feat: merge-train/barretenberg#23850
Merged
Merged
Conversation
…debug-build segfault (#23847) ## Problem The nightly barretenberg **debug** build has been failing (aztec-claude run [26935061960](https://github.com/AztecProtocol/aztec-claude/actions/runs/26935061960); same failure in aztec-packages runs #105/#106). The build dies with `exit status 139` (SIGSEGV) on: ``` FAILED ... ecc_tests PippengerConstantine.SimdX4MatchesScalarPathLanewise (code: 139) [ RUN ] PippengerConstantine.SimdX4MatchesScalarPathLanewise timeout: the monitored command dumped core ``` ## Root cause In `barretenberg/cpp/src/barretenberg/ecc/scalar_multiplication/pippenger_constantine.hpp`, `simd_u32x4_store` writes the result vector with: ```cpp *reinterpret_cast<SimdU32x4*>(dst) = v; ``` `SimdU32x4` is `uint32_t __attribute__((vector_size(16)))`, which carries **16-byte alignment**, so this is an *aligned* 128-bit store. But `dst` is an arbitrary `uint32_t*` — the test and fuzzer pass a stack `std::array<uint32_t, 4>` (4-byte aligned). At `-O0` (debug) the store lowers to an alignment-requiring `movaps`/`movdqa` and faults whenever `dst` is not 16-byte aligned. This only surfaces in the **debug** nightly: the helper is `[[gnu::always_inline]]`, so at `-O2` SROA promotes the local `out` array into registers and the memory store is elided — which is why the full (release) CI is green while the debug build segfaults. The SIMD x4 helpers are currently consumed only by the unit test and fuzzer (not yet wired into the MSM hot loop), so the blast radius is the test/fuzzer. ## Fix Store via `__builtin_memcpy`, which has no alignment precondition and lowers to the intended unaligned `movdqu` / NEON `st1` (the WASM `wasm_v128_store` path is unchanged). This matches the helper's documented intent. ## Verification (red/green, debug preset) Built `ecc_tests` with the `debug` CMake preset (`build-debug`, `-O0 -D_GLIBCXX_DEBUG`), matching the nightly: - **Without the fix:** `PippengerConstantine.SimdX4MatchesScalarPathLanewise` → exit **139** (SIGSEGV), reproducing the nightly. - **With the fix:** all 6 `PippengerConstantine.*` tests pass. A standalone repro confirmed the mechanism independently: the aligned store to a 4-byte-aligned destination segfaults at `-O0`; the `memcpy` form stores correctly. --- *Created by [claudebox](https://claudebox.work/v2/sessions/cadf49316638602b) · group: `slackbot`* --------- Co-authored-by: iakovenkos <sergey.s.yakovenko@gmail.com>
iakovenkos
approved these changes
Jun 4, 2026
Collaborator
Author
Flakey Tests🤖 says: This CI run detected 2 tests that failed, but were tolerated due to a .test_patterns.yml entry. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See merge-train-readme.md.
This is a merge-train.