Repository navigation
sign: Order rejection checks by rejection rate - #1431
mkannwischer wants to merge 1 commit into
Conversation
There was a problem hiding this comment.
Mac Mini (M1, 2020) benchmarks (opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
46483 cycles |
46495 cycles |
1.00 |
ML-DSA-44 sign |
126277 cycles |
131379 cycles |
0.96 |
ML-DSA-44 verify |
47328 cycles |
47329 cycles |
1.00 |
ML-DSA-65 keypair |
81684 cycles |
81688 cycles |
1.00 |
ML-DSA-65 sign |
203626 cycles |
215251 cycles |
0.95 |
ML-DSA-65 verify |
79296 cycles |
79300 cycles |
1.00 |
ML-DSA-87 keypair |
132436 cycles |
132342 cycles |
1.00 |
ML-DSA-87 sign |
264823 cycles |
277355 cycles |
0.95 |
ML-DSA-87 verify |
137040 cycles |
137009 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A55 (Snapdragon 888) benchmarks (opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
267076 cycles |
271362 cycles |
0.98 |
ML-DSA-44 sign |
768789 cycles |
821847 cycles |
0.94 |
ML-DSA-44 verify |
268922 cycles |
275728 cycles |
0.98 |
ML-DSA-65 keypair |
462364 cycles |
465556 cycles |
0.99 |
ML-DSA-65 sign |
1234780 cycles |
1335846 cycles |
0.92 |
ML-DSA-65 verify |
447107 cycles |
454969 cycles |
0.98 |
ML-DSA-87 keypair |
788186 cycles |
797505 cycles |
0.99 |
ML-DSA-87 sign |
1693254 cycles |
1842068 cycles |
0.92 |
ML-DSA-87 verify |
769006 cycles |
772926 cycles |
0.99 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Mac Mini (M1, 2020) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
113344 cycles |
112827 cycles |
1.00 |
ML-DSA-44 sign |
389337 cycles |
402496 cycles |
0.97 |
ML-DSA-44 verify |
120087 cycles |
119365 cycles |
1.01 |
ML-DSA-65 keypair |
192979 cycles |
193024 cycles |
1.00 |
ML-DSA-65 sign |
607732 cycles |
651437 cycles |
0.93 |
ML-DSA-65 verify |
193079 cycles |
192987 cycles |
1.00 |
ML-DSA-87 keypair |
318985 cycles |
318856 cycles |
1.00 |
ML-DSA-87 sign |
786008 cycles |
831314 cycles |
0.95 |
ML-DSA-87 verify |
325787 cycles |
325766 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Intel Xeon 4th gen (c7i)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
43504 cycles |
43490 cycles |
1.00 |
ML-DSA-44 sign |
125230 cycles |
130374 cycles |
0.96 |
ML-DSA-44 verify |
45467 cycles |
45520 cycles |
1.00 |
ML-DSA-65 keypair |
75839 cycles |
76087 cycles |
1.00 |
ML-DSA-65 sign |
199703 cycles |
213837 cycles |
0.93 |
ML-DSA-65 verify |
74685 cycles |
74906 cycles |
1.00 |
ML-DSA-87 keypair |
123539 cycles |
123169 cycles |
1.00 |
ML-DSA-87 sign |
257464 cycles |
271033 cycles |
0.95 |
ML-DSA-87 verify |
120914 cycles |
120541 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Intel Xeon 4th gen (c7i) (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
92027 cycles |
91856 cycles |
1.00 |
ML-DSA-44 sign |
335496 cycles |
347728 cycles |
0.96 |
ML-DSA-44 verify |
99573 cycles |
99470 cycles |
1.00 |
ML-DSA-65 keypair |
154185 cycles |
154288 cycles |
1.00 |
ML-DSA-65 sign |
528846 cycles |
565330 cycles |
0.94 |
ML-DSA-65 verify |
158879 cycles |
159320 cycles |
1.00 |
ML-DSA-87 keypair |
255300 cycles |
255280 cycles |
1.00 |
ML-DSA-87 sign |
683414 cycles |
724481 cycles |
0.94 |
ML-DSA-87 verify |
265647 cycles |
263908 cycles |
1.01 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
AMD EPYC 4th gen (c7a)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
46966 cycles |
47122 cycles |
1.00 |
ML-DSA-44 sign |
135193 cycles |
139708 cycles |
0.97 |
ML-DSA-44 verify |
49509 cycles |
49493 cycles |
1.00 |
ML-DSA-65 keypair |
82102 cycles |
82609 cycles |
0.99 |
ML-DSA-65 sign |
215204 cycles |
228035 cycles |
0.94 |
ML-DSA-65 verify |
82418 cycles |
82590 cycles |
1.00 |
ML-DSA-87 keypair |
130554 cycles |
135639 cycles |
0.96 |
ML-DSA-87 sign |
265794 cycles |
297917 cycles |
0.89 |
ML-DSA-87 verify |
128188 cycles |
138079 cycles |
0.93 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
AMD EPYC 3rd gen (c6a)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
52000 cycles |
51939 cycles |
1.00 |
ML-DSA-44 sign |
149041 cycles |
156000 cycles |
0.96 |
ML-DSA-44 verify |
54150 cycles |
54123 cycles |
1.00 |
ML-DSA-65 keypair |
90040 cycles |
89968 cycles |
1.00 |
ML-DSA-65 sign |
238806 cycles |
253225 cycles |
0.94 |
ML-DSA-65 verify |
90688 cycles |
89944 cycles |
1.01 |
ML-DSA-87 keypair |
144075 cycles |
144818 cycles |
0.99 |
ML-DSA-87 sign |
291390 cycles |
311705 cycles |
0.93 |
ML-DSA-87 verify |
138819 cycles |
139540 cycles |
0.99 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
AMD EPYC 4th gen (c7a) (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
130153 cycles |
118984 cycles |
1.09 |
ML-DSA-44 sign |
465144 cycles |
457223 cycles |
1.02 |
ML-DSA-44 verify |
136543 cycles |
131208 cycles |
1.04 |
ML-DSA-65 keypair |
201668 cycles |
201145 cycles |
1.00 |
ML-DSA-65 sign |
690959 cycles |
742379 cycles |
0.93 |
ML-DSA-65 verify |
210334 cycles |
210357 cycles |
1.00 |
ML-DSA-87 keypair |
330759 cycles |
331120 cycles |
1.00 |
ML-DSA-87 sign |
885955 cycles |
937547 cycles |
0.94 |
ML-DSA-87 verify |
343797 cycles |
344646 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
⚠️ Performance Alert ⚠️
Possible performance regression was detected for benchmark 'AMD EPYC 4th gen (c7a) (no-opt)'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.03.
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
130153 cycles |
118984 cycles |
1.09 |
ML-DSA-44 verify |
136543 cycles |
131208 cycles |
1.04 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
AMD EPYC 3rd gen (c6a) (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
136700 cycles |
133390 cycles |
1.02 |
ML-DSA-44 sign |
509354 cycles |
520419 cycles |
0.98 |
ML-DSA-44 verify |
149320 cycles |
147163 cycles |
1.01 |
ML-DSA-65 keypair |
226418 cycles |
224254 cycles |
1.01 |
ML-DSA-65 sign |
791338 cycles |
847424 cycles |
0.93 |
ML-DSA-65 verify |
236618 cycles |
234487 cycles |
1.01 |
ML-DSA-87 keypair |
368243 cycles |
374067 cycles |
0.98 |
ML-DSA-87 sign |
997011 cycles |
1077560 cycles |
0.93 |
ML-DSA-87 verify |
381585 cycles |
387046 cycles |
0.99 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Intel Xeon 3rd gen (c6i)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
61814 cycles |
61880 cycles |
1.00 |
ML-DSA-44 sign |
180420 cycles |
188999 cycles |
0.95 |
ML-DSA-44 verify |
66341 cycles |
66270 cycles |
1.00 |
ML-DSA-65 keypair |
111270 cycles |
108610 cycles |
1.02 |
ML-DSA-65 sign |
293308 cycles |
311200 cycles |
0.94 |
ML-DSA-65 verify |
109878 cycles |
108850 cycles |
1.01 |
ML-DSA-87 keypair |
172071 cycles |
170415 cycles |
1.01 |
ML-DSA-87 sign |
360965 cycles |
379059 cycles |
0.95 |
ML-DSA-87 verify |
173187 cycles |
169838 cycles |
1.02 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Intel Xeon 3rd gen (c6i) (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
153928 cycles |
154089 cycles |
1.00 |
ML-DSA-44 sign |
565915 cycles |
588454 cycles |
0.96 |
ML-DSA-44 verify |
168942 cycles |
168985 cycles |
1.00 |
ML-DSA-65 keypair |
261111 cycles |
262084 cycles |
1.00 |
ML-DSA-65 sign |
892477 cycles |
957523 cycles |
0.93 |
ML-DSA-65 verify |
271433 cycles |
271777 cycles |
1.00 |
ML-DSA-87 keypair |
431696 cycles |
432435 cycles |
1.00 |
ML-DSA-87 sign |
1144218 cycles |
1211727 cycles |
0.94 |
ML-DSA-87 verify |
446747 cycles |
446648 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton5
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
51436 cycles |
51545 cycles |
1.00 |
ML-DSA-44 sign |
155076 cycles |
162250 cycles |
0.96 |
ML-DSA-44 verify |
54819 cycles |
54707 cycles |
1.00 |
ML-DSA-65 keypair |
90196 cycles |
90155 cycles |
1.00 |
ML-DSA-65 sign |
248768 cycles |
266210 cycles |
0.93 |
ML-DSA-65 verify |
89976 cycles |
89817 cycles |
1.00 |
ML-DSA-87 keypair |
145946 cycles |
145919 cycles |
1.00 |
ML-DSA-87 sign |
316757 cycles |
335543 cycles |
0.94 |
ML-DSA-87 verify |
144938 cycles |
145468 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A55 (Snapdragon 888) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
462444 cycles |
465447 cycles |
0.99 |
ML-DSA-44 sign |
2045365 cycles |
2139526 cycles |
0.96 |
ML-DSA-44 verify |
555007 cycles |
556637 cycles |
1.00 |
ML-DSA-65 keypair |
781227 cycles |
784251 cycles |
1.00 |
ML-DSA-65 sign |
3197624 cycles |
3492484 cycles |
0.92 |
ML-DSA-65 verify |
864436 cycles |
867152 cycles |
1.00 |
ML-DSA-87 keypair |
1261446 cycles |
1264532 cycles |
1.00 |
ML-DSA-87 sign |
4010388 cycles |
4319116 cycles |
0.93 |
ML-DSA-87 verify |
1386616 cycles |
1392230 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A76 (Raspberry Pi 5) benchmarks (opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
112180 cycles |
112287 cycles |
1.00 |
ML-DSA-44 sign |
340501 cycles |
354101 cycles |
0.96 |
ML-DSA-44 verify |
117258 cycles |
117167 cycles |
1.00 |
ML-DSA-65 keypair |
194399 cycles |
194666 cycles |
1.00 |
ML-DSA-65 sign |
544990 cycles |
583246 cycles |
0.93 |
ML-DSA-65 verify |
193236 cycles |
193114 cycles |
1.00 |
ML-DSA-87 keypair |
320046 cycles |
321100 cycles |
1.00 |
ML-DSA-87 sign |
707638 cycles |
747925 cycles |
0.95 |
ML-DSA-87 verify |
317779 cycles |
318764 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton5 (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
100293 cycles |
99958 cycles |
1.00 |
ML-DSA-44 sign |
355113 cycles |
369441 cycles |
0.96 |
ML-DSA-44 verify |
110582 cycles |
110183 cycles |
1.00 |
ML-DSA-65 keypair |
175331 cycles |
176088 cycles |
1.00 |
ML-DSA-65 sign |
555087 cycles |
596658 cycles |
0.93 |
ML-DSA-65 verify |
177767 cycles |
177613 cycles |
1.00 |
ML-DSA-87 keypair |
285187 cycles |
286628 cycles |
0.99 |
ML-DSA-87 sign |
715875 cycles |
758172 cycles |
0.94 |
ML-DSA-87 verify |
296824 cycles |
295695 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton4
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
67422 cycles |
67200 cycles |
1.00 |
ML-DSA-44 sign |
190932 cycles |
198799 cycles |
0.96 |
ML-DSA-44 verify |
70378 cycles |
70440 cycles |
1.00 |
ML-DSA-65 keypair |
119289 cycles |
119573 cycles |
1.00 |
ML-DSA-65 sign |
305593 cycles |
325662 cycles |
0.94 |
ML-DSA-65 verify |
116785 cycles |
117090 cycles |
1.00 |
ML-DSA-87 keypair |
196705 cycles |
196352 cycles |
1.00 |
ML-DSA-87 sign |
401801 cycles |
420686 cycles |
0.96 |
ML-DSA-87 verify |
193245 cycles |
192912 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A76 (Raspberry Pi 5) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
211730 cycles |
212139 cycles |
1.00 |
ML-DSA-44 sign |
734098 cycles |
760267 cycles |
0.97 |
ML-DSA-44 verify |
229532 cycles |
229531 cycles |
1.00 |
ML-DSA-65 keypair |
375430 cycles |
376003 cycles |
1.00 |
ML-DSA-65 sign |
1162783 cycles |
1245016 cycles |
0.93 |
ML-DSA-65 verify |
371803 cycles |
371755 cycles |
1.00 |
ML-DSA-87 keypair |
601690 cycles |
601186 cycles |
1.00 |
ML-DSA-87 sign |
1500896 cycles |
1583247 cycles |
0.95 |
ML-DSA-87 verify |
617436 cycles |
616849 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton4 (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
127826 cycles |
127783 cycles |
1.00 |
ML-DSA-44 sign |
425885 cycles |
442225 cycles |
0.96 |
ML-DSA-44 verify |
136437 cycles |
136517 cycles |
1.00 |
ML-DSA-65 keypair |
223200 cycles |
223479 cycles |
1.00 |
ML-DSA-65 sign |
667955 cycles |
714516 cycles |
0.93 |
ML-DSA-65 verify |
220944 cycles |
220881 cycles |
1.00 |
ML-DSA-87 keypair |
365026 cycles |
365589 cycles |
1.00 |
ML-DSA-87 sign |
910819 cycles |
958728 cycles |
0.95 |
ML-DSA-87 verify |
371422 cycles |
371278 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton3
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
71587 cycles |
71487 cycles |
1.00 |
ML-DSA-44 sign |
202042 cycles |
210224 cycles |
0.96 |
ML-DSA-44 verify |
74958 cycles |
74917 cycles |
1.00 |
ML-DSA-65 keypair |
126032 cycles |
126188 cycles |
1.00 |
ML-DSA-65 sign |
324680 cycles |
344959 cycles |
0.94 |
ML-DSA-65 verify |
124159 cycles |
124325 cycles |
1.00 |
ML-DSA-87 keypair |
206662 cycles |
206913 cycles |
1.00 |
ML-DSA-87 sign |
423495 cycles |
444505 cycles |
0.95 |
ML-DSA-87 verify |
204723 cycles |
204416 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton3 (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
138343 cycles |
138325 cycles |
1.00 |
ML-DSA-44 sign |
470222 cycles |
488151 cycles |
0.96 |
ML-DSA-44 verify |
149475 cycles |
149578 cycles |
1.00 |
ML-DSA-65 keypair |
242575 cycles |
242073 cycles |
1.00 |
ML-DSA-65 sign |
739545 cycles |
792009 cycles |
0.93 |
ML-DSA-65 verify |
242027 cycles |
241892 cycles |
1.00 |
ML-DSA-87 keypair |
395811 cycles |
396711 cycles |
1.00 |
ML-DSA-87 sign |
1007776 cycles |
1059889 cycles |
0.95 |
ML-DSA-87 verify |
404966 cycles |
404586 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton2
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
112184 cycles |
112244 cycles |
1.00 |
ML-DSA-44 sign |
340803 cycles |
354631 cycles |
0.96 |
ML-DSA-44 verify |
117435 cycles |
117405 cycles |
1.00 |
ML-DSA-65 keypair |
194508 cycles |
194595 cycles |
1.00 |
ML-DSA-65 sign |
545448 cycles |
583758 cycles |
0.93 |
ML-DSA-65 verify |
193594 cycles |
193536 cycles |
1.00 |
ML-DSA-87 keypair |
320488 cycles |
320476 cycles |
1.00 |
ML-DSA-87 sign |
708659 cycles |
746827 cycles |
0.95 |
ML-DSA-87 verify |
318427 cycles |
317877 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Graviton2 (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
211908 cycles |
211929 cycles |
1.00 |
ML-DSA-44 sign |
735842 cycles |
760854 cycles |
0.97 |
ML-DSA-44 verify |
229369 cycles |
229874 cycles |
1.00 |
ML-DSA-65 keypair |
375079 cycles |
376701 cycles |
1.00 |
ML-DSA-65 sign |
1163391 cycles |
1247378 cycles |
0.93 |
ML-DSA-65 verify |
371247 cycles |
372579 cycles |
1.00 |
ML-DSA-87 keypair |
601033 cycles |
602349 cycles |
1.00 |
ML-DSA-87 sign |
1541924 cycles |
1587832 cycles |
0.97 |
ML-DSA-87 verify |
616742 cycles |
618624 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
CBMC Results (ML-DSA-87, REDUCE-RAM)
Full Results (433 configurations)
|
CBMC Results (ML-DSA-65, REDUCE-RAM)
Full Results (433 configurations)
|
CBMC Results (ML-DSA-44)
Full Results (433 configurations)
|
CBMC Results (ML-DSA-65)
Full Results (433 configurations)
|
CBMC Results (ML-DSA-87)
Full Results (433 configurations)
|
CBMC Results (ML-DSA-44, REDUCE-RAM)
Full Results (433 configurations)
|
There was a problem hiding this comment.
Arm Cortex-M55 (NUCLEO-N657X0-Q) benchmarks (opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
2129870 cycles |
2129863 cycles |
1.00 |
ML-DSA-44 sign |
16453128 cycles |
17791740 cycles |
0.92 |
ML-DSA-44 verify |
2398876 cycles |
2398890 cycles |
1.00 |
ML-DSA-65 keypair |
3731142 cycles |
3731150 cycles |
1.00 |
ML-DSA-65 sign |
13926258 cycles |
15034565 cycles |
0.93 |
ML-DSA-65 verify |
4012792 cycles |
4012806 cycles |
1.00 |
ML-DSA-87 keypair |
6483505 cycles |
6483258 cycles |
1.00 |
ML-DSA-87 sign |
23689890 cycles |
25052782 cycles |
0.95 |
ML-DSA-87 verify |
6865468 cycles |
6865084 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-M55 (NUCLEO-N657X0-Q) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
2328440 cycles |
2328425 cycles |
1.00 |
ML-DSA-44 sign |
16453121 cycles |
17791719 cycles |
0.92 |
ML-DSA-44 verify |
2398868 cycles |
2398854 cycles |
1.00 |
ML-DSA-65 keypair |
4025151 cycles |
4025137 cycles |
1.00 |
ML-DSA-65 sign |
13926279 cycles |
15034565 cycles |
0.93 |
ML-DSA-65 verify |
4012806 cycles |
4012799 cycles |
1.00 |
ML-DSA-87 keypair |
6864901 cycles |
6864894 cycles |
1.00 |
ML-DSA-87 sign |
23684886 cycles |
25049136 cycles |
0.95 |
ML-DSA-87 verify |
6864419 cycles |
6864405 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A72 (Raspberry Pi 4) benchmarks (opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
216636 cycles |
219078 cycles |
0.99 |
ML-DSA-44 sign |
575604 cycles |
603963 cycles |
0.95 |
ML-DSA-44 verify |
217827 cycles |
217691 cycles |
1.00 |
ML-DSA-65 keypair |
383224 cycles |
384101 cycles |
1.00 |
ML-DSA-65 sign |
943082 cycles |
994891 cycles |
0.95 |
ML-DSA-65 verify |
368499 cycles |
372051 cycles |
0.99 |
ML-DSA-87 keypair |
639519 cycles |
645476 cycles |
0.99 |
ML-DSA-87 sign |
1266823 cycles |
1330055 cycles |
0.95 |
ML-DSA-87 verify |
621041 cycles |
629238 cycles |
0.99 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Arm Cortex-A72 (Raspberry Pi 4) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
301525 cycles |
297514 cycles |
1.01 |
ML-DSA-44 sign |
1092324 cycles |
1130782 cycles |
0.97 |
ML-DSA-44 verify |
332434 cycles |
330092 cycles |
1.01 |
ML-DSA-65 keypair |
547265 cycles |
544902 cycles |
1.00 |
ML-DSA-65 sign |
1720491 cycles |
1850869 cycles |
0.93 |
ML-DSA-65 verify |
530876 cycles |
528082 cycles |
1.01 |
ML-DSA-87 keypair |
854708 cycles |
848422 cycles |
1.01 |
ML-DSA-87 sign |
2241005 cycles |
2353835 cycles |
0.95 |
ML-DSA-87 verify |
882179 cycles |
881818 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
SpacemiT K1 8 (Banana Pi F3) benchmarks (no-opt)
Details
| Benchmark suite | Current: 6840735 | Previous: 929b149 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
759381 cycles |
757723 cycles |
1.00 |
ML-DSA-44 sign |
3011050 cycles |
3138207 cycles |
0.96 |
ML-DSA-44 verify |
858873 cycles |
856800 cycles |
1.00 |
ML-DSA-65 keypair |
1288355 cycles |
1288667 cycles |
1.00 |
ML-DSA-65 sign |
4731285 cycles |
5093210 cycles |
0.93 |
ML-DSA-65 verify |
1367624 cycles |
1367550 cycles |
1.00 |
ML-DSA-87 keypair |
2105568 cycles |
2110630 cycles |
1.00 |
ML-DSA-87 sign |
5995343 cycles |
6351787 cycles |
0.94 |
ML-DSA-87 verify |
2220607 cycles |
2226957 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
hanno-becker
left a comment
There was a problem hiding this comment.
@mkannwischer This looks like an impressive improvement. What's the status here? Are you planning further work / cleanup, or should this be reviewed?
Ready for review and I think we should make this change. |
Check ||w0 - cs2||_inf < GAMMA2 - BETA before ||z||_inf < GAMMA1 - BETA, and compute ct0 along with its norm check in a separate loop after the check on z. Both loops and mld_compute_pack_z stop at the first polynomial that fails their check, so an attempt is cheapest when the check that rejects most often per polynomial runs first. Thanks to Georg Land for pointing out that this is faster. Signed-off-by: Matthias J. Kannwischer <matthias@zerorisc.com>
6840735 to
271b35a
Compare
Check ||w0 - cs2||_inf < GAMMA2 - BETA before ||z||_inf < GAMMA1 - BETA, and compute ct0 along with its norm check in a separate loop after the check on z.
Both loops and mld_compute_pack_z stop at the first polynomial that fails their check, so an attempt is cheapest when the check that rejects most often per polynomial runs first.
Thanks to Georg Land for pointing out that this is faster.