review round 3: close the two nonblocking nits

Round 3 returned fit-once-one-item-lands (first non-DO-NOT-ATTEST verdict).
The one blocker is an independently-executed toolchain session, which is
operator-side. This commit closes the two nonblocking recommendations:

- extract.sh: source the aeneas toolchain env AFTER the provenance guard, so
  a party without the toolchain gets the clean exit-2/3/4 diagnostics instead
  of a confusing env error (the reviewer hit exactly this). Re-verified: exit
  2/3 clean in an env-stripped shell, happy path byte-identical.
- RECORDED-RUN.md: append the locked cargo run under the pinned nightly —
  the differential bridge test mono_matches_deployed_verify passes with
  --locked --lib on nightly-2026-06-01, and the full default-features suite
  is green (13+3+12+37 passed, 0 failed). Also documents why single-feature
  integration tests cannot compile (upstream test files hardcode all twelve
  parameter sets), resolving the round-1 GPT observation.

No proof, model, or gate change.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
mrwulf 2026-07-24 21:31:05 +02:00
parent 522d8b2092
commit 77d6b2c195
2 changed files with 50 additions and 1 deletions

View file

@ -48,3 +48,48 @@ fips205.wots_csum_loop_eq :: propext, Classical.choice, Quot.sound
fips205.base2b_outer_loop_eq :: propext, Classical.choice, Quot.sound fips205.base2b_outer_loop_eq :: propext, Classical.choice, Quot.sound
fips205.slh_verify_128s_accepts_iff :: propext, Classical.choice, Quot.sound, verify_mono.oracle.f, verify_mono.oracle.h, verify_mono.oracle.h_msg, verify_mono.oracle.t_l, verify_mono.oracle.t_len fips205.slh_verify_128s_accepts_iff :: propext, Classical.choice, Quot.sound, verify_mono.oracle.f, verify_mono.oracle.h, verify_mono.oracle.h_msg, verify_mono.oracle.t_l, verify_mono.oracle.t_len
``` ```
## Locked cargo run under the pinned nightly (round-3 reviewer recommendation)
The differential test `mono_matches_deployed_verify` is the sole
verify_mono→deployed bridge; recorded here under `--locked` on the
toolchain pinned by `rust-toolchain.toml` (nightly-2026-06-01, the
transpiler's own channel). Note: the upstream integration-test files
hardcode all twelve parameter sets, so they only build with default
features — the single-feature run therefore uses `--lib` (which contains
the differential test), and the full suite runs with default features.
This explains the round-1 GPT observation that single-feature
`cargo test` fails to compile: an upstream test-layout property, not a
defect of the snapshot.
```
=== LOCKED CARGO RUN — fips205-source @ 797b4ef26338e27363683656f93cb065a77daa0e ===
date(UTC): 20260724T192246Z
toolchain: nightly-2026-06-01-x86_64-unknown-linux-gnu (overridden by '/home/oho/GitClone/FormalVerification/sources/fips205-source/rust-toolchain.toml')
--- (1) differential test (verify_mono vs deployed generic verifier), lib tests, SHA2-128s feature ---
running 2 tests
test verify_mono::tests::mono_matches_deployed_verify ... ok
test slh_dsa_sha2_128s::tests::simple_round_trips ... ok
test result: ok. 2 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 10.82s
--- (2) full upstream suite, default features (all parameter sets), --locked ---
test src/traits.rs - traits::Signer::try_sign_with_rng (line 284) ... ok
test src/traits.rs - traits::Verifier::hash_verify (line 455) ... ok
test src/traits.rs - traits::Verifier::verify (line 415) ... ok
test result: ok. 37 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 28.20s
--- (2b) full-suite summary lines (all test binaries + doctests) ---
running 13 tests
test result: ok. 13 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 39.25s
running 3 tests
test result: ok. 3 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 143.21s
running 1 test
test result: ok. 0 passed; 0 failed; 1 ignored; 0 measured; 0 filtered out; finished in 0.00s
running 12 tests
test result: ok. 12 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 6.39s
running 37 tests
test result: ok. 37 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 27.89s
```

View file

@ -28,7 +28,6 @@
# EXPECTED_SRC_COMMIT=<full-sha> ./extract.sh [PATH] # EXPECTED_SRC_COMMIT=<full-sha> ./extract.sh [PATH]
set -euo pipefail set -euo pipefail
source ~/aeneas-toolchain/env.sh
HERE="$(cd "$(dirname "$0")" && pwd)" HERE="$(cd "$(dirname "$0")" && pwd)"
CRATE="${1:-$HOME/GitClone/FormalVerification/sources/fips205-source}" CRATE="${1:-$HOME/GitClone/FormalVerification/sources/fips205-source}"
@ -53,6 +52,11 @@ if [ -n "$(git -C "$CRATE" status --porcelain)" ]; then
fi fi
echo "[0/2] provenance OK: fips205-source @ ${EXPECTED_SRC_COMMIT:0:12} (clean)" echo "[0/2] provenance OK: fips205-source @ ${EXPECTED_SRC_COMMIT:0:12} (clean)"
# Toolchain env is sourced only AFTER the provenance guard, so a party without
# the aeneas toolchain still gets the clean exit-2/3/4 diagnostics above
# (round-3 reviewer nit, 2026-07-24).
source ~/aeneas-toolchain/env.sh
echo "[1/2] charon: Rust -> LLBC (monomorphic SHA2-128s verify cone;" echo "[1/2] charon: Rust -> LLBC (monomorphic SHA2-128s verify cone;"
echo " crate::verify_mono::oracle is the opaque SHA-2 boundary)" echo " crate::verify_mono::oracle is the opaque SHA-2 boundary)"
cd "$CRATE" cd "$CRATE"