-
Notifications
You must be signed in to change notification settings - Fork 34
Verify memory usage: Re-use t1/w1 buffer #866
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
d324b92 to
099669a
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A76 (Raspberry Pi 5) benchmarks (opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
114041 cycles |
113940 cycles |
1.00 |
ML-DSA-44 sign |
358262 cycles |
358593 cycles |
1.00 |
ML-DSA-44 verify |
117990 cycles |
117948 cycles |
1.00 |
ML-DSA-65 keypair |
197906 cycles |
197605 cycles |
1.00 |
ML-DSA-65 sign |
593328 cycles |
593349 cycles |
1.00 |
ML-DSA-65 verify |
194602 cycles |
194728 cycles |
1.00 |
ML-DSA-87 keypair |
323373 cycles |
323551 cycles |
1.00 |
ML-DSA-87 sign |
758771 cycles |
759497 cycles |
1.00 |
ML-DSA-87 verify |
320597 cycles |
320221 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mac Mini (M1, 2020) benchmarks (opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
46414 cycles |
46414 cycles |
1 |
ML-DSA-44 sign |
132149 cycles |
132119 cycles |
1.00 |
ML-DSA-44 verify |
47770 cycles |
47770 cycles |
1 |
ML-DSA-65 keypair |
81322 cycles |
81324 cycles |
1.00 |
ML-DSA-65 sign |
217450 cycles |
217246 cycles |
1.00 |
ML-DSA-65 verify |
80063 cycles |
80053 cycles |
1.00 |
ML-DSA-87 keypair |
132975 cycles |
132577 cycles |
1.00 |
ML-DSA-87 sign |
278882 cycles |
278661 cycles |
1.00 |
ML-DSA-87 verify |
130846 cycles |
130402 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mac Mini (M1, 2020) benchmarks (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
114401 cycles |
114364 cycles |
1.00 |
ML-DSA-44 sign |
419014 cycles |
418649 cycles |
1.00 |
ML-DSA-44 verify |
122338 cycles |
122371 cycles |
1.00 |
ML-DSA-65 keypair |
196308 cycles |
195798 cycles |
1.00 |
ML-DSA-65 sign |
684170 cycles |
683876 cycles |
1.00 |
ML-DSA-65 verify |
197811 cycles |
197587 cycles |
1.00 |
ML-DSA-87 keypair |
322949 cycles |
323082 cycles |
1.00 |
ML-DSA-87 sign |
865849 cycles |
865971 cycles |
1.00 |
ML-DSA-87 verify |
328538 cycles |
328329 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A76 (Raspberry Pi 5) benchmarks (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
213501 cycles |
213269 cycles |
1.00 |
ML-DSA-44 sign |
762561 cycles |
762761 cycles |
1.00 |
ML-DSA-44 verify |
229215 cycles |
229231 cycles |
1.00 |
ML-DSA-65 keypair |
381784 cycles |
381458 cycles |
1.00 |
ML-DSA-65 sign |
1254764 cycles |
1255541 cycles |
1.00 |
ML-DSA-65 verify |
372158 cycles |
371461 cycles |
1.00 |
ML-DSA-87 keypair |
606643 cycles |
605334 cycles |
1.00 |
ML-DSA-87 sign |
1596557 cycles |
1598639 cycles |
1.00 |
ML-DSA-87 verify |
617389 cycles |
617579 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Intel Xeon 4th gen (c7i)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
34586 cycles |
34560 cycles |
1.00 |
ML-DSA-44 sign |
120748 cycles |
120458 cycles |
1.00 |
ML-DSA-44 verify |
38140 cycles |
38041 cycles |
1.00 |
ML-DSA-65 keypair |
61601 cycles |
61473 cycles |
1.00 |
ML-DSA-65 sign |
203120 cycles |
202795 cycles |
1.00 |
ML-DSA-65 verify |
62789 cycles |
62620 cycles |
1.00 |
ML-DSA-87 keypair |
94044 cycles |
93431 cycles |
1.01 |
ML-DSA-87 sign |
237111 cycles |
232760 cycles |
1.02 |
ML-DSA-87 verify |
94591 cycles |
94310 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Intel Xeon 4th gen (c7i) (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
94517 cycles |
94077 cycles |
1.00 |
ML-DSA-44 sign |
333634 cycles |
333569 cycles |
1.00 |
ML-DSA-44 verify |
100077 cycles |
99920 cycles |
1.00 |
ML-DSA-65 keypair |
161142 cycles |
160791 cycles |
1.00 |
ML-DSA-65 sign |
546320 cycles |
548071 cycles |
1.00 |
ML-DSA-65 verify |
161327 cycles |
161736 cycles |
1.00 |
ML-DSA-87 keypair |
267769 cycles |
267351 cycles |
1.00 |
ML-DSA-87 sign |
714142 cycles |
713822 cycles |
1.00 |
ML-DSA-87 verify |
271140 cycles |
269242 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AMD EPYC 3rd gen (c6a)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
69559 cycles |
68971 cycles |
1.01 |
ML-DSA-44 sign |
188330 cycles |
188227 cycles |
1.00 |
ML-DSA-44 verify |
69096 cycles |
68895 cycles |
1.00 |
ML-DSA-65 keypair |
119267 cycles |
119215 cycles |
1.00 |
ML-DSA-65 sign |
302594 cycles |
302342 cycles |
1.00 |
ML-DSA-65 verify |
115090 cycles |
114992 cycles |
1.00 |
ML-DSA-87 keypair |
203553 cycles |
210188 cycles |
0.97 |
ML-DSA-87 sign |
396203 cycles |
406594 cycles |
0.97 |
ML-DSA-87 verify |
192614 cycles |
200557 cycles |
0.96 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Intel Xeon 3rd gen (c6i)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
56618 cycles |
56710 cycles |
1.00 |
ML-DSA-44 sign |
182976 cycles |
182366 cycles |
1.00 |
ML-DSA-44 verify |
61201 cycles |
61343 cycles |
1.00 |
ML-DSA-65 keypair |
99131 cycles |
99152 cycles |
1.00 |
ML-DSA-65 sign |
300800 cycles |
302439 cycles |
0.99 |
ML-DSA-65 verify |
100582 cycles |
100434 cycles |
1.00 |
ML-DSA-87 keypair |
153242 cycles |
152988 cycles |
1.00 |
ML-DSA-87 sign |
358122 cycles |
356727 cycles |
1.00 |
ML-DSA-87 verify |
153576 cycles |
153085 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton4
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
68607 cycles |
68624 cycles |
1.00 |
ML-DSA-44 sign |
203714 cycles |
203573 cycles |
1.00 |
ML-DSA-44 verify |
70764 cycles |
70745 cycles |
1.00 |
ML-DSA-65 keypair |
121923 cycles |
121712 cycles |
1.00 |
ML-DSA-65 sign |
333969 cycles |
334444 cycles |
1.00 |
ML-DSA-65 verify |
117894 cycles |
117921 cycles |
1.00 |
ML-DSA-87 keypair |
198907 cycles |
199055 cycles |
1.00 |
ML-DSA-87 sign |
431170 cycles |
430519 cycles |
1.00 |
ML-DSA-87 verify |
194576 cycles |
195026 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AMD EPYC 3rd gen (c6a) (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
135046 cycles |
134935 cycles |
1.00 |
ML-DSA-44 sign |
526783 cycles |
525271 cycles |
1.00 |
ML-DSA-44 verify |
147525 cycles |
147402 cycles |
1.00 |
ML-DSA-65 keypair |
227378 cycles |
226843 cycles |
1.00 |
ML-DSA-65 sign |
862294 cycles |
861368 cycles |
1.00 |
ML-DSA-65 verify |
235341 cycles |
234653 cycles |
1.00 |
ML-DSA-87 keypair |
371208 cycles |
370947 cycles |
1.00 |
ML-DSA-87 sign |
1081442 cycles |
1080604 cycles |
1.00 |
ML-DSA-87 verify |
383138 cycles |
382813 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Intel Xeon 3rd gen (c6i) (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
157952 cycles |
157528 cycles |
1.00 |
ML-DSA-44 sign |
549923 cycles |
550092 cycles |
1.00 |
ML-DSA-44 verify |
169041 cycles |
169850 cycles |
1.00 |
ML-DSA-65 keypair |
268541 cycles |
268576 cycles |
1.00 |
ML-DSA-65 sign |
906096 cycles |
906359 cycles |
1.00 |
ML-DSA-65 verify |
274414 cycles |
273697 cycles |
1.00 |
ML-DSA-87 keypair |
449829 cycles |
448611 cycles |
1.00 |
ML-DSA-87 sign |
1161108 cycles |
1160142 cycles |
1.00 |
ML-DSA-87 verify |
457140 cycles |
458086 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton4 (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
128825 cycles |
128728 cycles |
1.00 |
ML-DSA-44 sign |
449451 cycles |
449453 cycles |
1.00 |
ML-DSA-44 verify |
138146 cycles |
136972 cycles |
1.01 |
ML-DSA-65 keypair |
220813 cycles |
220815 cycles |
1.00 |
ML-DSA-65 sign |
730337 cycles |
730757 cycles |
1.00 |
ML-DSA-65 verify |
222907 cycles |
221434 cycles |
1.01 |
ML-DSA-87 keypair |
365599 cycles |
365897 cycles |
1.00 |
ML-DSA-87 sign |
929258 cycles |
928934 cycles |
1.00 |
ML-DSA-87 verify |
372855 cycles |
370579 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AMD EPYC 4th gen (c7a)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
41908 cycles |
41746 cycles |
1.00 |
ML-DSA-44 sign |
133621 cycles |
133873 cycles |
1.00 |
ML-DSA-44 verify |
44579 cycles |
44753 cycles |
1.00 |
ML-DSA-65 keypair |
72806 cycles |
75286 cycles |
0.97 |
ML-DSA-65 sign |
213586 cycles |
215708 cycles |
0.99 |
ML-DSA-65 verify |
73057 cycles |
74765 cycles |
0.98 |
ML-DSA-87 keypair |
107522 cycles |
110095 cycles |
0.98 |
ML-DSA-87 sign |
252638 cycles |
256250 cycles |
0.99 |
ML-DSA-87 verify |
109549 cycles |
111605 cycles |
0.98 |
This comment was automatically generated by workflow using github-action-benchmark.
oqs-bot
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton3
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
72655 cycles |
72734 cycles |
1.00 |
ML-DSA-44 sign |
214088 cycles |
214040 cycles |
1.00 |
ML-DSA-44 verify |
75813 cycles |
75728 cycles |
1.00 |
ML-DSA-65 keypair |
128178 cycles |
128121 cycles |
1.00 |
ML-DSA-65 sign |
354357 cycles |
354117 cycles |
1.00 |
ML-DSA-65 verify |
125434 cycles |
125363 cycles |
1.00 |
ML-DSA-87 keypair |
206400 cycles |
209489 cycles |
0.99 |
ML-DSA-87 sign |
448036 cycles |
454515 cycles |
0.99 |
ML-DSA-87 verify |
205203 cycles |
205358 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton2
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
114270 cycles |
114038 cycles |
1.00 |
ML-DSA-44 sign |
359124 cycles |
358900 cycles |
1.00 |
ML-DSA-44 verify |
118342 cycles |
117956 cycles |
1.00 |
ML-DSA-65 keypair |
198694 cycles |
197540 cycles |
1.01 |
ML-DSA-65 sign |
595452 cycles |
593954 cycles |
1.00 |
ML-DSA-65 verify |
195260 cycles |
194746 cycles |
1.00 |
ML-DSA-87 keypair |
323679 cycles |
324592 cycles |
1.00 |
ML-DSA-87 sign |
759606 cycles |
760860 cycles |
1.00 |
ML-DSA-87 verify |
320697 cycles |
320931 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
AMD EPYC 4th gen (c7a) (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
119897 cycles |
120479 cycles |
1.00 |
ML-DSA-44 sign |
446804 cycles |
447990 cycles |
1.00 |
ML-DSA-44 verify |
129949 cycles |
130348 cycles |
1.00 |
ML-DSA-65 keypair |
204619 cycles |
204777 cycles |
1.00 |
ML-DSA-65 sign |
727135 cycles |
728464 cycles |
1.00 |
ML-DSA-65 verify |
210312 cycles |
209710 cycles |
1.00 |
ML-DSA-87 keypair |
337048 cycles |
341820 cycles |
0.99 |
ML-DSA-87 sign |
924977 cycles |
936896 cycles |
0.99 |
ML-DSA-87 verify |
347829 cycles |
349579 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton3 (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
139013 cycles |
138964 cycles |
1.00 |
ML-DSA-44 sign |
486029 cycles |
485918 cycles |
1.00 |
ML-DSA-44 verify |
148508 cycles |
148562 cycles |
1.00 |
ML-DSA-65 keypair |
242989 cycles |
242855 cycles |
1.00 |
ML-DSA-65 sign |
797054 cycles |
796582 cycles |
1.00 |
ML-DSA-65 verify |
240787 cycles |
241149 cycles |
1.00 |
ML-DSA-87 keypair |
396472 cycles |
397298 cycles |
1.00 |
ML-DSA-87 sign |
1016729 cycles |
1016612 cycles |
1.00 |
ML-DSA-87 verify |
402637 cycles |
402113 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.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Graviton2 (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
213770 cycles |
213497 cycles |
1.00 |
ML-DSA-44 sign |
763291 cycles |
763502 cycles |
1.00 |
ML-DSA-44 verify |
229425 cycles |
229441 cycles |
1.00 |
ML-DSA-65 keypair |
382272 cycles |
381881 cycles |
1.00 |
ML-DSA-65 sign |
1256084 cycles |
1257130 cycles |
1.00 |
ML-DSA-65 verify |
372530 cycles |
371845 cycles |
1.00 |
ML-DSA-87 keypair |
607903 cycles |
605946 cycles |
1.00 |
ML-DSA-87 sign |
1599026 cycles |
1600239 cycles |
1.00 |
ML-DSA-87 verify |
618349 cycles |
617930 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
SpacemiT K1 8 (Banana Pi F3) benchmarks (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
829072 cycles |
829753 cycles |
1.00 |
ML-DSA-44 sign |
3248535 cycles |
3243434 cycles |
1.00 |
ML-DSA-44 verify |
923349 cycles |
921839 cycles |
1.00 |
ML-DSA-65 keypair |
1412138 cycles |
1414144 cycles |
1.00 |
ML-DSA-65 sign |
5364520 cycles |
5333778 cycles |
1.01 |
ML-DSA-65 verify |
1479739 cycles |
1476828 cycles |
1.00 |
ML-DSA-87 keypair |
2311432 cycles |
2313670 cycles |
1.00 |
ML-DSA-87 sign |
6663285 cycles |
6677266 cycles |
1.00 |
ML-DSA-87 verify |
2405801 cycles |
2405289 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A55 (Snapdragon 888) benchmarks (opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
276776 cycles |
274848 cycles |
1.01 |
ML-DSA-44 sign |
819362 cycles |
810924 cycles |
1.01 |
ML-DSA-44 verify |
274034 cycles |
272523 cycles |
1.01 |
ML-DSA-65 keypair |
471949 cycles |
473918 cycles |
1.00 |
ML-DSA-65 sign |
1344711 cycles |
1350357 cycles |
1.00 |
ML-DSA-65 verify |
449926 cycles |
450974 cycles |
1.00 |
ML-DSA-87 keypair |
808114 cycles |
806468 cycles |
1.00 |
ML-DSA-87 sign |
1826738 cycles |
1813351 cycles |
1.01 |
ML-DSA-87 verify |
778882 cycles |
774394 cycles |
1.01 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A55 (Snapdragon 888) benchmarks (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
464741 cycles |
465764 cycles |
1.00 |
ML-DSA-44 sign |
2149417 cycles |
2155517 cycles |
1.00 |
ML-DSA-44 verify |
547085 cycles |
547931 cycles |
1.00 |
ML-DSA-65 keypair |
781127 cycles |
780509 cycles |
1.00 |
ML-DSA-65 sign |
3527778 cycles |
3529186 cycles |
1.00 |
ML-DSA-65 verify |
850554 cycles |
849102 cycles |
1.00 |
ML-DSA-87 keypair |
1261547 cycles |
1256693 cycles |
1.00 |
ML-DSA-87 sign |
4447830 cycles |
4361997 cycles |
1.02 |
ML-DSA-87 verify |
1367645 cycles |
1364343 cycles |
1.00 |
This comment was automatically generated by workflow using github-action-benchmark.
This commit is the first of a series of commits reducing the stack usage of verification. It is hoisted out from #751 This commit places the t1 and w1 buffers into a union saving K KiB of memory. Operations using it are slightly reordered such that their lifetime does not overlap. As CBMC struggles with unions (issue 8813), we use the same workaround present in signing: Use a struct by default, and a union when MLD_CONFIG_REDUCE_RAM is set. Signed-off-by: Matthias J. Kannwischer <[email protected]>
099669a to
4e6eca3
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A72 (Raspberry Pi 4) benchmarks (opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
225942 cycles |
225497 cycles |
1.00 |
ML-DSA-44 sign |
643730 cycles |
626811 cycles |
1.03 |
ML-DSA-44 verify |
228639 cycles |
224639 cycles |
1.02 |
ML-DSA-65 keypair |
404935 cycles |
400938 cycles |
1.01 |
ML-DSA-65 sign |
1082998 cycles |
1030229 cycles |
1.05 |
ML-DSA-65 verify |
389823 cycles |
376238 cycles |
1.04 |
ML-DSA-87 keypair |
700313 cycles |
668182 cycles |
1.05 |
ML-DSA-87 sign |
1422286 cycles |
1383907 cycles |
1.03 |
ML-DSA-87 verify |
686409 cycles |
640032 cycles |
1.07 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
⚠️ 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: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-65 sign |
1082998 cycles |
1030229 cycles |
1.05 |
ML-DSA-65 verify |
389823 cycles |
376238 cycles |
1.04 |
ML-DSA-87 keypair |
700313 cycles |
668182 cycles |
1.05 |
ML-DSA-87 verify |
686409 cycles |
640032 cycles |
1.07 |
This comment was automatically generated by workflow using github-action-benchmark.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Arm Cortex-A72 (Raspberry Pi 4) benchmarks (no-opt)
Details
| Benchmark suite | Current: 099669a | Previous: f22bae3 | Ratio |
|---|---|---|---|
ML-DSA-44 keypair |
302400 cycles |
316786 cycles |
0.95 |
ML-DSA-44 sign |
1142931 cycles |
1200912 cycles |
0.95 |
ML-DSA-44 verify |
331020 cycles |
343490 cycles |
0.96 |
ML-DSA-65 keypair |
551404 cycles |
567375 cycles |
0.97 |
ML-DSA-65 sign |
1911875 cycles |
1980533 cycles |
0.97 |
ML-DSA-65 verify |
526969 cycles |
541638 cycles |
0.97 |
ML-DSA-87 keypair |
865083 cycles |
881612 cycles |
0.98 |
ML-DSA-87 sign |
2416111 cycles |
2490533 cycles |
0.97 |
ML-DSA-87 verify |
887865 cycles |
905931 cycles |
0.98 |
This comment was automatically generated by workflow using github-action-benchmark.
This commit is the first of a series of commits reducing the stack usage of verification.
It is hoisted out from #751
This commit places the t1 and w1 buffers into a union saving K KiB of memory. Operations using it are slightly reordered such that their lifetime does not overlap.
As CBMC struggles with unions (issue 8813), we use the same workaround present in signing: Use a struct by default, and a union when MLD_CONFIG_REDUCE_RAM is set.