Additional testing for changes due to rayon being replaced by p3::maybe-rayon

[Sahilgill24]

Motivation

#861

This is a test PR to check the changes made in the gkr-backend PR. The dependencies in the Cargo.toml here are pointing to the feat/p3-maybe-rayon branch.

Status

1. cargo make tests
   The tests are working completely fine .

2. cargo bench -p ceno_zkvm
benches/bitwise_keccakf.rs is giving an error for both the master branch and this branch test/p3-maybe-rayon , rest all are same.

Now for comparing the traces for both the cases, I was confused about which tests to change that use rayon and would show the difference in the stack trace.

[Sahilgill24]

Hey @hero78119 , I have made the PR here and mentioned about the tests, please confirm the stack trace related part once.

[hero78119]

There is a commit to complete overall feature
hero78119@e5598ed

which I had tested with new commit here scroll-tech/gkr-backend#8 (comment)

cargo bench -p ceno_zkvm benches/bitwise_keccakf.rs is giving an error for both the master branch and this branch test/p3-maybe-rayon , rest all are same

Yes this bench currently not work as expected as it's a bit out dated, and we already switch tolookup_keccakf

The benchmark command I run in local

master branch

JEMALLOC_SYS_WITH_MALLOC_CONF=retain:true,metadata_thp:always,thp:always,dirty_decay_ms:-1,muzzy_decay_ms:-1,abort_conf:true cargo bench --features jemalloc --bench fibonacci --package ceno_zkvm -- --save-baseline baseline

feature branch, with new parallel features.

JEMALLOC_SYS_WITH_MALLOC_CONF=retain:true,metadata_thp:always,thp:always,dirty_decay_ms:-1,muzzy_decay_ms:-1,abort_conf:true cargo bench --features parallel --features jemalloc --bench fibonacci --package ceno_zkvm -- --baseline baseline

And here is benchmark result on my local, overall no change as expected

CPU: 5900XT 32 cores
RAM: 64GB

Benchmark	Median Time (s)	Median Change (%)
fibonacci_max_steps_1048576	2.8407	+1.27% (Change within noise)
fibonacci_max_steps_2097152	4.9774	+0.80% (Change within noise)
fibonacci_max_steps_4194304	9.2673	+0.19% (No significant change)