mirror of
https://github.com/colbymchenry/codegraph.git
synced 2026-09-19 07:34:57 +08:00
Colby McHenry
cbf84855e7
A C/C++ (or any other kernel-routed) file with extremely deep nesting — clang's 16,384-brace `parser_overflow.c`, fuzzer corpora — parsed fine (tree-sitter is iterative) and then overflowed the native stack of the kernel's recursive walker. A native overflow is uncatchable: the parse worker is a thread of the `codegraph` process, so the SIGSEGV took the whole indexer down with no message, no partial index and no per-file fallback. Worker threads get Node's 4 MiB default stack; the 8 MiB main thread only moved the cliff (100k levels still died), so a bigger `resourceLimits.stackSizeMb` was never a fix. The walkers now guard their own recursion against the CALLING THREAD's real stack bounds (`codegraph-kernel/src/stack.rs`: glibc/musl `pthread_getattr_np`, macOS `pthread_get_stackaddr_np`, Win32 `GetCurrentThreadStackLimits`; one thread-local load + one compare per recursive entry, inserted by the `stack_guard!` macro at all 150 self-recursive / on-cycle walker functions). Within 256 KiB of the limit the walk stops descending and latches a flag; `stack::run_guarded` turns a tripped walk into the kernel's existing `defer:` routing signal, so the file takes the wasm path — whose walker catches its own JS `RangeError` per file — and lands as a partial result with a recorded parse error while the rest of the repository indexes normally. Platforms without a bounds query fall back to a fixed descent budget that is safe on any stack ≥ 2 MiB. No Worker stack bump; no new crates beyond `libc` (already in the lock file transitively). Validated: the reporter's `deep.c` inside a default 4 MiB worker goes from rc=132/139 to a clean `deferred` exit; `codegraph init` on a repo holding it exits 0 with the file recorded; 60k-deep expressions in every default-routed language survive on the main thread and in a worker; Rust unit tests drive the walkers on a 1 MiB thread; all 15 existing kernel parity suites unchanged; index wall-clock on express and redis within run-to-run noise with identical node/edge counts; Linux verified in Docker (node:22-bookworm, glibc bounds path). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01LxZj6W6Y1SHXwvpT3uwJpK
79 lines
3.3 KiB
TOML
79 lines
3.3 KiB
TOML
[package]
|
|
name = "codegraph-kernel"
|
|
version = "0.1.0"
|
|
edition = "2021"
|
|
license = "MIT"
|
|
publish = false
|
|
description = "Native extraction kernel for CodeGraph — tree-sitter parse+extract with one JS boundary crossing per file"
|
|
|
|
[lib]
|
|
crate-type = ["cdylib"]
|
|
|
|
[dependencies]
|
|
napi = { version = "3", default-features = false, features = ["napi8"] }
|
|
napi-derive = "3"
|
|
tree-sitter = "0.25"
|
|
sha2 = "0.10"
|
|
regex = "1"
|
|
|
|
# Grammars — MUST stay revision-matched with the wasm grammars the fallback
|
|
# path loads (tree-sitter-wasms npm package / src/extraction/wasm/). The
|
|
# kernel-grammar-parity test asserts node-kind-table equality at test time;
|
|
# bump these together with the wasm side or that gate fails.
|
|
tree-sitter-typescript = "0.23"
|
|
tree-sitter-javascript = "0.25"
|
|
tree-sitter-java = "0.23"
|
|
tree-sitter-python = "0.23"
|
|
tree-sitter-go = "0.23"
|
|
# Pinned exact: the vendored wasm (src/extraction/wasm/) was built from these
|
|
# tags' checked-in parser.c, sha-matched against these registry tarballs
|
|
# (R7a prep, #1345). A patch bump here without re-vendoring breaks the match.
|
|
tree-sitter-c = "=0.24.2"
|
|
tree-sitter-cpp = "=0.23.4"
|
|
tree-sitter-rust = "=0.24.2"
|
|
# csharp: the vendored wasm (#717) predates the kernel and was verified
|
|
# table-identical to this crate's tarball (ABI 15, STATE_COUNT 8053, node-kind
|
|
# and field tables — csharp checklist header). Bump crate + wasm together.
|
|
tree-sitter-c-sharp = "=0.23.5"
|
|
# ruby: content bump, ABI stays 14 (the v0.23.1 tag predates the ABI-15
|
|
# generator) — kernel-grammar-parity asserts same-revision, not same-ABI.
|
|
tree-sitter-ruby = "=0.23.1"
|
|
# php: the walker calls LANGUAGE_PHP (the full HTML-interleaving variant the
|
|
# wasm ships) — NEVER LANGUAGE_PHP_ONLY, which errors on leading HTML.
|
|
tree-sitter-php = "=0.24.2"
|
|
# swift: the vendored wasm is built from THIS crate's tarball src/ (the tag's
|
|
# checked-in parser.c is an older ABI-14 generation that can never sha-match;
|
|
# grammar.json rules are JSON-equal — swift checklist header). parser.c is
|
|
# ~20MB generated — expect slow compiles.
|
|
tree-sitter-swift = "=0.7.3"
|
|
# r: the vendored wasm IS r-lib v1.2.0 and this crate's tarball ships both
|
|
# generated artifacts sha-identical to the tag (parser.c 6221657347…,
|
|
# scanner.c 1209d11076… — r checklist §Grammar prep). ABI 14 (the tag
|
|
# predates the ABI-15 generator) — parity asserts same-revision, not
|
|
# same-ABI (ruby precedent). Bump crate + wasm together.
|
|
tree-sitter-r = "=1.2.0"
|
|
# luau: the vendored wasm IS tree-sitter-grammars v1.2.0 and this crate's
|
|
# tarball ships parser.c/scanner.c sha-identical to the tag (8f25bc17… /
|
|
# a157bb52… — lua-luau checklist §Grammar prep). ABI 14. Lua itself is
|
|
# vendored C (build.rs) — its v0.4.1 revision is not on crates.io.
|
|
tree-sitter-luau = "=1.2.0"
|
|
|
|
# tree-sitter-language: the version-agnostic LanguageFn shim for the vendored
|
|
# kotlin grammar C (see build.rs — no kotlin crate dep is possible).
|
|
tree-sitter-language = "0.1"
|
|
|
|
# Stack-bounds queries for the walker stack guard (src/stack.rs, #1581):
|
|
# pthread_getattr_np / pthread_get_stackaddr_np. Already in the lock file
|
|
# transitively; Windows uses a hand-declared kernel32 extern instead.
|
|
[target.'cfg(unix)'.dependencies]
|
|
libc = "0.2"
|
|
|
|
[build-dependencies]
|
|
napi-build = "2"
|
|
cc = "1"
|
|
|
|
[profile.release]
|
|
lto = true
|
|
codegen-units = 1
|
|
strip = "symbols"
|