Add optimized NTT and invNTT in C #683

rod-chapman · 2025-01-22T13:38:14Z

Re-introduces fast NTT C code, following source re-organization in #674

See comments on older #610 . This commit addresses comments in that PR.

All tests, proofs, and lint OK.

rod-chapman · 2025-01-22T14:04:41Z

Code size metrics using gcc 14.2.0 on Apple M1
gcc -c -O3 -I. poly.c
size poly.o

yields

main: 4368 bytes
fastntt3: 8320 bytes

so increase of 3952 bytes

oqs-bot

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`28950` cycles	`28965` cycles	`1.00`
`ML-KEM-512 encaps`	`34259` cycles	`34252` cycles	`1.00`
`ML-KEM-512 decaps`	`44775` cycles	`44728` cycles	`1.00`
`ML-KEM-768 keypair`	`49201` cycles	`49307` cycles	`1.00`
`ML-KEM-768 encaps`	`54560` cycles	`54571` cycles	`1.00`
`ML-KEM-768 decaps`	`69387` cycles	`69426` cycles	`1.00`
`ML-KEM-1024 keypair`	`71916` cycles	`71915` cycles	`1.00`
`ML-KEM-1024 encaps`	`80489` cycles	`80618` cycles	`1.00`
`ML-KEM-1024 decaps`	`100492` cycles	`100358` cycles	`1.00`

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

oqs-bot

Intel Xeon 4th gen (c7i)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`9211` cycles	`9314` cycles	`0.99`
`ML-KEM-512 encaps`	`10728` cycles	`10833` cycles	`0.99`
`ML-KEM-512 decaps`	`14908` cycles	`14762` cycles	`1.01`
`ML-KEM-768 keypair`	`16054` cycles	`16125` cycles	`1.00`
`ML-KEM-768 encaps`	`17332` cycles	`17222` cycles	`1.01`
`ML-KEM-768 decaps`	`23077` cycles	`23277` cycles	`0.99`
`ML-KEM-1024 keypair`	`21364` cycles	`21371` cycles	`1.00`
`ML-KEM-1024 encaps`	`23332` cycles	`23306` cycles	`1.00`
`ML-KEM-1024 decaps`	`30953` cycles	`30896` cycles	`1.00`

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

oqs-bot

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

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`25837` cycles	`29371` cycles	`0.88`
`ML-KEM-512 encaps`	`29332` cycles	`34937` cycles	`0.84`
`ML-KEM-512 decaps`	`35699` cycles	`45404` cycles	`0.79`
`ML-KEM-768 keypair`	`43851` cycles	`47083` cycles	`0.93`
`ML-KEM-768 encaps`	`49019` cycles	`55467` cycles	`0.88`
`ML-KEM-768 decaps`	`58058` cycles	`67463` cycles	`0.86`
`ML-KEM-1024 keypair`	`66203` cycles	`71606` cycles	`0.92`
`ML-KEM-1024 encaps`	`73160` cycles	`81980` cycles	`0.89`
`ML-KEM-1024 decaps`	`85151` cycles	`99279` cycles	`0.86`

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

oqs-bot

AMD EPYC 3rd gen (c6a)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`16974` cycles	`16960` cycles	`1.00`
`ML-KEM-512 encaps`	`18642` cycles	`18692` cycles	`1.00`
`ML-KEM-512 decaps`	`24038` cycles	`24036` cycles	`1.00`
`ML-KEM-768 keypair`	`28692` cycles	`28708` cycles	`1.00`
`ML-KEM-768 encaps`	`29775` cycles	`29783` cycles	`1.00`
`ML-KEM-768 decaps`	`37563` cycles	`37567` cycles	`1.00`
`ML-KEM-1024 keypair`	`41423` cycles	`41708` cycles	`0.99`
`ML-KEM-1024 encaps`	`43669` cycles	`43976` cycles	`0.99`
`ML-KEM-1024 decaps`	`53971` cycles	`54287` cycles	`0.99`

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

oqs-bot

Intel Xeon 3rd gen (c6i)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`15924` cycles	`15930` cycles	`1.00`
`ML-KEM-512 encaps`	`17996` cycles	`18001` cycles	`1.00`
`ML-KEM-512 decaps`	`24553` cycles	`24514` cycles	`1.00`
`ML-KEM-768 keypair`	`27362` cycles	`27359` cycles	`1.00`
`ML-KEM-768 encaps`	`28912` cycles	`28918` cycles	`1.00`
`ML-KEM-768 decaps`	`38335` cycles	`38352` cycles	`1.00`
`ML-KEM-1024 keypair`	`36969` cycles	`36955` cycles	`1.00`
`ML-KEM-1024 encaps`	`39911` cycles	`39871` cycles	`1.00`
`ML-KEM-1024 decaps`	`52402` cycles	`52380` cycles	`1.00`

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

oqs-bot

AMD EPYC 4th gen (c7a)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`11286` cycles	`11282` cycles	`1.00`
`ML-KEM-512 encaps`	`12823` cycles	`12844` cycles	`1.00`
`ML-KEM-512 decaps`	`17665` cycles	`17668` cycles	`1.00`
`ML-KEM-768 keypair`	`19597` cycles	`19634` cycles	`1.00`
`ML-KEM-768 encaps`	`20581` cycles	`20625` cycles	`1.00`
`ML-KEM-768 decaps`	`27623` cycles	`27639` cycles	`1.00`
`ML-KEM-1024 keypair`	`26273` cycles	`26296` cycles	`1.00`
`ML-KEM-1024 encaps`	`28173` cycles	`28183` cycles	`1.00`
`ML-KEM-1024 decaps`	`37627` cycles	`37611` cycles	`1.00`

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

oqs-bot

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

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`32889` cycles	`39593` cycles	`0.83`
`ML-KEM-512 encaps`	`37424` cycles	`47505` cycles	`0.79`
`ML-KEM-512 decaps`	`45996` cycles	`61785` cycles	`0.74`
`ML-KEM-768 keypair`	`54907` cycles	`63996` cycles	`0.86`
`ML-KEM-768 encaps`	`61323` cycles	`75296` cycles	`0.81`
`ML-KEM-768 decaps`	`72918` cycles	`93787` cycles	`0.78`
`ML-KEM-1024 keypair`	`83219` cycles	`95491` cycles	`0.87`
`ML-KEM-1024 encaps`	`91543` cycles	`108990` cycles	`0.84`
`ML-KEM-1024 decaps`	`106377` cycles	`132483` cycles	`0.80`

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

oqs-bot

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

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`40288` cycles	`46437` cycles	`0.87`
`ML-KEM-512 encaps`	`45413` cycles	`54623` cycles	`0.83`
`ML-KEM-512 decaps`	`55654` cycles	`70298` cycles	`0.79`
`ML-KEM-768 keypair`	`67249` cycles	`76666` cycles	`0.88`
`ML-KEM-768 encaps`	`73541` cycles	`87243` cycles	`0.84`
`ML-KEM-768 decaps`	`86943` cycles	`107715` cycles	`0.81`
`ML-KEM-1024 keypair`	`98605` cycles	`112111` cycles	`0.88`
`ML-KEM-1024 encaps`	`107754` cycles	`126070` cycles	`0.85`
`ML-KEM-1024 decaps`	`125295` cycles	`152172` cycles	`0.82`

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

oqs-bot

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

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`27587` cycles	`36058` cycles	`0.77`
`ML-KEM-512 encaps`	`30933` cycles	`42302` cycles	`0.73`
`ML-KEM-512 decaps`	`37402` cycles	`55452` cycles	`0.67`
`ML-KEM-768 keypair`	`46118` cycles	`58497` cycles	`0.79`
`ML-KEM-768 encaps`	`51040` cycles	`66956` cycles	`0.76`
`ML-KEM-768 decaps`	`59688` cycles	`84104` cycles	`0.71`
`ML-KEM-1024 keypair`	`70178` cycles	`86469` cycles	`0.81`
`ML-KEM-1024 encaps`	`77299` cycles	`97202` cycles	`0.80`
`ML-KEM-1024 decaps`	`88097` cycles	`118782` cycles	`0.74`

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

oqs-bot

Graviton2

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`28965` cycles	`28970` cycles	`1.00`
`ML-KEM-512 encaps`	`34274` cycles	`34254` cycles	`1.00`
`ML-KEM-512 decaps`	`44811` cycles	`44742` cycles	`1.00`
`ML-KEM-768 keypair`	`49251` cycles	`49315` cycles	`1.00`
`ML-KEM-768 encaps`	`54566` cycles	`54582` cycles	`1.00`
`ML-KEM-768 decaps`	`69410` cycles	`69448` cycles	`1.00`
`ML-KEM-1024 keypair`	`71883` cycles	`71939` cycles	`1.00`
`ML-KEM-1024 encaps`	`80532` cycles	`80628` cycles	`1.00`
`ML-KEM-1024 decaps`	`100525` cycles	`100412` cycles	`1.00`

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

oqs-bot

Graviton3

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`18916` cycles	`18921` cycles	`1.00`
`ML-KEM-512 encaps`	`22412` cycles	`22423` cycles	`1.00`
`ML-KEM-512 decaps`	`29685` cycles	`29675` cycles	`1.00`
`ML-KEM-768 keypair`	`32293` cycles	`32313` cycles	`1.00`
`ML-KEM-768 encaps`	`35807` cycles	`35801` cycles	`1.00`
`ML-KEM-768 decaps`	`46236` cycles	`46189` cycles	`1.00`
`ML-KEM-1024 keypair`	`46634` cycles	`46631` cycles	`1.00`
`ML-KEM-1024 encaps`	`52306` cycles	`52345` cycles	`1.00`
`ML-KEM-1024 decaps`	`66377` cycles	`66378` cycles	`1.00`

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

oqs-bot

Graviton3 (no-opt)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`34586` cycles	`38694` cycles	`0.89`
`ML-KEM-512 encaps`	`37809` cycles	`44315` cycles	`0.85`
`ML-KEM-512 decaps`	`46108` cycles	`56145` cycles	`0.82`
`ML-KEM-768 keypair`	`57990` cycles	`63849` cycles	`0.91`
`ML-KEM-768 encaps`	`62149` cycles	`70986` cycles	`0.88`
`ML-KEM-768 decaps`	`73596` cycles	`86944` cycles	`0.85`
`ML-KEM-1024 keypair`	`87478` cycles	`95398` cycles	`0.92`
`ML-KEM-1024 encaps`	`93780` cycles	`105306` cycles	`0.89`
`ML-KEM-1024 decaps`	`108813` cycles	`125683` cycles	`0.87`

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

oqs-bot

Graviton2 (no-opt)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`52203` cycles	`58968` cycles	`0.89`
`ML-KEM-512 encaps`	`57092` cycles	`67423` cycles	`0.85`
`ML-KEM-512 decaps`	`68968` cycles	`85993` cycles	`0.80`
`ML-KEM-768 keypair`	`87155` cycles	`98315` cycles	`0.89`
`ML-KEM-768 encaps`	`93740` cycles	`109146` cycles	`0.86`
`ML-KEM-768 decaps`	`110565` cycles	`133916` cycles	`0.83`
`ML-KEM-1024 keypair`	`133004` cycles	`147104` cycles	`0.90`
`ML-KEM-1024 encaps`	`143042` cycles	`162179` cycles	`0.88`
`ML-KEM-1024 decaps`	`164927` cycles	`193746` cycles	`0.85`

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

oqs-bot

Graviton4

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`17732` cycles	`17730` cycles	`1.00`
`ML-KEM-512 encaps`	`20979` cycles	`20979` cycles	`1`
`ML-KEM-512 decaps`	`27652` cycles	`27658` cycles	`1.00`
`ML-KEM-768 keypair`	`30494` cycles	`30519` cycles	`1.00`
`ML-KEM-768 encaps`	`33416` cycles	`33426` cycles	`1.00`
`ML-KEM-768 decaps`	`42983` cycles	`42976` cycles	`1.00`
`ML-KEM-1024 keypair`	`44133` cycles	`44139` cycles	`1.00`
`ML-KEM-1024 encaps`	`49418` cycles	`49439` cycles	`1.00`
`ML-KEM-1024 decaps`	`62368` cycles	`62362` cycles	`1.00`

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

oqs-bot

Graviton4 (no-opt)

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`31879` cycles	`35503` cycles	`0.90`
`ML-KEM-512 encaps`	`34748` cycles	`40647` cycles	`0.85`
`ML-KEM-512 decaps`	`42589` cycles	`51644` cycles	`0.82`
`ML-KEM-768 keypair`	`53039` cycles	`58480` cycles	`0.91`
`ML-KEM-768 encaps`	`57058` cycles	`65246` cycles	`0.87`
`ML-KEM-768 decaps`	`68178` cycles	`80462` cycles	`0.85`
`ML-KEM-1024 keypair`	`80881` cycles	`88186` cycles	`0.92`
`ML-KEM-1024 encaps`	`86428` cycles	`96964` cycles	`0.89`
`ML-KEM-1024 decaps`	`101128` cycles	`116641` cycles	`0.87`

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

oqs-bot

Bananapi bpi-f3 benchmarks

Benchmark suite	Current: `9c2f6a3`	Previous: `143daca`	Ratio
`ML-KEM-512 keypair`	`255062` cycles	`309304` cycles	`0.82`
`ML-KEM-512 encaps`	`284238` cycles	`413143` cycles	`0.69`
`ML-KEM-512 decaps`	`365465` cycles	`555068` cycles	`0.66`
`ML-KEM-768 keypair`	`425530` cycles	`506933` cycles	`0.84`
`ML-KEM-768 encaps`	`460484` cycles	`637031` cycles	`0.72`
`ML-KEM-768 decaps`	`568427` cycles	`819119` cycles	`0.69`
`ML-KEM-1024 keypair`	`625346` cycles	`733246` cycles	`0.85`
`ML-KEM-1024 encaps`	`670884` cycles	`893738` cycles	`0.75`
`ML-KEM-1024 decaps`	`798212` cycles	`1109650` cycles	`0.72`

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

oqs-bot

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`51636` cycles	`51555` cycles	`1.00`
`ML-KEM-512 encaps`	`59276` cycles	`59153` cycles	`1.00`
`ML-KEM-512 decaps`	`75080` cycles	`75710` cycles	`0.99`
`ML-KEM-768 keypair`	`87651` cycles	`88925` cycles	`0.99`
`ML-KEM-768 encaps`	`95164` cycles	`95966` cycles	`0.99`
`ML-KEM-768 decaps`	`119313` cycles	`120002` cycles	`0.99`
`ML-KEM-1024 keypair`	`132944` cycles	`132553` cycles	`1.00`
`ML-KEM-1024 encaps`	`144629` cycles	`144793` cycles	`1.00`
`ML-KEM-1024 decaps`	`176707` cycles	`176528` cycles	`1.00`

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

oqs-bot

Arm Cortex-A55 (Snapdragon 888) benchmarks

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`58116` cycles	`58108` cycles	`1.00`
`ML-KEM-512 encaps`	`64945` cycles	`64926` cycles	`1.00`
`ML-KEM-512 decaps`	`83854` cycles	`83675` cycles	`1.00`
`ML-KEM-768 keypair`	`98918` cycles	`98806` cycles	`1.00`
`ML-KEM-768 encaps`	`109470` cycles	`109470` cycles	`1`
`ML-KEM-768 decaps`	`136509` cycles	`136273` cycles	`1.00`
`ML-KEM-1024 keypair`	`149678` cycles	`149673` cycles	`1.00`
`ML-KEM-1024 encaps`	`166087` cycles	`165917` cycles	`1.00`
`ML-KEM-1024 decaps`	`201802` cycles	`202149` cycles	`1.00`

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

rod-chapman · 2025-01-31T14:46:35Z

The benchmarks seem to be a bit better again following adoption of GCC 14...

…g source re-organization in PR#674. See comments on older PR#610. All tests, proofs, and lint OK. Signed-off-by: Rod Chapman <rodchap@amazon.com> Correct list of called functions for this proof. Signed-off-by: Rod Chapman <rodchap@amazon.com> Update autogenerated files aftre rebase Signed-off-by: Rod Chapman <rodchap@amazon.com> Re-generate autogenerated files following rebase Signed-off-by: Rod Chapman <rodchap@amazon.com> Update auto-generated files and copyright messages following rebase Signed-off-by: Rod Chapman <rodchap@amazon.com> Update copyright notices for these news files Signed-off-by: Rod Chapman <rodchap@amazon.com> Update one more copyright notice Signed-off-by: Rod Chapman <rodchap@amazon.com> Update MLKEM_NAMESPACE to MLK_NAMESPACE for all new proofs Signed-off-by: Rod Chapman <rodchap@amazon.com> Rename INLINE to MLK_INLINE for new functions here. Signed-off-by: Rod Chapman <rodchap@amazon.com> rename NTT_BOUNDx macros to MLK_NTT_BOUNDx Signed-off-by: Rod Chapman <rodchap@amazon.com>

oqs-bot

SpacemiT K1 8 (Banana Pi F3) benchmarks

Benchmark suite	Current: `c8c3692`	Previous: `94c8b47`	Ratio
`ML-KEM-512 keypair`	`195915` cycles	`225118` cycles	`0.87`
`ML-KEM-512 encaps`	`210980` cycles	`269679` cycles	`0.78`
`ML-KEM-512 decaps`	`255226` cycles	`343289` cycles	`0.74`
`ML-KEM-768 keypair`	`327524` cycles	`371250` cycles	`0.88`
`ML-KEM-768 encaps`	`349417` cycles	`430008` cycles	`0.81`
`ML-KEM-768 decaps`	`410007` cycles	`527834` cycles	`0.78`
`ML-KEM-1024 keypair`	`496262` cycles	`555712` cycles	`0.89`
`ML-KEM-1024 encaps`	`527973` cycles	`631753` cycles	`0.84`
`ML-KEM-1024 decaps`	`605394` cycles	`752320` cycles	`0.80`

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

rod-chapman requested a review from a team January 22, 2025 13:38

rod-chapman mentioned this pull request Jan 22, 2025

Add fast, layer-merged forward and inverse NTT code. #610

Closed

rod-chapman force-pushed the fastntt3 branch from a7755d5 to e96eefe Compare January 22, 2025 14:12

rod-chapman added the benchmark this PR should be benchmarked in CI label Jan 22, 2025

oqs-bot reviewed Jan 22, 2025

View reviewed changes

hanno-becker changed the title ~~Re-baseline and introduce new fast NTT and inverse NTT~~ Add optimized NTT and invNTT in C Jan 23, 2025

rod-chapman force-pushed the fastntt3 branch 6 times, most recently from e60839c to eac7e91 Compare January 28, 2025 13:54

rod-chapman force-pushed the fastntt3 branch 2 times, most recently from a5b1265 to 0bc46b1 Compare January 31, 2025 14:01

rod-chapman force-pushed the fastntt3 branch from 0bc46b1 to 81877e5 Compare January 31, 2025 17:33

hanno-becker force-pushed the fastntt3 branch from 81877e5 to af0f31d Compare January 31, 2025 17:53

hanno-becker added benchmark this PR should be benchmarked in CI and removed benchmark this PR should be benchmarked in CI labels Jan 31, 2025

hanno-becker force-pushed the fastntt3 branch from af0f31d to 9c2f6a3 Compare February 1, 2025 06:52

hanno-becker added benchmark this PR should be benchmarked in CI and removed benchmark this PR should be benchmarked in CI labels Feb 1, 2025

rod-chapman force-pushed the fastntt3 branch 4 times, most recently from bac386a to 8583efb Compare February 4, 2025 09:40

hanno-becker force-pushed the fastntt3 branch from 8583efb to 676a80c Compare February 5, 2025 05:04

rod-chapman force-pushed the fastntt3 branch from 676a80c to ca2bff1 Compare February 5, 2025 09:09

hanno-becker added benchmark this PR should be benchmarked in CI and removed benchmark this PR should be benchmarked in CI labels Feb 5, 2025

rod-chapman force-pushed the fastntt3 branch 6 times, most recently from 4429485 to f25ae98 Compare February 6, 2025 11:55

rod-chapman force-pushed the fastntt3 branch from f25ae98 to c8c3692 Compare February 6, 2025 14:07

rod-chapman added benchmark this PR should be benchmarked in CI and removed benchmark this PR should be benchmarked in CI labels Feb 6, 2025

oqs-bot reviewed Feb 6, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add optimized NTT and invNTT in C #683

Add optimized NTT and invNTT in C #683

rod-chapman commented Jan 22, 2025 •

edited

Loading

rod-chapman commented Jan 22, 2025 •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

rod-chapman commented Jan 31, 2025

oqs-bot left a comment

Add optimized NTT and invNTT in C #683

Are you sure you want to change the base?

Add optimized NTT and invNTT in C #683

Conversation

rod-chapman commented Jan 22, 2025 • edited Loading

rod-chapman commented Jan 22, 2025 • edited Loading

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Arm Cortex-A76 (Raspberry Pi 5) benchmarks

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Intel Xeon 4th gen (c7i)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

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

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

AMD EPYC 3rd gen (c6a)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Intel Xeon 3rd gen (c6i)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

AMD EPYC 4th gen (c7a)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

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

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

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

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

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

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton2

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton3

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton3 (no-opt)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton2 (no-opt)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton4

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Graviton4 (no-opt)

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Bananapi bpi-f3 benchmarks

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Arm Cortex-A72 (Raspberry Pi 4) benchmarks

oqs-bot left a comment • edited Loading

Choose a reason for hiding this comment

Arm Cortex-A55 (Snapdragon 888) benchmarks

rod-chapman commented Jan 31, 2025

oqs-bot left a comment

Choose a reason for hiding this comment

SpacemiT K1 8 (Banana Pi F3) benchmarks

rod-chapman commented Jan 22, 2025 •

edited

Loading

rod-chapman commented Jan 22, 2025 •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading

oqs-bot left a comment •

edited

Loading