Files
precompile/sr25519/accel_parity_test.go
zeekay ada296185c fix(accel): eliminate CPU/GPU verdict divergence in verify precompiles (consensus-split + forgery)
Under the enable-everything precompile policy these verify-only precompiles are
no longer refused under strict-PQ, so any divergence between the luxfi/accel
GPU/SIMD path and the pure-CPU path is a LIVE consensus split: GPU-equipped and
CPU-only validators would return different verify verdicts on the same input and
the chain forks. Several accel paths were also outright wrong-primitive forgery
surfaces.

LAW applied: the GPU verdict must be byte-identical to the CPU verdict on every
input, or the accel path is a consensus hazard and is removed. CPU (pure-Go,
CGO_ENABLED=0) is now the single source of truth for every on-chain verdict.
None of these GPU kernels can be proven byte-identical here, and several are
provably NOT, so every verdict-trusting accel branch is deleted and routed to
the canonical CPU verifier. Determinism over speed.

Active forgery/wrong-output in the DEFAULT build (accelcrypto CPU fallback
leaked non-CPU semantics, err==nil):
- sr25519: verified via the accel Ed25519 kernel -> a valid Ed25519 signature
  was ACCEPTED as a valid sr25519/Schnorrkel signature (wrong-curve forgery).
- blake3: accelcrypto.Hash(HashBlake3) returns BLAKE2b, not BLAKE3; the "real
  blake3" fallback was dead code. OpHash256 + MerkleRoot emitted BLAKE2b.

Latent splits (reachable only under -tags accel + a GPU backend):
- frost: FROST-Schnorr verified through the SigECDSA kernel (wrong primitive +
  32B x-only key mis-framed as a 33B ECDSA key).
- cggmp21: SigECDSA kernel truncated the 65B key to 33B and skipped the CPU
  path's recover-and-compare binding (ignored the v byte).
- ed25519: GPU Ed25519 kernel not proven byte-identical to the canonical
  verifier on cofactor / non-canonical-S / small-order edge cases.
- secp256r1: P-256 inputs sent to the fixed-curve secp256k1 ECDSA kernel.
- mldsa: GPU verified with EMPTY context; precompile binds the FIPS-204 ctx
  "lux-evm-precompile-mldsa-v1" -> different mu, different verdict.
- bls12381 / zk(groth16): GPU MSM result trusted without re-verification and
  with no curve binding carried to the kernel.
- kzg4844: GPU commitment used the backend's OWN embedded trusted setup (empty
  SRS) instead of the canonical EIP-4844 SRS -> breaks KZG binding.
- ai_mining (BatchVerifyMLDSA/BatchAccumulator): accel batch kernel truncates
  messages to 32 bytes and is mode-fixed -> diverges from VerifyMLDSA.
- fhe: GPU computed BFV arithmetic while CPU computes TFHE (different scheme,
  incompatible ciphertext, nil relin key) -> corrupt + divergent ciphertext.

Cleanups (no split, vestigial): removed dead mlkem encapsulateGPU; removed the
hpke GPU pre-warm that discarded its result and returned the CPU output anyway.
slhdsa was already CPU-only (prior identical fix).

Tests (differential / forgery proofs, all green CGO_ENABLED=0 and CGO -race):
- sr25519: a valid Ed25519 sig is now REJECTED; real sr25519 still verifies.
- blake3: OpHash256/MerkleRoot output == BLAKE3 and != BLAKE2b.
- cggmp21: flipped recovery byte is rejected (recover-compare authoritative).
- frost: a valid ECDSA sig is rejected as FROST-Schnorr.
- ed25519: verdict matches the canonical ed25519.Verify oracle over 64x4 vectors.
- mldsa: empty-ctx sig (the accel framing) rejected; precompileCtx sig accepted.

Net -593 LOC. No consensus code touched.
2026-06-27 19:46:11 -07:00

63 lines
2.5 KiB
Go

// Copyright (C) 2025, Lux Industries, Inc. All rights reserved.
// See the file LICENSE for licensing terms.
package sr25519
import (
stded25519 "crypto/ed25519"
"crypto/rand"
"testing"
"github.com/luxfi/geth/common"
"github.com/stretchr/testify/require"
)
// TestSR25519_RejectsEd25519Forgery is the regression proof for the
// wrong-curve accel divergence: the precompile used to route through
// luxfi/accel's Ed25519 GPU kernel ("closest GPU kernel"), so a *valid
// Ed25519* signature (same 32-byte key / 64-byte sig shape) was accepted as a
// valid sr25519/Schnorrkel signature, and accel-equipped validators returned
// a different verdict than CPU-only ones.
//
// sr25519 is Schnorrkel over Ristretto255 with a Merlin "substrate"
// transcript; an Ed25519 signature can never satisfy that verification
// equation. After the fix the precompile MUST reject this input.
func TestSR25519_RejectsEd25519Forgery(t *testing.T) {
pub, priv, err := stded25519.GenerateKey(rand.Reader)
require.NoError(t, err)
msg := []byte("cross-scheme forgery: an ed25519 signature presented as sr25519")
sig := stded25519.Sign(priv, msg)
// The input is, by construction, a genuinely valid Ed25519 signature with
// the exact byte-shape sr25519 expects -- precisely what the deleted accel
// (Ed25519 kernel) path verified and accepted.
require.True(t, stded25519.Verify(pub, msg, sig),
"vector must be a valid Ed25519 signature (the input the accel path accepted)")
require.Len(t, []byte(pub), PublicKeySize)
require.Len(t, sig, SignatureSize)
input := buildInput(pub, sig, msg)
gas := SR25519VerifyPrecompile.RequiredGas(input)
result, _, err := SR25519VerifyPrecompile.Run(
nil, common.Address{}, ContractAddress, input, gas, true,
)
require.NoError(t, err)
require.Equal(t, byte(0), result[31],
"FORGERY: ed25519 signature must NOT verify as sr25519 (wrong-curve acceptance closed)")
}
// TestSR25519_RealSignatureStillVerifies pins the positive direction: the CPU
// Schnorrkel path (now the single source of truth) still accepts Alice's
// genuine sr25519 signature. Together with the forgery test this proves the
// verdict is the CPU verdict, byte-stable across the validator set.
func TestSR25519_RealSignatureStillVerifies(t *testing.T) {
input := buildInput(alicePubKey, aliceSignature, aliceMessage)
gas := SR25519VerifyPrecompile.RequiredGas(input)
result, _, err := SR25519VerifyPrecompile.Run(
nil, common.Address{}, ContractAddress, input, gas, true,
)
require.NoError(t, err)
require.Equal(t, byte(1), result[31], "genuine sr25519 signature must verify")
}