Add tooling to compare execution cycles between two MASP packages
#1.031 geöffnet am 25.03.2026
Repository-Metriken
- Stars
- (110 Sterne)
- PR-Merge-Metriken
- (Durchschn. Merge 4T 7h) (51 gemergte PRs in 30 T)
Beschreibung
Continuation of protocol-1920
The goal is comparing the performance of hand-written MASM vs code generated by the compiler.
We have the benchmark for the MASM P2ID note in the protocol repo (see the results at https://github.com/0xMiden/protocol/blob/next/bin/bench-transaction/bench-tx.json). We should replicate at least one scenario in tests/integration-network/src/mockchain/basic_wallet.rs and measure the number of cycles for the Rust version of the P2ID note. Ideally we want the latest numbers posted in an easy to find place (in a README file?).
Potential follow-up
Use profiling infra to report not only total cycle count diffs but also provide insights into where cycles are spent.