shreyas-londhe:perf/drop-linear-forms - Branch - WizardOfMenlo/whir

perf(whir_zk): IRS-coeff residency + output-pruned NTT for f̂ opens

#252

Comparing

shreyas-londhe:perf/drop-linear-forms

(

c39ce01

) with

(

ec7aa32

)

-19%

Regression: 4

Untouched: 6

Skipped: 22

No successful run was found on v1 ( 0a68627) during the generation of this report, so main ( ec7aa32) was used instead as the comparison base. There might be some changes unrelated to this pull request in this report.

Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments, which may affect the accuracy of the results. Learn more about regression causes.

Benchmarks

32 total

interleaved_rs_encode[(20, 4, 4)]

benches/expand_from_coeff.rs

-24%

613.3 ms807.9 ms

interleaved_rs_encode[(18, 2, 2)]

benches/expand_from_coeff.rs

-20%

68.8 ms86.3 ms

interleaved_rs_encode[(22, 4, 4)]

benches/expand_from_coeff.rs

-17%

3.4 s4.1 s

interleaved_rs_encode[(18, 4, 3)]

benches/expand_from_coeff.rs

-14%

155.2 ms179.9 ms

sumcheck_first_round[1048576]

benches/sumcheck.rs

34.3 ms34.3 ms

sumcheck_first_round[262144]

benches/sumcheck.rs

6.9 ms6.9 ms

interleaved_rs_encode[(16, 2, 2)]

benches/expand_from_coeff.rs

-3%

16.9 ms17.4 ms

interleaved_rs_encode[(16, 4, 3)]

benches/expand_from_coeff.rs

-3%

32.4 ms33.4 ms

sumcheck_first_round[65536]

benches/sumcheck.rs

-3%

1.8 ms1.9 ms

interleaved_rs_encode[(20, 2, 3)]

benches/expand_from_coeff.rs

-8%

373.2 ms407.7 ms

The benchmarks below were skipped, so their baseline results are used instead. If they were deleted in your codebase, archive them to remove them from the performance reports.
Learn more about archiving benchmarks

expand_from_coeff[(20, 2)]

benches/expand_from_coeff.rs

Skipped

69 ms*

expand_from_coeff[(24, 4)]

benches/expand_from_coeff.rs

Skipped

2.8 s*

sumcheck_first_round[67108864]

benches/sumcheck.rs

Skipped

1.2 s*

expand_from_coeff[(20, 4)]

benches/expand_from_coeff.rs

Skipped

144.3 ms*

sumcheck_first_round[1048576]

benches/sumcheck.rs

Skipped

9.5 ms*

expand_from_coeff[(26, 4)]

benches/expand_from_coeff.rs

Skipped

13 s*

expand_from_coeff[(18, 4)]

benches/expand_from_coeff.rs

Skipped

26.2 ms*

expand_from_coeff[(26, 2)]

benches/expand_from_coeff.rs

Skipped

6.5 s*

sumcheck_first_round[262144]

benches/sumcheck.rs

Skipped

4.7 ms*

expand_from_coeff[(18, 2)]

benches/expand_from_coeff.rs

Skipped

18.2 ms*

sumcheck_first_round[65536]

benches/sumcheck.rs

Skipped

866.3 µs*

expand_from_coeff[(22, 4)]

benches/expand_from_coeff.rs

Skipped

662.4 ms*

expand_from_coeff[(22, 2)]

benches/expand_from_coeff.rs

Skipped

333.9 ms*

sumcheck_first_round[16777216]

benches/sumcheck.rs

Skipped

266.6 ms*

expand_from_coeff[(24, 2)]

benches/expand_from_coeff.rs

Skipped

1.4 s*

Commits

Click on a commit to change the comparison range

Base

main

ec7aa32

-18.77%

perf(whir_zk): drop linear_forms after covector build

c183108

17 days ago

by shreyas-londhe

-0.03%

perf(whir_zk): hold IRS coefficients, re-encode codeword on demand

fc3e614

16 days ago

by shreyas-londhe

-0.02%

chore(whir_zk): clippy + signature cleanup post-pruned-NTT

c39ce01

16 days ago

by shreyas-londhe

Home Terms Privacy Docs