair`	`112010` cycles	`111473` cycles	`1.00`
`ML-DSA-65 sign`	`316174` cycles	`312617` cycles	`1.01`
`ML-DSA-65 verify`	`110578` cycles	`110558` cycles	`1.00`
`ML-DSA-87 keypair`	`171339` cycles	`171460` cycles	`1.00`
`ML-DSA-87 sign`	`378831` cycles	`377513` cycles	`1.00`
`ML-DSA-87 verify`	`170072` cycles	`169102` cycles	`1.01`

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

oqs-bot

Graviton2 (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`212222` cycles	`211802` cycles	`1.00`
`ML-DSA-44 sign`	`761067` cycles	`759783` cycles	`1.00`
`ML-DSA-44 verify`	`229929` cycles	`229307` cycles	`1.00`
`ML-DSA-65 keypair`	`378728` cycles	`377288` cycles	`1.00`
`ML-DSA-65 sign`	`1248002` cycles	`1247192` cycles	`1.00`
`ML-DSA-65 verify`	`373359` cycles	`371605` cycles	`1.00`
`ML-DSA-87 keypair`	`602631` cycles	`601230` cycles	`1.00`
`ML-DSA-87 sign`	`1585048` cycles	`1584827` cycles	`1.00`
`ML-DSA-87 verify`	`618528` cycles	`616717` cycles	`1.00`

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

oqs-bot

Graviton3

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`71305` cycles	`71561` cycles	`1.00`
`ML-DSA-44 sign`	`209071` cycles	`209019` cycles	`1.00`
`ML-DSA-44 verify`	`74710` cycles	`74745` cycles	`1.00`
`ML-DSA-65 keypair`	`125924` cycles	`125918` cycles	`1.00`
`ML-DSA-65 sign`	`345759` cycles	`345393` cycles	`1.00`
`ML-DSA-65 verify`	`124066` cycles	`124184` cycles	`1.00`
`ML-DSA-87 keypair`	`206974` cycles	`206649` cycles	`1.00`
`ML-DSA-87 sign`	`444175` cycles	`439798` cycles	`1.01`
`ML-DSA-87 verify`	`204042` cycles	`204451` cycles	`1.00`

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

oqs-bot

Intel Xeon 3rd gen (c6i) (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`154456` cycles	`154753` cycles	`1.00`
`ML-DSA-44 sign`	`587935` cycles	`591158` cycles	`0.99`
`ML-DSA-44 verify`	`169516` cycles	`169947` cycles	`1.00`
`ML-DSA-65 keypair`	`262430` cycles	`261726` cycles	`1.00`
`ML-DSA-65 sign`	`964711` cycles	`963022` cycles	`1.00`
`ML-DSA-65 verify`	`272608` cycles	`271733` cycles	`1.00`
`ML-DSA-87 keypair`	`431590` cycles	`431775` cycles	`1.00`
`ML-DSA-87 sign`	`1212135` cycles	`1213443` cycles	`1.00`
`ML-DSA-87 verify`	`447757` cycles	`447885` cycles	`1.00`

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

github-actions

Arm Cortex-A76 (Raspberry Pi 5) benchmarks (opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`112223` cycles	`112128` cycles	`1.00`
`ML-DSA-44 sign`	`353019` cycles	`353788` cycles	`1.00`
`ML-DSA-44 verify`	`117145` cycles	`117189` cycles	`1.00`
`ML-DSA-65 keypair`	`194591` cycles	`194358` cycles	`1.00`
`ML-DSA-65 sign`	`582967` cycles	`583733` cycles	`1.00`
`ML-DSA-65 verify`	`192794` cycles	`193111` cycles	`1.00`
`ML-DSA-87 keypair`	`320571` cycles	`320083` cycles	`1.00`
`ML-DSA-87 sign`	`745952` cycles	`747201` cycles	`1.00`
`ML-DSA-87 verify`	`318435` cycles	`317895` cycles	`1.00`

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

oqs-bot

Graviton3 (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`137918` cycles	`138038` cycles	`1.00`
`ML-DSA-44 sign`	`485921` cycles	`486082` cycles	`1.00`
`ML-DSA-44 verify`	`149024` cycles	`149093` cycles	`1.00`
`ML-DSA-65 keypair`	`241712` cycles	`241929` cycles	`1.00`
`ML-DSA-65 sign`	`791958` cycles	`791534` cycles	`1.00`
`ML-DSA-65 verify`	`242168` cycles	`241280` cycles	`1.00`
`ML-DSA-87 keypair`	`395759` cycles	`396309` cycles	`1.00`
`ML-DSA-87 sign`	`1013511` cycles	`1019346` cycles	`0.99`
`ML-DSA-87 verify`	`403617` cycles	`403741` cycles	`1.00`

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

github-actions

Arm Cortex-A55 (Snapdragon 888) benchmarks (opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`267129` cycles	`266639` cycles	`1.00`
`ML-DSA-44 sign`	`804495` cycles	`808808` cycles	`0.99`
`ML-DSA-44 verify`	`270621` cycles	`270153` cycles	`1.00`
`ML-DSA-65 keypair`	`461167` cycles	`461300` cycles	`1.00`
`ML-DSA-65 sign`	`1310863` cycles	`1323365` cycles	`0.99`
`ML-DSA-65 verify`	`448034` cycles	`447313` cycles	`1.00`
`ML-DSA-87 keypair`	`788871` cycles	`789934` cycles	`1.00`
`ML-DSA-87 sign`	`1782995` cycles	`1827662` cycles	`0.98`
`ML-DSA-87 verify`	`768306` cycles	`769879` cycles	`1.00`

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

github-actions

Arm Cortex-A76 (Raspberry Pi 5) benchmarks (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`211528` cycles	`211527` cycles	`1.00`
`ML-DSA-44 sign`	`758665` cycles	`759858` cycles	`1.00`
`ML-DSA-44 verify`	`228972` cycles	`229351` cycles	`1.00`
`ML-DSA-65 keypair`	`377109` cycles	`378548` cycles	`1.00`
`ML-DSA-65 sign`	`1247274` cycles	`1247648` cycles	`1.00`
`ML-DSA-65 verify`	`371224` cycles	`372375` cycles	`1.00`
`ML-DSA-87 keypair`	`600381` cycles	`601887` cycles	`1.00`
`ML-DSA-87 sign`	`1582615` cycles	`1582472` cycles	`1.00`
`ML-DSA-87 verify`	`616036` cycles	`617767` cycles	`1.00`

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

github-actions

Arm Cortex-A55 (Snapdragon 888) benchmarks (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`463003` cycles	`462947` cycles	`1.00`
`ML-DSA-44 sign`	`2134532` cycles	`2132913` cycles	`1.00`
`ML-DSA-44 verify`	`555002` cycles	`555065` cycles	`1.00`
`ML-DSA-65 keypair`	`782019` cycles	`781178` cycles	`1.00`
`ML-DSA-65 sign`	`3483165` cycles	`3495803` cycles	`1.00`
`ML-DSA-65 verify`	`865196` cycles	`863872` cycles	`1.00`
`ML-DSA-87 keypair`	`1261848` cycles	`1265586` cycles	`1.00`
`ML-DSA-87 sign`	`4308043` cycles	`4308959` cycles	`1.00`
`ML-DSA-87 verify`	`1385965` cycles	`1390453` cycles	`1.00`

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

github-actions

Arm Cortex-A72 (Raspberry Pi 4) benchmarks (opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`233714` cycles	`214312` cycles	`1.09`
`ML-DSA-44 sign`	`634614` cycles	`605496` cycles	`1.05`
`ML-DSA-44 verify`	`228174` cycles	`222932` cycles	`1.02`
`ML-DSA-65 keypair`	`400744` cycles	`390142` cycles	`1.03`
`ML-DSA-65 sign`	`1028807` cycles	`1004679` cycles	`1.02`
`ML-DSA-65 verify`	`375697` cycles	`374059` cycles	`1.00`
`ML-DSA-87 keypair`	`669839` cycles	`652450` cycles	`1.03`
`ML-DSA-87 sign`	`1386322` cycles	`1336385` cycles	`1.04`
`ML-DSA-87 verify`	`650101` cycles	`633167` cycles	`1.03`

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

github-actions

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Arm Cortex-A72 (Raspberry Pi 4) benchmarks (opt)'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.03.

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`233714` cycles	`214312` cycles	`1.09`
`ML-DSA-44 sign`	`634614` cycles	`605496` cycles	`1.05`
`ML-DSA-87 sign`	`1386322` cycles	`1336385` cycles	`1.04`

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

oqs-bot · 2026-06-15T00:17:07Z

CBMC Results (ML-DSA-65, REDUCE-RAM)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	1581s	1504s	+5.1%
`mld_invntt_layer`	✅	174s	163s	+7%
`poly_pointwise_montgomery_c`	✅	134s	126s	+6%
`rej_uniform_native`	✅	130s	119s	+9%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	80s	81s	-1%
`mld_ct_memcmp`	✅	67s	62s	+8%
`mld_ntt_layer`	✅	43s	42s	+2%
`fqmul`	✅	42s	37s	+14%
`polyveck_chknorm`	✅	38s	39s	-3%
`mld_ntt_butterfly_block`	✅	23s	21s	+10%
`keccakf1600x4_permute_native`	✅	22s	22s	+0%
`mld_attempt_signature_generation`	✅	21s	24s	-12%
`poly_chknorm_c`	✅	19s	21s	-10%
`polyveck_decompose`	✅	17s	20s	-15%
`polyt0_unpack`	✅	16s	17s	-6%
`rej_uniform_c`	✅	16s	17s	-6%
`mld_check_pct`	✅	15s	13s	+15%
`polyvecl_chknorm`	✅	15s	13s	+15%
`sign_verify_internal`	✅	15s	13s	+15%
`poly_add`	✅	13s	12s	+8%
`poly_uniform_eta_4x`	✅	12s	11s	+9%
`polyvec_matrix_pointwise_montgomery_row`	✅	10s	7s	+43%
`polyveck_invntt_tomont`	✅	10s	6s	+67%
`keccak_absorb_once_x4`	✅	9s	10s	-10%
`poly_invntt_tomont_c`	✅	9s	6s	+50%
`polyvecl_ntt`	✅	8s	7s	+14%
`compute_pack_t0_t1`	✅	7s	8s	-12%
`keccak_absorb`	✅	7s	7s	+0%
`pointwise_acc_native_aarch64`	✅	7s	6s	+17%
`polyveck_caddq`	✅	7s	9s	-22%
`polyz_unpack_c`	✅	7s	6s	+17%
`sign`	✅	7s	8s	-12%
`keccak_init`	✅	6s	3s	+100%
`keccak_squeezeblocks_x4`	✅	6s	5s	+20%
`keccakf1600x4_extract_bytes`	✅	6s	5s	+20%
`mld_h`	✅	6s	4s	+50%
`mld_keccakf1600_permute_c`	✅	6s	7s	-14%
`pointwise_acc_native_x86_64`	✅	6s	6s	+0%
`poly_reduce`	✅	6s	5s	+20%
`polyeta_unpack`	✅	6s	4s	+50%
`polyveck_reduce`	✅	6s	4s	+50%
`rej_uniform`	✅	6s	8s	-25%
`sign_pk_from_sk`	✅	6s	6s	+0%
`sign_signature_extmu`	✅	6s	3s	+100%
`sign_verify_pre_hash_shake256`	✅	6s	5s	+20%
`make_hint`	✅	5s	2s	+150%
`mld_compute_pack_z`	✅	5s	6s	-17%
`mld_ct_abs_i32`	✅	5s	3s	+67%
`montgomery_reduce`	✅	5s	3s	+67%
`ntt_native_x86_64`	✅	5s	3s	+67%
`nttunpack_native_x86_64`	✅	5s	2s	+150%
`pack_sig_z`	✅	5s	5s	+0%
`poly_caddq`	✅	5s	2s	+150%
`poly_challenge`	✅	5s	3s	+67%
`poly_permute_bitrev_to_custom_optional_native`	✅	5s	2s	+150%
`poly_power2round`	✅	5s	7s	-29%
`poly_uniform`	✅	5s	5s	+0%
`poly_uniform_eta`	✅	5s	3s	+67%
`polyz_unpack_17_native_aarch64`	✅	5s	1s	+400%
`sign_keypair`	✅	5s	4s	+25%
`sign_keypair_internal`	✅	5s	3s	+67%
`sign_signature`	✅	5s	4s	+25%
`sign_signature_pre_hash_shake256`	✅	5s	6s	-17%
`sign_verify`	✅	5s	3s	+67%
`sign_verify_pre_hash_internal`	✅	5s	5s	+0%
`yvec_get_poly`	✅	5s	1s	+400%
`fqscale`	✅	4s	2s	+100%
`keccak_f1600_x4_native_aarch64_v84a`	✅	4s	3s	+33%
`keccak_squeeze`	✅	4s	2s	+100%
`keccakf1600x4_xor_bytes`	✅	4s	3s	+33%
`keccakf1600x4_xor_bytes_native`	✅	4s	2s	+100%
`mld_ct_cmask_neg_i32`	✅	4s	2s	+100%
`mld_sample_s1_s2`	✅	4s	4s	+0%
`ntt_native_aarch64`	✅	4s	3s	+33%
`pack_sk_rho_key_tr_s2`	✅	4s	4s	+0%
`poly_caddq_c`	✅	4s	4s	+0%
`poly_caddq_native`	✅	4s	3s	+33%
`poly_caddq_native_aarch64`	✅	4s	2s	+100%
`poly_caddq_native_x86_64`	✅	4s	2s	+100%
`poly_chknorm_native`	✅	4s	3s	+33%
`poly_chknorm_native_x86_64`	✅	4s	2s	+100%
`poly_decompose`	✅	4s	3s	+33%
`poly_decompose_c`	✅	4s	2s	+100%
`poly_ntt`	✅	4s	2s	+100%
`poly_permute_bitrev_to_custom_optional`	✅	4s	2s	+100%
`poly_pointwise_montgomery`	✅	4s	3s	+33%
`poly_shiftl`	✅	4s	6s	-33%
`poly_use_hint_c`	✅	4s	5s	-20%
`poly_use_hint_native`	✅	4s	2s	+100%
`polyveck_ntt`	✅	4s	2s	+100%
`polyw1_pack_88`	✅	4s	4s	+0%
`polyz_unpack_19_native_aarch64`	✅	4s	2s	+100%
`rej_eta`	✅	4s	3s	+33%
`rej_eta_native`	✅	4s	2s	+100%
`rej_uniform_eta_native_aarch64`	✅	4s	4s	+0%
`shake256_finalize`	✅	4s	2s	+100%
`shake256_release`	✅	4s	3s	+33%
`sig_unpack_hints`	✅	4s	3s	+33%
`sign_signature_internal`	✅	4s	3s	+33%
`sign_signature_pre_hash_internal`	✅	4s	4s	+0%
`sign_verify_extmu`	✅	4s	3s	+33%
`unpack_sk_s2hat`	✅	4s	3s	+33%
`yvec_init`	✅	4s	3s	+33%
`caddq`	✅	3s	4s	-25%
`decompose`	✅	3s	2s	+50%
`keccak_f1600_x1_native_aarch64`	✅	3s	3s	+0%
`keccak_f1600_x1_native_aarch64_v84a`	✅	3s	3s	+0%
`keccak_f1600_x4_native_avx2`	✅	3s	2s	+50%
`keccakf1600_extract_bytes (big endian)`	✅	3s	3s	+0%
`mld_ct_cmask_nonzero_u32`	✅	3s	3s	+0%
`mld_ct_get_optblocker_u8`	✅	3s	3s	+0%
`mld_prepare_domain_separation_prefix`	✅	3s	3s	+0%
`mld_sample_s1_s2_serial`	✅	3s	2s	+50%
`mld_value_barrier_i64`	✅	3s	4s	-25%
`mld_value_barrier_u32`	✅	3s	2s	+50%
`pack_sig_c`	✅	3s	5s	-40%
`pointwise_native_aarch64`	✅	3s	4s	-25%
`pointwise_native_x86_64`	✅	3s	3s	+0%
`poly_chknorm`	✅	3s	1s	+200%
`poly_chknorm_native_aarch64`	✅	3s	1s	+200%
`poly_decompose_32_native_aarch64`	✅	3s	3s	+0%
`poly_decompose_88_native_aarch64`	✅	3s	1s	+200%
`poly_decompose_native`	✅	3s	4s	-25%
`poly_invntt_tomont`	✅	3s	2s	+50%
`poly_invntt_tomont_native`	✅	3s	4s	-25%
`poly_ntt_native`	✅	3s	3s	+0%
`poly_pointwise_montgomery_native`	✅	3s	4s	-25%
`poly_uniform_gamma1_4x`	✅	3s	2s	+50%
`poly_use_hint`	✅	3s	2s	+50%
`poly_use_hint_native_aarch64`	✅	3s	2s	+50%
`polyeta_pack`	✅	3s	6s	-50%
`polyt0_pack`	✅	3s	4s	-25%
`polyt1_unpack`	✅	3s	4s	-25%
`polyvec_matrix_expand_serial`	✅	3s	2s	+50%
`polyvecl_pack_eta`	✅	3s	4s	-25%
`polyvecl_pointwise_acc_montgomery_c`	✅	3s	4s	-25%
`polyvecl_pointwise_acc_montgomery_native`	✅	3s	3s	+0%
`polyvecl_uniform_gamma1`	✅	3s	2s	+50%
`polyvecl_unpack_eta`	✅	3s	2s	+50%
`polyvecl_unpack_z`	✅	3s	2s	+50%
`polyz_pack`	✅	3s	3s	+0%
`power2round`	✅	3s	3s	+0%
`rej_eta_c`	✅	3s	2s	+50%
`rej_uniform_native_aarch64`	✅	3s	2s	+50%
`shake128_init`	✅	3s	4s	-25%
`shake128_release`	✅	3s	1s	+200%
`shake128x4_absorb_once`	✅	3s	6s	-50%
`shake256x4_absorb_once`	✅	3s	4s	-25%
`shake256x4_squeezeblocks`	✅	3s	3s	+0%
`sign_open`	✅	3s	2s	+50%
`sk_t0hat_get_poly`	✅	3s	2s	+50%
`sys_check_capability`	✅	3s	4s	-25%
`unpack_pk_t1`	✅	3s	4s	-25%
`unpack_sk`	✅	3s	2s	+50%
`use_hint`	✅	3s	3s	+0%
`intt_native_aarch64`	✅	2s	2s	+0%
`intt_native_x86_64`	✅	2s	4s	-50%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	2s	2s	+0%
`keccak_finalize`	✅	2s	5s	-60%
`keccakf1600_permute_native`	✅	2s	1s	+100%
`keccakf1600_xor_bytes`	✅	2s	1s	+100%
`keccakf1600_xor_bytes (big endian)`	✅	2s	2s	+0%
`keccakf1600x4_extract_bytes_native`	✅	2s	3s	-33%
`keccakf1600x4_permute`	✅	2s	2s	+0%
`mld_ct_cmask_nonzero_u8`	✅	2s	5s	-60%
`mld_ct_get_optblocker_i64`	✅	2s	2s	+0%
`mld_ct_get_optblocker_u32`	✅	2s	3s	-33%
`mld_ct_sel_int32`	✅	2s	1s	+100%
`mld_keccakf1600_extract_bytes`	✅	2s	5s	-60%
`mld_keccakf1600x4_extract_bytes_c`	✅	2s	2s	+0%
`mld_keccakf1600x4_xor_bytes_c`	✅	2s	3s	-33%
`mld_polymat_expand_entry`	✅	2s	3s	-33%
`mld_value_barrier_u8`	✅	2s	2s	+0%
`pack_sig_h`	✅	2s	5s	-60%
`pack_sk_s1`	✅	2s	4s	-50%
`poly_ntt_c`	✅	2s	4s	-50%
`poly_sub`	✅	2s	3s	-33%
`poly_uniform_4x`	✅	2s	2s	+0%
`polyt1_pack`	✅	2s	2s	+0%
`polyveck_pack_eta`	✅	2s	3s	-33%
`polyveck_pack_w1`	✅	2s	6s	-67%
`polyveck_unpack_eta`	✅	2s	4s	-50%
`polyvecl_pointwise_acc_montgomery`	✅	2s	2s	+0%
`polyw1_pack`	✅	2s	7s	-71%
`polyw1_pack_32`	✅	2s	4s	-50%
`polyz_unpack`	✅	2s	3s	-33%
`polyz_unpack_native`	✅	2s	2s	+0%
`reduce32`	✅	2s	4s	-50%
`shake128_finalize`	✅	2s	2s	+0%
`shake128_squeeze`	✅	2s	4s	-50%
`shake128x4_squeezeblocks`	✅	2s	3s	-33%
`shake256`	✅	2s	3s	-33%
`shake256_absorb`	✅	2s	5s	-60%
`shake256_init`	✅	2s	2s	+0%
`shake256_squeeze`	✅	2s	3s	-33%
`sk_s1hat_get_poly`	✅	2s	1s	+100%
`sk_s2hat_get_poly`	✅	2s	3s	-33%
`unpack_sk_s1hat`	✅	2s	3s	-33%
`unpack_sk_t0hat`	✅	2s	3s	-33%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	1s	3s	-67%
`keccakf1600_permute`	✅	1s	2s	-50%
`poly_uniform_gamma1`	✅	1s	2s	-50%
`polyvec_matrix_expand`	✅	1s	4s	-75%
`polyvecl_uniform_gamma1_serial`	✅	1s	3s	-67%
`shake128_absorb`	✅	1s	4s	-75%

oqs-bot · 2026-06-15T00:17:10Z

CBMC Results (ML-DSA-87, REDUCE-RAM)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	1560s	1624s	-3.9%
`mld_invntt_layer`	✅	166s	172s	-3%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	152s	155s	-2%
`poly_pointwise_montgomery_c`	✅	130s	136s	-4%
`rej_uniform_native`	✅	123s	125s	-2%
`mld_ct_memcmp`	✅	63s	65s	-3%
`mld_ntt_layer`	✅	42s	43s	-2%
`fqmul`	✅	39s	43s	-9%
`mld_attempt_signature_generation`	✅	33s	37s	-11%
`sign_verify_internal`	✅	23s	22s	+5%
`keccakf1600x4_permute_native`	✅	22s	26s	-15%
`mld_ntt_butterfly_block`	✅	22s	23s	-4%
`poly_chknorm_c`	✅	19s	18s	+6%
`polyveck_decompose`	✅	17s	18s	-6%
`mld_check_pct`	✅	16s	13s	+23%
`polyt0_unpack`	✅	16s	17s	-6%
`rej_uniform_c`	✅	15s	16s	-6%
`poly_uniform_eta_4x`	✅	13s	13s	+0%
`polyeta_unpack`	✅	13s	14s	-7%
`poly_add`	✅	11s	11s	+0%
`poly_invntt_tomont_c`	✅	11s	9s	+22%
`compute_pack_t0_t1`	✅	10s	11s	-9%
`mld_keccakf1600_permute_c`	✅	9s	6s	+50%
`polyvecl_chknorm`	✅	9s	10s	-10%
`polyvecl_ntt`	✅	9s	9s	+0%
`sign`	✅	9s	8s	+12%
`keccak_absorb_once_x4`	✅	8s	10s	-20%
`poly_decompose_c`	✅	8s	5s	+60%
`rej_uniform`	✅	8s	8s	+0%
`keccak_absorb`	✅	7s	7s	+0%
`pointwise_acc_native_aarch64`	✅	7s	8s	-12%
`polyvec_matrix_pointwise_montgomery_row`	✅	7s	8s	-12%
`polyveck_invntt_tomont`	✅	7s	7s	+0%
`rej_eta_native`	✅	7s	3s	+133%
`sign_keypair_internal`	✅	7s	7s	+0%
`mld_sample_s1_s2_serial`	✅	6s	5s	+20%
`pointwise_acc_native_x86_64`	✅	6s	6s	+0%
`sign_signature_internal`	✅	6s	5s	+20%
`sign_signature_pre_hash_internal`	✅	6s	3s	+100%
`mld_ct_cmask_nonzero_u32`	✅	5s	5s	+0%
`mld_polymat_expand_entry`	✅	5s	5s	+0%
`pack_sk_rho_key_tr_s2`	✅	5s	3s	+67%
`poly_challenge`	✅	5s	6s	-17%
`poly_decompose_native`	✅	5s	2s	+150%
`poly_invntt_tomont_native`	✅	5s	2s	+150%
`poly_pointwise_montgomery`	✅	5s	3s	+67%
`poly_power2round`	✅	5s	5s	+0%
`polyveck_pack_w1`	✅	5s	3s	+67%
`polyz_unpack_c`	✅	5s	6s	-17%
`rej_uniform_eta_native_aarch64`	✅	5s	5s	+0%
`sign_pk_from_sk`	✅	5s	6s	-17%
`caddq`	✅	4s	2s	+100%
`fqscale`	✅	4s	3s	+33%
`intt_native_aarch64`	✅	4s	3s	+33%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	4s	2s	+100%
`keccakf1600_xor_bytes (big endian)`	✅	4s	2s	+100%
`keccakf1600x4_permute`	✅	4s	2s	+100%
`make_hint`	✅	4s	2s	+100%
`mld_compute_pack_z`	✅	4s	3s	+33%
`mld_ct_cmask_nonzero_u8`	✅	4s	3s	+33%
`mld_ct_get_optblocker_u8`	✅	4s	2s	+100%
`mld_keccakf1600x4_xor_bytes_c`	✅	4s	3s	+33%
`mld_sample_s1_s2`	✅	4s	4s	+0%
`ntt_native_aarch64`	✅	4s	8s	-50%
`pack_sig_h`	✅	4s	3s	+33%
`poly_caddq_native_aarch64`	✅	4s	3s	+33%
`poly_chknorm`	✅	4s	2s	+100%
`poly_decompose_88_native_aarch64`	✅	4s	4s	+0%
`poly_ntt_c`	✅	4s	3s	+33%
`poly_reduce`	✅	4s	4s	+0%
`poly_sub`	✅	4s	1s	+300%
`poly_uniform_gamma1_4x`	✅	4s	5s	-20%
`poly_use_hint_c`	✅	4s	2s	+100%
`polyeta_pack`	✅	4s	4s	+0%
`polyveck_caddq`	✅	4s	5s	-20%
`polyveck_unpack_eta`	✅	4s	4s	+0%
`polyvecl_uniform_gamma1_serial`	✅	4s	5s	-20%
`polyw1_pack_32`	✅	4s	3s	+33%
`polyz_unpack_17_native_aarch64`	✅	4s	2s	+100%
`rej_eta_c`	✅	4s	4s	+0%
`rej_uniform_native_aarch64`	✅	4s	4s	+0%
`shake256`	✅	4s	4s	+0%
`shake256_init`	✅	4s	4s	+0%
`sign_keypair`	✅	4s	3s	+33%
`sign_signature`	✅	4s	3s	+33%
`sign_signature_extmu`	✅	4s	4s	+0%
`sign_signature_pre_hash_shake256`	✅	4s	5s	-20%
`sign_verify`	✅	4s	6s	-33%
`sign_verify_extmu`	✅	4s	3s	+33%
`sign_verify_pre_hash_internal`	✅	4s	4s	+0%
`yvec_get_poly`	✅	4s	2s	+100%
`decompose`	✅	3s	3s	+0%
`keccak_f1600_x1_native_aarch64_v84a`	✅	3s	3s	+0%
`keccak_f1600_x4_native_avx2`	✅	3s	4s	-25%
`keccak_init`	✅	3s	3s	+0%
`keccakf1600x4_extract_bytes_native`	✅	3s	2s	+50%
`mld_h`	✅	3s	4s	-25%
`mld_keccakf1600x4_extract_bytes_c`	✅	3s	2s	+50%
`mld_value_barrier_i64`	✅	3s	3s	+0%
`mld_value_barrier_u32`	✅	3s	2s	+50%
`ntt_native_x86_64`	✅	3s	6s	-50%
`nttunpack_native_x86_64`	✅	3s	4s	-25%
`pack_sig_c`	✅	3s	3s	+0%
`pack_sig_z`	✅	3s	3s	+0%
`pointwise_native_aarch64`	✅	3s	3s	+0%
`pointwise_native_x86_64`	✅	3s	4s	-25%
`poly_caddq`	✅	3s	4s	-25%
`poly_caddq_c`	✅	3s	3s	+0%
`poly_chknorm_native_aarch64`	✅	3s	4s	-25%
`poly_decompose_32_native_aarch64`	✅	3s	2s	+50%
`poly_invntt_tomont`	✅	3s	6s	-50%
`poly_ntt`	✅	3s	4s	-25%
`poly_permute_bitrev_to_custom_optional`	✅	3s	2s	+50%
`poly_shiftl`	✅	3s	3s	+0%
`poly_use_hint_native`	✅	3s	3s	+0%
`polyt1_pack`	✅	3s	4s	-25%
`polyt1_unpack`	✅	3s	3s	+0%
`polyvec_matrix_expand`	✅	3s	2s	+50%
`polyveck_ntt`	✅	3s	2s	+50%
`polyveck_pack_eta`	✅	3s	2s	+50%
`polyveck_reduce`	✅	3s	6s	-50%
`polyvecl_pointwise_acc_montgomery_native`	✅	3s	3s	+0%
`polyvecl_uniform_gamma1`	✅	3s	3s	+0%
`polyvecl_unpack_eta`	✅	3s	2s	+50%
`polyw1_pack_88`	✅	3s	2s	+50%
`power2round`	✅	3s	3s	+0%
`rej_eta`	✅	3s	3s	+0%
`shake128_absorb`	✅	3s	2s	+50%
`shake128_init`	✅	3s	2s	+50%
`shake256x4_absorb_once`	✅	3s	2s	+50%
`sign_open`	✅	3s	5s	-40%
`sign_verify_pre_hash_shake256`	✅	3s	3s	+0%
`sk_t0hat_get_poly`	✅	3s	2s	+50%
`unpack_sk`	✅	3s	4s	-25%
`unpack_sk_s2hat`	✅	3s	3s	+0%
`intt_native_x86_64`	✅	2s	4s	-50%
`keccak_f1600_x4_native_aarch64_v84a`	✅	2s	2s	+0%
`keccak_squeeze`	✅	2s	3s	-33%
`keccak_squeezeblocks_x4`	✅	2s	4s	-50%
`keccakf1600_extract_bytes (big endian)`	✅	2s	4s	-50%
`keccakf1600_permute`	✅	2s	2s	+0%
`keccakf1600_permute_native`	✅	2s	3s	-33%
`keccakf1600x4_xor_bytes`	✅	2s	2s	+0%
`keccakf1600x4_xor_bytes_native`	✅	2s	1s	+100%
`mld_ct_abs_i32`	✅	2s	3s	-33%
`mld_ct_get_optblocker_i64`	✅	2s	1s	+100%
`mld_ct_get_optblocker_u32`	✅	2s	2s	+0%
`mld_ct_sel_int32`	✅	2s	2s	+0%
`mld_keccakf1600_extract_bytes`	✅	2s	3s	-33%
`mld_prepare_domain_separation_prefix`	✅	2s	3s	-33%
`mld_value_barrier_u8`	✅	2s	2s	+0%
`montgomery_reduce`	✅	2s	2s	+0%
`pack_sk_s1`	✅	2s	2s	+0%
`poly_caddq_native`	✅	2s	4s	-50%
`poly_caddq_native_x86_64`	✅	2s	3s	-33%
`poly_chknorm_native`	✅	2s	4s	-50%
`poly_chknorm_native_x86_64`	✅	2s	4s	-50%
`poly_decompose`	✅	2s	2s	+0%
`poly_ntt_native`	✅	2s	4s	-50%
`poly_permute_bitrev_to_custom_optional_native`	✅	2s	4s	-50%
`poly_pointwise_montgomery_native`	✅	2s	5s	-60%
`poly_uniform`	✅	2s	4s	-50%
`poly_uniform_4x`	✅	2s	3s	-33%
`poly_uniform_eta`	✅	2s	4s	-50%
`poly_uniform_gamma1`	✅	2s	3s	-33%
`poly_use_hint`	✅	2s	3s	-33%
`poly_use_hint_native_aarch64`	✅	2s	2s	+0%
`polyt0_pack`	✅	2s	4s	-50%
`polyvec_matrix_expand_serial`	✅	2s	4s	-50%
`polyveck_chknorm`	✅	2s	3s	-33%
`polyvecl_pack_eta`	✅	2s	2s	+0%
`polyvecl_pointwise_acc_montgomery`	✅	2s	4s	-50%
`polyvecl_unpack_z`	✅	2s	3s	-33%
`polyw1_pack`	✅	2s	2s	+0%
`polyz_pack`	✅	2s	1s	+100%
`polyz_unpack`	✅	2s	3s	-33%
`polyz_unpack_19_native_aarch64`	✅	2s	3s	-33%
`polyz_unpack_native`	✅	2s	3s	-33%
`reduce32`	✅	2s	5s	-60%
`shake128_squeeze`	✅	2s	1s	+100%
`shake128x4_absorb_once`	✅	2s	2s	+0%
`shake128x4_squeezeblocks`	✅	2s	2s	+0%
`shake256_absorb`	✅	2s	2s	+0%
`shake256_squeeze`	✅	2s	3s	-33%
`shake256x4_squeezeblocks`	✅	2s	3s	-33%
`sig_unpack_hints`	✅	2s	3s	-33%
`sk_s1hat_get_poly`	✅	2s	2s	+0%
`sk_s2hat_get_poly`	✅	2s	3s	-33%
`unpack_pk_t1`	✅	2s	3s	-33%
`unpack_sk_t0hat`	✅	2s	7s	-71%
`use_hint`	✅	2s	4s	-50%
`keccak_f1600_x1_native_aarch64`	✅	1s	2s	-50%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	1s	3s	-67%
`keccak_finalize`	✅	1s	2s	-50%
`keccakf1600_xor_bytes`	✅	1s	3s	-67%
`keccakf1600x4_extract_bytes`	✅	1s	2s	-50%
`mld_ct_cmask_neg_i32`	✅	1s	2s	-50%
`polyvecl_pointwise_acc_montgomery_c`	✅	1s	3s	-67%
`shake128_finalize`	✅	1s	2s	-50%
`shake128_release`	✅	1s	2s	-50%
`shake256_finalize`	✅	1s	1s	+0%
`shake256_release`	✅	1s	1s	+0%
`sys_check_capability`	✅	1s	4s	-75%
`unpack_sk_s1hat`	✅	1s	3s	-67%
`yvec_init`	✅	1s	2s	-50%

github-actions

Arm Cortex-A72 (Raspberry Pi 4) benchmarks (no-opt)

Details

Benchmark suite	Current: `24a79f0`	Previous: `9c5eceb`	Ratio
`ML-DSA-44 keypair`	`299345` cycles	`312767` cycles	`0.96`
`ML-DSA-44 sign`	`1146844` cycles	`1207539` cycles	`0.95`
`ML-DSA-44 verify`	`340467` cycles	`342752` cycles	`0.99`
`ML-DSA-65 keypair`	`551047` cycles	`567893` cycles	`0.97`
`ML-DSA-65 sign`	`1875176` cycles	`1961247` cycles	`0.96`
`ML-DSA-65 verify`	`537953` cycles	`560470` cycles	`0.96`
`ML-DSA-87 keypair`	`851210` cycles	`849145` cycles	`1.00`
`ML-DSA-87 sign`	`2389528` cycles	`2376735` cycles	`1.01`
`ML-DSA-87 verify`	`881000` cycles	`890655` cycles	`0.99`

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

oqs-bot · 2026-06-15T00:17:54Z

CBMC Results (ML-DSA-44, REDUCE-RAM)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	1613s	1453s	+11.0%
`mld_invntt_layer`	✅	183s	147s	+24%
`poly_pointwise_montgomery_c`	✅	144s	125s	+15%
`rej_uniform_native`	✅	133s	119s	+12%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	130s	112s	+16%
`mld_ct_memcmp`	✅	72s	65s	+11%
`mld_ntt_layer`	✅	48s	42s	+14%
`fqmul`	✅	45s	40s	+12%
`mld_attempt_signature_generation`	✅	25s	25s	+0%
`keccakf1600x4_permute_native`	✅	24s	23s	+4%
`mld_ntt_butterfly_block`	✅	24s	21s	+14%
`sign_verify_internal`	✅	22s	20s	+10%
`poly_chknorm_c`	✅	18s	18s	+0%
`polyeta_unpack`	✅	16s	16s	+0%
`polyt0_unpack`	✅	15s	16s	-6%
`rej_uniform_c`	✅	15s	15s	+0%
`poly_uniform_eta_4x`	✅	13s	14s	-7%
`compute_pack_t0_t1`	✅	11s	9s	+22%
`mld_check_pct`	✅	11s	12s	-8%
`polyvec_matrix_pointwise_montgomery_row`	✅	11s	8s	+38%
`poly_add`	✅	10s	9s	+11%
`polyveck_chknorm`	✅	10s	6s	+67%
`polyz_unpack_c`	✅	10s	11s	-9%
`keccak_absorb_once_x4`	✅	9s	10s	-10%
`poly_invntt_tomont_c`	✅	9s	8s	+12%
`polyveck_decompose`	✅	9s	6s	+50%
`rej_uniform`	✅	8s	10s	-20%
`keccak_squeezeblocks_x4`	✅	7s	4s	+75%
`polyvecl_ntt`	✅	7s	3s	+133%
`sign_signature_internal`	✅	7s	4s	+75%
`keccak_absorb`	✅	6s	6s	+0%
`mld_ct_cmask_nonzero_u32`	✅	6s	1s	+500%
`mld_keccakf1600_permute_c`	✅	6s	8s	-25%
`pointwise_acc_native_aarch64`	✅	6s	4s	+50%
`pointwise_acc_native_x86_64`	✅	6s	5s	+20%
`poly_chknorm_native`	✅	6s	4s	+50%
`poly_ntt_c`	✅	6s	3s	+100%
`poly_power2round`	✅	6s	6s	+0%
`rej_eta`	✅	6s	4s	+50%
`sign`	✅	6s	7s	-14%
`sk_s1hat_get_poly`	✅	6s	3s	+100%
`decompose`	✅	5s	1s	+400%
`keccakf1600x4_xor_bytes_native`	✅	5s	3s	+67%
`mld_compute_pack_z`	✅	5s	5s	+0%
`mld_polymat_expand_entry`	✅	5s	2s	+150%
`mld_sample_s1_s2`	✅	5s	8s	-38%
`mld_sample_s1_s2_serial`	✅	5s	3s	+67%
`ntt_native_x86_64`	✅	5s	1s	+400%
`pack_sig_z`	✅	5s	3s	+67%
`poly_decompose_c`	✅	5s	7s	-29%
`poly_shiftl`	✅	5s	4s	+25%
`poly_uniform_eta`	✅	5s	5s	+0%
`polyvec_matrix_expand_serial`	✅	5s	5s	+0%
`polyveck_caddq`	✅	5s	6s	-17%
`polyveck_invntt_tomont`	✅	5s	3s	+67%
`polyveck_reduce`	✅	5s	6s	-17%
`polyveck_unpack_eta`	✅	5s	2s	+150%
`polyvecl_unpack_eta`	✅	5s	2s	+150%
`rej_eta_c`	✅	5s	3s	+67%
`sign_open`	✅	5s	5s	+0%
`sign_signature`	✅	5s	3s	+67%
`sign_signature_pre_hash_internal`	✅	5s	4s	+25%
`sign_verify_pre_hash_shake256`	✅	5s	4s	+25%
`unpack_pk_t1`	✅	5s	2s	+150%
`yvec_init`	✅	5s	4s	+25%
`caddq`	✅	4s	1s	+300%
`intt_native_x86_64`	✅	4s	2s	+100%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	4s	3s	+33%
`keccakf1600x4_extract_bytes_native`	✅	4s	1s	+300%
`mld_ct_abs_i32`	✅	4s	2s	+100%
`mld_keccakf1600_extract_bytes`	✅	4s	2s	+100%
`montgomery_reduce`	✅	4s	2s	+100%
`ntt_native_aarch64`	✅	4s	3s	+33%
`pack_sk_s1`	✅	4s	3s	+33%
`pointwise_native_x86_64`	✅	4s	3s	+33%
`poly_invntt_tomont`	✅	4s	3s	+33%
`poly_uniform`	✅	4s	5s	-20%
`poly_uniform_gamma1`	✅	4s	3s	+33%
`poly_use_hint_native`	✅	4s	4s	+0%
`polyeta_pack`	✅	4s	2s	+100%
`polyt1_pack`	✅	4s	5s	-20%
`polyveck_pack_w1`	✅	4s	2s	+100%
`polyvecl_chknorm`	✅	4s	9s	-56%
`polyvecl_pack_eta`	✅	4s	2s	+100%
`polyvecl_pointwise_acc_montgomery_native`	✅	4s	1s	+300%
`polyz_unpack`	✅	4s	3s	+33%
`polyz_unpack_17_native_aarch64`	✅	4s	2s	+100%
`polyz_unpack_native`	✅	4s	3s	+33%
`reduce32`	✅	4s	3s	+33%
`shake128_absorb`	✅	4s	3s	+33%
`shake128_release`	✅	4s	2s	+100%
`sig_unpack_hints`	✅	4s	2s	+100%
`sign_keypair`	✅	4s	4s	+0%
`sign_keypair_internal`	✅	4s	3s	+33%
`sign_pk_from_sk`	✅	4s	7s	-43%
`sign_signature_extmu`	✅	4s	4s	+0%
`fqscale`	✅	3s	2s	+50%
`intt_native_aarch64`	✅	3s	2s	+50%
`keccak_squeeze`	✅	3s	1s	+200%
`keccakf1600x4_xor_bytes`	✅	3s	2s	+50%
`mld_ct_cmask_nonzero_u8`	✅	3s	1s	+200%
`mld_ct_get_optblocker_i64`	✅	3s	2s	+50%
`mld_ct_get_optblocker_u8`	✅	3s	3s	+0%
`mld_h`	✅	3s	3s	+0%
`mld_keccakf1600x4_extract_bytes_c`	✅	3s	1s	+200%
`mld_value_barrier_u32`	✅	3s	1s	+200%
`nttunpack_native_x86_64`	✅	3s	2s	+50%
`pack_sig_h`	✅	3s	5s	-40%
`pointwise_native_aarch64`	✅	3s	3s	+0%
`poly_caddq_c`	✅	3s	4s	-25%
`poly_caddq_native_aarch64`	✅	3s	1s	+200%
`poly_caddq_native_x86_64`	✅	3s	1s	+200%
`poly_chknorm`	✅	3s	2s	+50%
`poly_chknorm_native_x86_64`	✅	3s	4s	-25%
`poly_decompose_32_native_aarch64`	✅	3s	1s	+200%
`poly_decompose_88_native_aarch64`	✅	3s	4s	-25%
`poly_decompose_native`	✅	3s	5s	-40%
`poly_invntt_tomont_native`	✅	3s	2s	+50%
`poly_pointwise_montgomery_native`	✅	3s	2s	+50%
`poly_reduce`	✅	3s	4s	-25%
`poly_sub`	✅	3s	4s	-25%
`poly_use_hint_native_aarch64`	✅	3s	4s	-25%
`polyt0_pack`	✅	3s	4s	-25%
`polyvec_matrix_expand`	✅	3s	4s	-25%
`polyveck_pack_eta`	✅	3s	3s	+0%
`polyvecl_pointwise_acc_montgomery_c`	✅	3s	3s	+0%
`polyvecl_unpack_z`	✅	3s	2s	+50%
`polyw1_pack`	✅	3s	1s	+200%
`polyw1_pack_32`	✅	3s	3s	+0%
`polyw1_pack_88`	✅	3s	3s	+0%
`power2round`	✅	3s	3s	+0%
`rej_eta_native`	✅	3s	3s	+0%
`rej_uniform_native_aarch64`	✅	3s	5s	-40%
`shake128_finalize`	✅	3s	2s	+50%
`shake128_squeeze`	✅	3s	2s	+50%
`shake256_init`	✅	3s	5s	-40%
`shake256_release`	✅	3s	1s	+200%
`shake256_squeeze`	✅	3s	4s	-25%
`sign_signature_pre_hash_shake256`	✅	3s	3s	+0%
`sign_verify`	✅	3s	3s	+0%
`sign_verify_extmu`	✅	3s	2s	+50%
`sk_s2hat_get_poly`	✅	3s	4s	-25%
`sk_t0hat_get_poly`	✅	3s	2s	+50%
`sys_check_capability`	✅	3s	3s	+0%
`unpack_sk`	✅	3s	2s	+50%
`unpack_sk_s2hat`	✅	3s	3s	+0%
`keccak_f1600_x1_native_aarch64`	✅	2s	3s	-33%
`keccak_f1600_x1_native_aarch64_v84a`	✅	2s	3s	-33%
`keccak_f1600_x4_native_aarch64_v84a`	✅	2s	2s	+0%
`keccak_f1600_x4_native_avx2`	✅	2s	4s	-50%
`keccak_finalize`	✅	2s	3s	-33%
`keccakf1600_extract_bytes (big endian)`	✅	2s	4s	-50%
`keccakf1600_permute_native`	✅	2s	1s	+100%
`keccakf1600_xor_bytes (big endian)`	✅	2s	2s	+0%
`keccakf1600x4_extract_bytes`	✅	2s	4s	-50%
`make_hint`	✅	2s	4s	-50%
`mld_ct_sel_int32`	✅	2s	3s	-33%
`mld_prepare_domain_separation_prefix`	✅	2s	4s	-50%
`mld_value_barrier_u8`	✅	2s	2s	+0%
`pack_sig_c`	✅	2s	3s	-33%
`poly_caddq`	✅	2s	3s	-33%
`poly_caddq_native`	✅	2s	2s	+0%
`poly_challenge`	✅	2s	5s	-60%
`poly_chknorm_native_aarch64`	✅	2s	5s	-60%
`poly_decompose`	✅	2s	3s	-33%
`poly_ntt`	✅	2s	3s	-33%
`poly_ntt_native`	✅	2s	2s	+0%
`poly_permute_bitrev_to_custom_optional`	✅	2s	2s	+0%
`poly_permute_bitrev_to_custom_optional_native`	✅	2s	5s	-60%
`poly_pointwise_montgomery`	✅	2s	3s	-33%
`poly_uniform_4x`	✅	2s	3s	-33%
`poly_uniform_gamma1_4x`	✅	2s	3s	-33%
`poly_use_hint`	✅	2s	4s	-50%
`poly_use_hint_c`	✅	2s	3s	-33%
`polyt1_unpack`	✅	2s	2s	+0%
`polyveck_ntt`	✅	2s	4s	-50%
`polyvecl_pointwise_acc_montgomery`	✅	2s	3s	-33%
`polyvecl_uniform_gamma1`	✅	2s	2s	+0%
`polyvecl_uniform_gamma1_serial`	✅	2s	4s	-50%
`polyz_pack`	✅	2s	2s	+0%
`polyz_unpack_19_native_aarch64`	✅	2s	2s	+0%
`rej_uniform_eta_native_aarch64`	✅	2s	2s	+0%
`shake128_init`	✅	2s	6s	-67%
`shake128x4_absorb_once`	✅	2s	3s	-33%
`shake128x4_squeezeblocks`	✅	2s	3s	-33%
`shake256`	✅	2s	2s	+0%
`shake256_absorb`	✅	2s	3s	-33%
`shake256_finalize`	✅	2s	2s	+0%
`shake256x4_squeezeblocks`	✅	2s	1s	+100%
`sign_verify_pre_hash_internal`	✅	2s	3s	-33%
`unpack_sk_s1hat`	✅	2s	2s	+0%
`unpack_sk_t0hat`	✅	2s	1s	+100%
`use_hint`	✅	2s	1s	+100%
`yvec_get_poly`	✅	2s	1s	+100%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	1s	1s	+0%
`keccak_init`	✅	1s	4s	-75%
`keccakf1600_permute`	✅	1s	4s	-75%
`keccakf1600_xor_bytes`	✅	1s	3s	-67%
`keccakf1600x4_permute`	✅	1s	2s	-50%
`mld_ct_cmask_neg_i32`	✅	1s	1s	+0%
`mld_ct_get_optblocker_u32`	✅	1s	2s	-50%
`mld_keccakf1600x4_xor_bytes_c`	✅	1s	1s	+0%
`mld_value_barrier_i64`	✅	1s	2s	-50%
`pack_sk_rho_key_tr_s2`	✅	1s	4s	-75%
`shake256x4_absorb_once`	✅	1s	2s	-50%

oqs-bot · 2026-06-15T00:19:08Z

CBMC Results (ML-DSA-65)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	2111s	2040s	+3.5%
`mld_invntt_layer`	✅	297s	286s	+4%
`polyvecl_pointwise_acc_montgomery_c`	✅	214s	193s	+11%
`rej_uniform_native`	✅	150s	149s	+1%
`polyvec_matrix_expand`	✅	135s	130s	+4%
`poly_pointwise_montgomery_c`	✅	95s	99s	-4%
`mld_ct_memcmp`	✅	69s	67s	+3%
`mld_attempt_signature_generation`	✅	66s	66s	+0%
`sign_verify_internal`	✅	65s	64s	+2%
`sign_signature_internal`	✅	49s	48s	+2%
`mld_ntt_layer`	✅	46s	43s	+7%
`fqmul`	✅	40s	43s	-7%
`polyvec_matrix_expand_serial`	✅	25s	25s	+0%
`rej_uniform`	✅	23s	22s	+5%
`mld_ntt_butterfly_block`	✅	22s	22s	+0%
`keccakf1600x4_permute_native`	✅	21s	23s	-9%
`poly_chknorm_c`	✅	20s	19s	+5%
`polyt0_unpack`	✅	17s	18s	-6%
`compute_pack_t0_t1`	✅	16s	16s	+0%
`polyvecl_chknorm`	✅	16s	16s	+0%
`polyveck_decompose`	✅	15s	15s	+0%
`poly_uniform_4x`	✅	14s	11s	+27%
`mld_check_pct`	✅	13s	13s	+0%
`rej_uniform_c`	✅	13s	12s	+8%
`mld_compute_pack_z`	✅	12s	10s	+20%
`poly_add`	✅	11s	10s	+10%
`poly_uniform_eta_4x`	✅	11s	12s	-8%
`keccak_absorb_once_x4`	✅	10s	12s	-17%
`sign`	✅	10s	7s	+43%
`poly_caddq_c`	✅	9s	7s	+29%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	9s	10s	-10%
`polyveck_chknorm`	✅	9s	11s	-18%
`polyveck_invntt_tomont`	✅	9s	8s	+12%
`poly_invntt_tomont_c`	✅	8s	8s	+0%
`polyveck_ntt`	✅	8s	9s	-11%
`pointwise_acc_native_x86_64`	✅	7s	6s	+17%
`polyvecl_ntt`	✅	7s	6s	+17%
`sk_s1hat_get_poly`	✅	7s	2s	+250%
`keccak_squeezeblocks_x4`	✅	6s	5s	+20%
`keccakf1600x4_xor_bytes_native`	✅	6s	2s	+200%
`mld_keccakf1600_permute_c`	✅	6s	7s	-14%
`pack_sk_rho_key_tr_s2`	✅	6s	5s	+20%
`polyz_unpack_c`	✅	6s	7s	-14%
`sign_verify`	✅	6s	5s	+20%
`sk_s2hat_get_poly`	✅	6s	2s	+200%
`decompose`	✅	5s	4s	+25%
`intt_native_x86_64`	✅	5s	3s	+67%
`keccak_absorb`	✅	5s	5s	+0%
`keccakf1600x4_permute`	✅	5s	2s	+150%
`pointwise_acc_native_aarch64`	✅	5s	6s	-17%
`poly_caddq_native_x86_64`	✅	5s	2s	+150%
`poly_chknorm_native_aarch64`	✅	5s	4s	+25%
`poly_power2round`	✅	5s	4s	+25%
`poly_reduce`	✅	5s	2s	+150%
`polyeta_unpack`	✅	5s	1s	+400%
`polyw1_pack_32`	✅	5s	3s	+67%
`reduce32`	✅	5s	3s	+67%
`rej_uniform_native_aarch64`	✅	5s	4s	+25%
`shake256`	✅	5s	3s	+67%
`sign_keypair_internal`	✅	5s	2s	+150%
`sign_verify_extmu`	✅	5s	4s	+25%
`sign_verify_pre_hash_internal`	✅	5s	5s	+0%
`unpack_sk`	✅	5s	1s	+400%
`unpack_sk_t0hat`	✅	5s	6s	-17%
`fqscale`	✅	4s	5s	-20%
`keccakf1600_xor_bytes`	✅	4s	2s	+100%
`mld_ct_cmask_neg_i32`	✅	4s	3s	+33%
`mld_prepare_domain_separation_prefix`	✅	4s	3s	+33%
`mld_sample_s1_s2`	✅	4s	4s	+0%
`montgomery_reduce`	✅	4s	2s	+100%
`ntt_native_x86_64`	✅	4s	3s	+33%
`pack_sig_c`	✅	4s	4s	+0%
`pack_sig_h`	✅	4s	4s	+0%
`poly_caddq`	✅	4s	2s	+100%
`poly_decompose_32_native_aarch64`	✅	4s	3s	+33%
`poly_ntt`	✅	4s	3s	+33%
`poly_permute_bitrev_to_custom_optional_native`	✅	4s	3s	+33%
`poly_pointwise_montgomery`	✅	4s	2s	+100%
`poly_pointwise_montgomery_native`	✅	4s	3s	+33%
`poly_shiftl`	✅	4s	7s	-43%
`polyt0_pack`	✅	4s	6s	-33%
`polyveck_caddq`	✅	4s	7s	-43%
`polyveck_unpack_eta`	✅	4s	4s	+0%
`polyvecl_unpack_z`	✅	4s	3s	+33%
`polyz_unpack`	✅	4s	2s	+100%
`polyz_unpack_17_native_aarch64`	✅	4s	3s	+33%
`polyz_unpack_19_native_aarch64`	✅	4s	3s	+33%
`polyz_unpack_native`	✅	4s	2s	+100%
`power2round`	✅	4s	2s	+100%
`rej_eta_native`	✅	4s	4s	+0%
`rej_uniform_eta_native_aarch64`	✅	4s	4s	+0%
`shake128_finalize`	✅	4s	3s	+33%
`shake128_release`	✅	4s	2s	+100%
`shake128x4_absorb_once`	✅	4s	2s	+100%
`shake128x4_squeezeblocks`	✅	4s	2s	+100%
`shake256x4_absorb_once`	✅	4s	2s	+100%
`sig_unpack_hints`	✅	4s	3s	+33%
`sign_keypair`	✅	4s	3s	+33%
`sign_pk_from_sk`	✅	4s	7s	-43%
`sign_signature_extmu`	✅	4s	2s	+100%
`sign_signature_pre_hash_internal`	✅	4s	7s	-43%
`sign_verify_pre_hash_shake256`	✅	4s	4s	+0%
`use_hint`	✅	4s	2s	+100%
`yvec_get_poly`	✅	4s	4s	+0%
`yvec_init`	✅	4s	1s	+300%
`intt_native_aarch64`	✅	3s	2s	+50%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	3s	2s	+50%
`keccak_finalize`	✅	3s	2s	+50%
`keccakf1600_permute`	✅	3s	3s	+0%
`keccakf1600_permute_native`	✅	3s	3s	+0%
`mld_ct_cmask_nonzero_u8`	✅	3s	2s	+50%
`mld_ct_sel_int32`	✅	3s	1s	+200%
`mld_h`	✅	3s	3s	+0%
`mld_keccakf1600_extract_bytes`	✅	3s	2s	+50%
`mld_keccakf1600x4_xor_bytes_c`	✅	3s	3s	+0%
`mld_sample_s1_s2_serial`	✅	3s	9s	-67%
`mld_value_barrier_i64`	✅	3s	3s	+0%
`nttunpack_native_x86_64`	✅	3s	6s	-50%
`pack_sig_z`	✅	3s	2s	+50%
`pack_sk_s1`	✅	3s	3s	+0%
`pointwise_native_x86_64`	✅	3s	3s	+0%
`poly_caddq_native`	✅	3s	3s	+0%
`poly_caddq_native_aarch64`	✅	3s	2s	+50%
`poly_challenge`	✅	3s	6s	-50%
`poly_chknorm`	✅	3s	2s	+50%
`poly_chknorm_native_x86_64`	✅	3s	6s	-50%
`poly_decompose`	✅	3s	2s	+50%
`poly_decompose_c`	✅	3s	5s	-40%
`poly_decompose_native`	✅	3s	2s	+50%
`poly_invntt_tomont`	✅	3s	4s	-25%
`poly_invntt_tomont_native`	✅	3s	2s	+50%
`poly_ntt_c`	✅	3s	2s	+50%
`poly_ntt_native`	✅	3s	2s	+50%
`poly_permute_bitrev_to_custom_optional`	✅	3s	3s	+0%
`poly_sub`	✅	3s	3s	+0%
`poly_uniform`	✅	3s	3s	+0%
`poly_uniform_eta`	✅	3s	3s	+0%
`poly_use_hint`	✅	3s	3s	+0%
`poly_use_hint_native`	✅	3s	3s	+0%
`poly_use_hint_native_aarch64`	✅	3s	2s	+50%
`polyeta_pack`	✅	3s	3s	+0%
`polyvec_matrix_pointwise_montgomery_row`	✅	3s	2s	+50%
`polyveck_pack_eta`	✅	3s	1s	+200%
`polyveck_pack_w1`	✅	3s	3s	+0%
`polyvecl_pack_eta`	✅	3s	5s	-40%
`polyvecl_pointwise_acc_montgomery_native`	✅	3s	3s	+0%
`polyvecl_uniform_gamma1`	✅	3s	4s	-25%
`polyw1_pack`	✅	3s	3s	+0%
`polyz_pack`	✅	3s	4s	-25%
`rej_eta`	✅	3s	3s	+0%
`shake128_absorb`	✅	3s	3s	+0%
`shake128_init`	✅	3s	5s	-40%
`shake256_absorb`	✅	3s	2s	+50%
`shake256_finalize`	✅	3s	3s	+0%
`shake256_init`	✅	3s	1s	+200%
`shake256x4_squeezeblocks`	✅	3s	3s	+0%
`sign_signature`	✅	3s	3s	+0%
`unpack_pk_t1`	✅	3s	3s	+0%
`unpack_sk_s1hat`	✅	3s	3s	+0%
`caddq`	✅	2s	6s	-67%
`keccak_f1600_x1_native_aarch64`	✅	2s	1s	+100%
`keccak_f1600_x4_native_aarch64_v84a`	✅	2s	2s	+0%
`keccak_init`	✅	2s	3s	-33%
`keccak_squeeze`	✅	2s	1s	+100%
`keccakf1600_xor_bytes (big endian)`	✅	2s	3s	-33%
`keccakf1600x4_extract_bytes`	✅	2s	2s	+0%
`keccakf1600x4_extract_bytes_native`	✅	2s	7s	-71%
`mld_ct_abs_i32`	✅	2s	2s	+0%
`mld_ct_get_optblocker_i64`	✅	2s	1s	+100%
`mld_ct_get_optblocker_u32`	✅	2s	4s	-50%
`mld_ct_get_optblocker_u8`	✅	2s	1s	+100%
`mld_polymat_expand_entry`	✅	2s	2s	+0%
`ntt_native_aarch64`	✅	2s	5s	-60%
`pointwise_native_aarch64`	✅	2s	3s	-33%
`poly_chknorm_native`	✅	2s	4s	-50%
`poly_decompose_88_native_aarch64`	✅	2s	2s	+0%
`poly_uniform_gamma1`	✅	2s	2s	+0%
`poly_uniform_gamma1_4x`	✅	2s	3s	-33%
`poly_use_hint_c`	✅	2s	3s	-33%
`polyt1_pack`	✅	2s	2s	+0%
`polyt1_unpack`	✅	2s	4s	-50%
`polyveck_reduce`	✅	2s	3s	-33%
`polyvecl_pointwise_acc_montgomery`	✅	2s	2s	+0%
`polyvecl_uniform_gamma1_serial`	✅	2s	3s	-33%
`polyvecl_unpack_eta`	✅	2s	3s	-33%
`polyw1_pack_88`	✅	2s	3s	-33%
`rej_eta_c`	✅	2s	2s	+0%
`shake128_squeeze`	✅	2s	1s	+100%
`shake256_release`	✅	2s	2s	+0%
`sign_open`	✅	2s	3s	-33%
`sign_signature_pre_hash_shake256`	✅	2s	4s	-50%
`sys_check_capability`	✅	2s	2s	+0%
`unpack_sk_s2hat`	✅	2s	4s	-50%
`keccak_f1600_x1_native_aarch64_v84a`	✅	1s	2s	-50%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	1s	1s	+0%
`keccak_f1600_x4_native_avx2`	✅	1s	2s	-50%
`keccakf1600_extract_bytes (big endian)`	✅	1s	2s	-50%
`keccakf1600x4_xor_bytes`	✅	1s	3s	-67%
`make_hint`	✅	1s	3s	-67%
`mld_ct_cmask_nonzero_u32`	✅	1s	4s	-75%
`mld_keccakf1600x4_extract_bytes_c`	✅	1s	2s	-50%
`mld_value_barrier_u32`	✅	1s	2s	-50%
`mld_value_barrier_u8`	✅	1s	2s	-50%
`shake256_squeeze`	✅	1s	3s	-67%
`sk_t0hat_get_poly`	✅	1s	3s	-67%

oqs-bot · 2026-06-15T00:19:40Z

CBMC Results (ML-DSA-87)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	2318s	2406s	-3.7%
`polyvecl_pointwise_acc_montgomery_c`	✅	300s	319s	-6%
`mld_invntt_layer`	✅	281s	291s	-3%
`polyvec_matrix_expand`	✅	208s	223s	-7%
`rej_uniform_native`	✅	139s	153s	-9%
`mld_attempt_signature_generation`	✅	103s	105s	-2%
`poly_pointwise_montgomery_c`	✅	97s	94s	+3%
`mld_ct_memcmp`	✅	68s	64s	+6%
`sign_signature_internal`	✅	64s	68s	-6%
`sign_verify_internal`	✅	58s	61s	-5%
`polyvec_matrix_expand_serial`	✅	51s	48s	+6%
`mld_ntt_layer`	✅	42s	44s	-5%
`fqmul`	✅	39s	40s	-3%
`compute_pack_t0_t1`	✅	31s	30s	+3%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	29s	30s	-3%
`keccakf1600x4_permute_native`	✅	22s	25s	-12%
`mld_ntt_butterfly_block`	✅	22s	22s	+0%
`rej_uniform`	✅	22s	23s	-4%
`poly_chknorm_c`	✅	19s	17s	+12%
`polyt0_unpack`	✅	18s	19s	-5%
`polyeta_unpack`	✅	16s	14s	+14%
`mld_check_pct`	✅	13s	15s	-13%
`poly_add`	✅	12s	11s	+9%
`poly_uniform_4x`	✅	12s	12s	+0%
`rej_uniform_c`	✅	12s	13s	-8%
`poly_uniform_eta_4x`	✅	11s	12s	-8%
`polyveck_decompose`	✅	11s	8s	+38%
`polyveck_invntt_tomont`	✅	10s	9s	+11%
`keccak_absorb_once_x4`	✅	9s	10s	-10%
`pointwise_acc_native_aarch64`	✅	9s	7s	+29%
`polyvecl_ntt`	✅	9s	8s	+12%
`sign`	✅	9s	11s	-18%
`poly_decompose_c`	✅	8s	8s	+0%
`polyveck_caddq`	✅	8s	8s	+0%
`mld_compute_pack_z`	✅	7s	7s	+0%
`mld_keccakf1600_permute_c`	✅	7s	7s	+0%
`poly_invntt_tomont_c`	✅	7s	8s	-12%
`sign_pk_from_sk`	✅	7s	5s	+40%
`sign_signature_extmu`	✅	7s	4s	+75%
`pointwise_acc_native_x86_64`	✅	6s	8s	-25%
`poly_caddq_c`	✅	6s	7s	-14%
`poly_caddq_native_aarch64`	✅	6s	4s	+50%
`polyveck_ntt`	✅	6s	7s	-14%
`sign_open`	✅	6s	3s	+100%
`sk_s1hat_get_poly`	✅	6s	3s	+100%
`caddq`	✅	5s	3s	+67%
`decompose`	✅	5s	2s	+150%
`fqscale`	✅	5s	3s	+67%
`keccak_absorb`	✅	5s	6s	-17%
`mld_ct_cmask_nonzero_u8`	✅	5s	2s	+150%
`mld_prepare_domain_separation_prefix`	✅	5s	4s	+25%
`poly_chknorm_native_aarch64`	✅	5s	4s	+25%
`poly_permute_bitrev_to_custom_optional_native`	✅	5s	4s	+25%
`poly_shiftl`	✅	5s	3s	+67%
`poly_use_hint_native_aarch64`	✅	5s	2s	+150%
`polyveck_unpack_eta`	✅	5s	5s	+0%
`polyvecl_unpack_z`	✅	5s	4s	+25%
`polyz_unpack_c`	✅	5s	4s	+25%
`rej_eta_native`	✅	5s	5s	+0%
`sign_signature_pre_hash_internal`	✅	5s	5s	+0%
`sign_verify_extmu`	✅	5s	5s	+0%
`sign_verify_pre_hash_shake256`	✅	5s	4s	+25%
`unpack_sk_t0hat`	✅	5s	5s	+0%
`intt_native_aarch64`	✅	4s	3s	+33%
`keccak_f1600_x4_native_aarch64_v84a`	✅	4s	3s	+33%
`keccakf1600_extract_bytes (big endian)`	✅	4s	2s	+100%
`keccakf1600x4_permute`	✅	4s	3s	+33%
`keccakf1600x4_xor_bytes_native`	✅	4s	3s	+33%
`mld_ct_get_optblocker_u32`	✅	4s	1s	+300%
`mld_polymat_expand_entry`	✅	4s	2s	+100%
`mld_sample_s1_s2`	✅	4s	8s	-50%
`mld_sample_s1_s2_serial`	✅	4s	5s	-20%
`ntt_native_aarch64`	✅	4s	3s	+33%
`pack_sk_s1`	✅	4s	3s	+33%
`poly_caddq_native_x86_64`	✅	4s	2s	+100%
`poly_challenge`	✅	4s	5s	-20%
`poly_power2round`	✅	4s	4s	+0%
`poly_use_hint_native`	✅	4s	2s	+100%
`polyvec_matrix_pointwise_montgomery_row`	✅	4s	7s	-43%
`polyveck_chknorm`	✅	4s	5s	-20%
`polyveck_pack_eta`	✅	4s	2s	+100%
`polyveck_reduce`	✅	4s	4s	+0%
`polyvecl_chknorm`	✅	4s	4s	+0%
`polyvecl_unpack_eta`	✅	4s	3s	+33%
`rej_uniform_native_aarch64`	✅	4s	3s	+33%
`shake128x4_absorb_once`	✅	4s	1s	+300%
`shake256_absorb`	✅	4s	3s	+33%
`shake256_init`	✅	4s	3s	+33%
`sig_unpack_hints`	✅	4s	3s	+33%
`sign_keypair`	✅	4s	5s	-20%
`sign_signature_pre_hash_shake256`	✅	4s	7s	-43%
`unpack_sk`	✅	4s	4s	+0%
`yvec_get_poly`	✅	4s	4s	+0%
`yvec_init`	✅	4s	3s	+33%
`intt_native_x86_64`	✅	3s	4s	-25%
`keccak_f1600_x1_native_aarch64_v84a`	✅	3s	3s	+0%
`keccak_f1600_x4_native_avx2`	✅	3s	3s	+0%
`keccak_init`	✅	3s	2s	+50%
`keccak_squeezeblocks_x4`	✅	3s	5s	-40%
`keccakf1600_permute_native`	✅	3s	5s	-40%
`keccakf1600_xor_bytes`	✅	3s	3s	+0%
`keccakf1600_xor_bytes (big endian)`	✅	3s	3s	+0%
`keccakf1600x4_extract_bytes`	✅	3s	2s	+50%
`keccakf1600x4_extract_bytes_native`	✅	3s	2s	+50%
`keccakf1600x4_xor_bytes`	✅	3s	2s	+50%
`make_hint`	✅	3s	4s	-25%
`mld_ct_abs_i32`	✅	3s	3s	+0%
`mld_ct_get_optblocker_i64`	✅	3s	3s	+0%
`mld_h`	✅	3s	2s	+50%
`mld_keccakf1600x4_extract_bytes_c`	✅	3s	2s	+50%
`mld_keccakf1600x4_xor_bytes_c`	✅	3s	3s	+0%
`mld_value_barrier_u32`	✅	3s	3s	+0%
`montgomery_reduce`	✅	3s	2s	+50%
`ntt_native_x86_64`	✅	3s	3s	+0%
`pack_sig_h`	✅	3s	3s	+0%
`pack_sig_z`	✅	3s	2s	+50%
`pack_sk_rho_key_tr_s2`	✅	3s	1s	+200%
`pointwise_native_x86_64`	✅	3s	3s	+0%
`poly_chknorm_native`	✅	3s	4s	-25%
`poly_decompose`	✅	3s	3s	+0%
`poly_decompose_32_native_aarch64`	✅	3s	6s	-50%
`poly_decompose_native`	✅	3s	5s	-40%
`poly_ntt_c`	✅	3s	2s	+50%
`poly_ntt_native`	✅	3s	3s	+0%
`poly_permute_bitrev_to_custom_optional`	✅	3s	2s	+50%
`poly_pointwise_montgomery_native`	✅	3s	3s	+0%
`poly_sub`	✅	3s	4s	-25%
`poly_uniform`	✅	3s	4s	-25%
`poly_uniform_eta`	✅	3s	4s	-25%
`poly_uniform_gamma1`	✅	3s	4s	-25%
`polyveck_pack_w1`	✅	3s	3s	+0%
`polyvecl_pack_eta`	✅	3s	3s	+0%
`polyvecl_pointwise_acc_montgomery`	✅	3s	3s	+0%
`polyvecl_uniform_gamma1`	✅	3s	3s	+0%
`polyw1_pack`	✅	3s	1s	+200%
`polyw1_pack_88`	✅	3s	3s	+0%
`polyz_unpack`	✅	3s	3s	+0%
`polyz_unpack_17_native_aarch64`	✅	3s	1s	+200%
`polyz_unpack_19_native_aarch64`	✅	3s	3s	+0%
`rej_eta_c`	✅	3s	2s	+50%
`rej_uniform_eta_native_aarch64`	✅	3s	2s	+50%
`shake128_absorb`	✅	3s	3s	+0%
`shake128_finalize`	✅	3s	4s	-25%
`shake128_init`	✅	3s	3s	+0%
`shake128x4_squeezeblocks`	✅	3s	3s	+0%
`shake256_release`	✅	3s	2s	+50%
`shake256_squeeze`	✅	3s	4s	-25%
`sign_verify`	✅	3s	3s	+0%
`sign_verify_pre_hash_internal`	✅	3s	5s	-40%
`sk_s2hat_get_poly`	✅	3s	4s	-25%
`sk_t0hat_get_poly`	✅	3s	4s	-25%
`sys_check_capability`	✅	3s	3s	+0%
`unpack_pk_t1`	✅	3s	3s	+0%
`unpack_sk_s2hat`	✅	3s	3s	+0%
`use_hint`	✅	3s	2s	+50%
`keccak_f1600_x1_native_aarch64`	✅	2s	3s	-33%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	2s	1s	+100%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	2s	1s	+100%
`keccakf1600_permute`	✅	2s	2s	+0%
`mld_ct_cmask_neg_i32`	✅	2s	3s	-33%
`mld_ct_cmask_nonzero_u32`	✅	2s	4s	-50%
`mld_ct_sel_int32`	✅	2s	5s	-60%
`mld_keccakf1600_extract_bytes`	✅	2s	5s	-60%
`mld_value_barrier_u8`	✅	2s	2s	+0%
`nttunpack_native_x86_64`	✅	2s	5s	-60%
`pack_sig_c`	✅	2s	5s	-60%
`pointwise_native_aarch64`	✅	2s	4s	-50%
`poly_caddq`	✅	2s	2s	+0%
`poly_caddq_native`	✅	2s	5s	-60%
`poly_chknorm_native_x86_64`	✅	2s	3s	-33%
`poly_decompose_88_native_aarch64`	✅	2s	4s	-50%
`poly_invntt_tomont`	✅	2s	3s	-33%
`poly_invntt_tomont_native`	✅	2s	4s	-50%
`poly_ntt`	✅	2s	5s	-60%
`poly_uniform_gamma1_4x`	✅	2s	4s	-50%
`poly_use_hint`	✅	2s	2s	+0%
`polyt0_pack`	✅	2s	3s	-33%
`polyt1_pack`	✅	2s	3s	-33%
`polyt1_unpack`	✅	2s	2s	+0%
`polyvecl_pointwise_acc_montgomery_native`	✅	2s	4s	-50%
`polyvecl_uniform_gamma1_serial`	✅	2s	2s	+0%
`polyw1_pack_32`	✅	2s	3s	-33%
`polyz_pack`	✅	2s	3s	-33%
`polyz_unpack_native`	✅	2s	2s	+0%
`rej_eta`	✅	2s	3s	-33%
`shake128_squeeze`	✅	2s	3s	-33%
`shake256`	✅	2s	2s	+0%
`shake256_finalize`	✅	2s	2s	+0%
`shake256x4_absorb_once`	✅	2s	2s	+0%
`shake256x4_squeezeblocks`	✅	2s	3s	-33%
`sign_keypair_internal`	✅	2s	5s	-60%
`sign_signature`	✅	2s	4s	-50%
`unpack_sk_s1hat`	✅	2s	3s	-33%
`keccak_finalize`	✅	1s	1s	+0%
`keccak_squeeze`	✅	1s	5s	-80%
`mld_ct_get_optblocker_u8`	✅	1s	3s	-67%
`mld_value_barrier_i64`	✅	1s	4s	-75%
`poly_chknorm`	✅	1s	3s	-67%
`poly_pointwise_montgomery`	✅	1s	2s	-50%
`poly_reduce`	✅	1s	3s	-67%
`poly_use_hint_c`	✅	1s	3s	-67%
`polyeta_pack`	✅	1s	3s	-67%
`power2round`	✅	1s	2s	-50%
`reduce32`	✅	1s	3s	-67%
`shake128_release`	✅	1s	3s	-67%

oqs-bot · 2026-06-15T00:19:45Z

CBMC Results (ML-DSA-44)

Full Results (204 proofs)

Proof	Status	Current	Previous	Change
`TOTAL`	✅	1853s	1752s	+5.8%
`mld_invntt_layer`	✅	296s	284s	+4%
`rej_uniform_native`	✅	149s	143s	+4%
`polyvecl_pointwise_acc_montgomery_c`	✅	130s	123s	+6%
`poly_pointwise_montgomery_c`	✅	104s	88s	+18%
`mld_ct_memcmp`	✅	73s	67s	+9%
`mld_attempt_signature_generation`	✅	63s	62s	+2%
`mld_ntt_layer`	✅	46s	47s	-2%
`fqmul`	✅	44s	43s	+2%
`sign_verify_internal`	✅	29s	27s	+7%
`mld_ntt_butterfly_block`	✅	26s	22s	+18%
`polyvec_matrix_expand`	✅	26s	26s	+0%
`rej_uniform`	✅	24s	23s	+4%
`keccakf1600x4_permute_native`	✅	23s	24s	-4%
`sign_signature_internal`	✅	20s	18s	+11%
`poly_chknorm_c`	✅	19s	19s	+0%
`polyeta_unpack`	✅	18s	15s	+20%
`polyt0_unpack`	✅	18s	15s	+20%
`compute_pack_t0_t1`	✅	14s	15s	-7%
`rej_uniform_c`	✅	14s	12s	+17%
`mld_check_pct`	✅	13s	14s	-7%
`poly_uniform_4x`	✅	13s	10s	+30%
`poly_uniform_eta_4x`	✅	12s	12s	+0%
`keccak_absorb_once_x4`	✅	11s	14s	-21%
`poly_add`	✅	11s	10s	+10%
`poly_invntt_tomont_c`	✅	10s	7s	+43%
`polyvec_matrix_pointwise_montgomery_yvec`	✅	10s	10s	+0%
`polyveck_chknorm`	✅	10s	11s	-9%
`sign`	✅	10s	7s	+43%
`poly_caddq_c`	✅	9s	4s	+125%
`polyz_unpack_c`	✅	9s	10s	-10%
`mld_compute_pack_z`	✅	8s	11s	-27%
`sign_verify_pre_hash_shake256`	✅	8s	4s	+100%
`keccak_absorb`	✅	7s	6s	+17%
`mld_ct_cmask_nonzero_u8`	✅	7s	2s	+250%
`poly_power2round`	✅	7s	5s	+40%
`polyvec_matrix_expand_serial`	✅	7s	6s	+17%
`polyveck_decompose`	✅	7s	10s	-30%
`mld_keccakf1600_permute_c`	✅	6s	8s	-25%
`mld_sample_s1_s2_serial`	✅	6s	3s	+100%
`pack_sig_h`	✅	6s	1s	+500%
`pack_sk_rho_key_tr_s2`	✅	6s	2s	+200%
`poly_uniform_gamma1_4x`	✅	6s	4s	+50%
`polyt0_pack`	✅	6s	6s	+0%
`rej_eta_native`	✅	6s	4s	+50%
`sign_keypair`	✅	6s	4s	+50%
`sign_verify`	✅	6s	6s	+0%
`make_hint`	✅	5s	4s	+25%
`mld_ct_get_optblocker_u32`	✅	5s	2s	+150%
`mld_keccakf1600x4_xor_bytes_c`	✅	5s	4s	+25%
`pack_sk_s1`	✅	5s	2s	+150%
`pointwise_acc_native_x86_64`	✅	5s	4s	+25%
`poly_caddq_native_x86_64`	✅	5s	4s	+25%
`poly_challenge`	✅	5s	4s	+25%
`poly_ntt_c`	✅	5s	3s	+67%
`poly_permute_bitrev_to_custom_optional`	✅	5s	6s	-17%
`poly_shiftl`	✅	5s	2s	+150%
`poly_uniform`	✅	5s	7s	-29%
`poly_use_hint_c`	✅	5s	6s	-17%
`polyt1_unpack`	✅	5s	2s	+150%
`polyvecl_ntt`	✅	5s	5s	+0%
`rej_uniform_eta_native_aarch64`	✅	5s	2s	+150%
`shake256_release`	✅	5s	5s	+0%
`sign_open`	✅	5s	6s	-17%
`sign_pk_from_sk`	✅	5s	6s	-17%
`sign_signature_extmu`	✅	5s	3s	+67%
`sign_verify_pre_hash_internal`	✅	5s	6s	-17%
`sys_check_capability`	✅	5s	2s	+150%
`fqscale`	✅	4s	3s	+33%
`intt_native_x86_64`	✅	4s	4s	+0%
`keccak_f1600_x4_native_avx2`	✅	4s	2s	+100%
`keccak_squeeze`	✅	4s	2s	+100%
`keccak_squeezeblocks_x4`	✅	4s	5s	-20%
`keccakf1600_permute_native`	✅	4s	6s	-33%
`mld_h`	✅	4s	6s	-33%
`mld_polymat_expand_entry`	✅	4s	2s	+100%
`mld_prepare_domain_separation_prefix`	✅	4s	4s	+0%
`mld_value_barrier_u32`	✅	4s	3s	+33%
`ntt_native_aarch64`	✅	4s	4s	+0%
`pack_sig_c`	✅	4s	2s	+100%
`pointwise_acc_native_aarch64`	✅	4s	8s	-50%
`pointwise_native_x86_64`	✅	4s	2s	+100%
`poly_caddq_native_aarch64`	✅	4s	2s	+100%
`poly_chknorm_native`	✅	4s	2s	+100%
`poly_decompose_native`	✅	4s	2s	+100%
`poly_invntt_tomont`	✅	4s	3s	+33%
`poly_invntt_tomont_native`	✅	4s	5s	-20%
`poly_ntt`	✅	4s	4s	+0%
`poly_pointwise_montgomery_native`	✅	4s	1s	+300%
`poly_reduce`	✅	4s	4s	+0%
`poly_uniform_eta`	✅	4s	3s	+33%
`poly_use_hint_native`	✅	4s	2s	+100%
`poly_use_hint_native_aarch64`	✅	4s	4s	+0%
`polyt1_pack`	✅	4s	4s	+0%
`polyvec_matrix_pointwise_montgomery_row`	✅	4s	3s	+33%
`polyveck_invntt_tomont`	✅	4s	6s	-33%
`polyveck_ntt`	✅	4s	3s	+33%
`polyvecl_pointwise_acc_montgomery`	✅	4s	3s	+33%
`polyvecl_uniform_gamma1`	✅	4s	3s	+33%
`polyz_unpack_17_native_aarch64`	✅	4s	2s	+100%
`rej_eta_c`	✅	4s	5s	-20%
`shake128_absorb`	✅	4s	2s	+100%
`shake128_squeeze`	✅	4s	2s	+100%
`shake256_absorb`	✅	4s	2s	+100%
`shake256_init`	✅	4s	1s	+300%
`sign_signature`	✅	4s	4s	+0%
`sign_signature_pre_hash_shake256`	✅	4s	5s	-20%
`sign_verify_extmu`	✅	4s	4s	+0%
`sk_s1hat_get_poly`	✅	4s	1s	+300%
`unpack_sk`	✅	4s	3s	+33%
`unpack_sk_s1hat`	✅	4s	3s	+33%
`use_hint`	✅	4s	2s	+100%
`yvec_get_poly`	✅	4s	2s	+100%
`yvec_init`	✅	4s	3s	+33%
`decompose`	✅	3s	3s	+0%
`intt_native_aarch64`	✅	3s	3s	+0%
`keccak_f1600_x1_native_aarch64_v84a`	✅	3s	2s	+50%
`keccak_f1600_x4_native_aarch64_v8a_v84a_scalar_hybrid`	✅	3s	2s	+50%
`keccakf1600_permute`	✅	3s	1s	+200%
`keccakf1600_xor_bytes`	✅	3s	1s	+200%
`keccakf1600x4_extract_bytes`	✅	3s	4s	-25%
`keccakf1600x4_extract_bytes_native`	✅	3s	3s	+0%
`keccakf1600x4_xor_bytes`	✅	3s	2s	+50%
`mld_ct_get_optblocker_i64`	✅	3s	2s	+50%
`mld_ct_get_optblocker_u8`	✅	3s	2s	+50%
`mld_ct_sel_int32`	✅	3s	2s	+50%
`mld_keccakf1600_extract_bytes`	✅	3s	2s	+50%
`mld_keccakf1600x4_extract_bytes_c`	✅	3s	2s	+50%
`ntt_native_x86_64`	✅	3s	3s	+0%
`pack_sig_z`	✅	3s	3s	+0%
`poly_caddq_native`	✅	3s	4s	-25%
`poly_chknorm_native_x86_64`	✅	3s	3s	+0%
`poly_decompose_88_native_aarch64`	✅	3s	2s	+50%
`poly_decompose_c`	✅	3s	4s	-25%
`poly_uniform_gamma1`	✅	3s	4s	-25%
`polyeta_pack`	✅	3s	4s	-25%
`polyveck_caddq`	✅	3s	2s	+50%
`polyveck_pack_eta`	✅	3s	2s	+50%
`polyveck_reduce`	✅	3s	4s	-25%
`polyvecl_chknorm`	✅	3s	5s	-40%
`polyvecl_pointwise_acc_montgomery_native`	✅	3s	2s	+50%
`polyvecl_unpack_eta`	✅	3s	4s	-25%
`polyz_unpack_19_native_aarch64`	✅	3s	4s	-25%
`polyz_unpack_native`	✅	3s	4s	-25%
`reduce32`	✅	3s	1s	+200%
`rej_eta`	✅	3s	1s	+200%
`shake128_init`	✅	3s	3s	+0%
`shake128x4_absorb_once`	✅	3s	3s	+0%
`shake256`	✅	3s	1s	+200%
`shake256x4_squeezeblocks`	✅	3s	1s	+200%
`sig_unpack_hints`	✅	3s	3s	+0%
`unpack_pk_t1`	✅	3s	2s	+50%
`unpack_sk_t0hat`	✅	3s	4s	-25%
`keccak_f1600_x1_native_aarch64`	✅	2s	3s	-33%
`keccak_f1600_x4_native_aarch64_v84a`	✅	2s	3s	-33%
`keccak_f1600_x4_native_aarch64_v8a_scalar_hybrid`	✅	2s	2s	+0%
`keccak_init`	✅	2s	1s	+100%
`keccakf1600_extract_bytes (big endian)`	✅	2s	2s	+0%
`keccakf1600_xor_bytes (big endian)`	✅	2s	3s	-33%
`keccakf1600x4_permute`	✅	2s	3s	-33%
`mld_ct_abs_i32`	✅	2s	2s	+0%
`mld_ct_cmask_neg_i32`	✅	2s	4s	-50%
`mld_ct_cmask_nonzero_u32`	✅	2s	2s	+0%
`mld_sample_s1_s2`	✅	2s	3s	-33%
`mld_value_barrier_i64`	✅	2s	1s	+100%
`montgomery_reduce`	✅	2s	2s	+0%
`nttunpack_native_x86_64`	✅	2s	5s	-60%
`pointwise_native_aarch64`	✅	2s	3s	-33%
`poly_chknorm`	✅	2s	4s	-50%
`poly_chknorm_native_aarch64`	✅	2s	4s	-50%
`poly_decompose`	✅	2s	4s	-50%
`poly_ntt_native`	✅	2s	1s	+100%
`poly_permute_bitrev_to_custom_optional_native`	✅	2s	6s	-67%
`poly_pointwise_montgomery`	✅	2s	2s	+0%
`poly_use_hint`	✅	2s	3s	-33%
`polyveck_pack_w1`	✅	2s	4s	-50%
`polyveck_unpack_eta`	✅	2s	2s	+0%
`polyvecl_pack_eta`	✅	2s	3s	-33%
`polyw1_pack`	✅	2s	4s	-50%
`polyw1_pack_88`	✅	2s	4s	-50%
`polyz_pack`	✅	2s	4s	-50%
`polyz_unpack`	✅	2s	2s	+0%
`rej_uniform_native_aarch64`	✅	2s	2s	+0%
`shake128_finalize`	✅	2s	2s	+0%
`shake128_release`	✅	2s	3s	-33%
`shake256_finalize`	✅	2s	3s	-33%
`sign_keypair_internal`	✅	2s	4s	-50%
`sign_signature_pre_hash_internal`	✅	2s	4s	-50%
`sk_t0hat_get_poly`	✅	2s	4s	-50%
`caddq`	✅	1s	2s	-50%
`keccak_finalize`	✅	1s	5s	-80%
`keccakf1600x4_xor_bytes_native`	✅	1s	2s	-50%
`mld_value_barrier_u8`	✅	1s	4s	-75%
`poly_caddq`	✅	1s	3s	-67%
`poly_decompose_32_native_aarch64`	✅	1s	1s	+0%
`poly_sub`	✅	1s	2s	-50%
`polyvecl_uniform_gamma1_serial`	✅	1s	3s	-67%
`polyvecl_unpack_z`	✅	1s	3s	-67%
`polyw1_pack_32`	✅	1s	4s	-75%
`power2round`	✅	1s	2s	-50%
`shake128x4_squeezeblocks`	✅	1s	3s	-67%
`shake256_squeeze`	✅	1s	1s	+0%
`shake256x4_absorb_once`	✅	1s	4s	-75%
`sk_s2hat_get_poly`	✅	1s	3s	-67%
`unpack_sk_s2hat`	✅	1s	1s	+0%

mkannwischer added the benchmark label Jun 15, 2026

github-actions Bot reviewed Jun 15, 2026

View reviewed changes

oqs-bot reviewed Jun 15, 2026

View reviewed changes

github-actions Bot reviewed Jun 15, 2026

View reviewed changes

oqs-bot reviewed Jun 15, 2026

View reviewed changes

github-actions Bot reviewed Jun 15, 2026

View reviewed changes

Uh oh!

Conversation

mkannwischer commented Jun 14, 2026

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

Mac Mini (M1, 2020) benchmarks (opt)

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

Mac Mini (M1, 2020) benchmarks (no-opt)

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

SpacemiT K1 8 (Banana Pi F3) benchmarks (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 4th gen (c7i)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 4th gen (c7i) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 3rd gen (c6a)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 3rd gen (c6a) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 4th gen (c7a)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton4

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton2

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

AMD EPYC 4th gen (c7a) (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton4 (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 3rd gen (c6i)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton2 (no-opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton3

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Intel Xeon 3rd gen (c6i) (no-opt)

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

Arm Cortex-A76 (Raspberry Pi 5) benchmarks (opt)

Uh oh!

oqs-bot left a comment

Choose a reason for hiding this comment

Graviton3 (no-opt)

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

Arm Cortex-A55 (Snapdragon 888) benchmarks (opt)

Uh oh!