Compare commits
| Author | SHA1 | Date | |
|---|---|---|---|
 Kaoru0822andClaude Opus 5.5
|
733c160d4f |
No files matched your search
@@ -4,7 +4,6 @@ import argparse
|
||||
import hashlib
|
||||
import json
|
||||
import os
|
||||
import re
|
||||
from pathlib import Path
|
||||
import shutil
|
||||
import signal
|
||||
@@ -23,31 +22,7 @@ def file_digest(path):
|
||||
return digest.hexdigest()
|
||||
|
||||
|
||||
def pinned_experimental_engine(settings):
|
||||
expected = settings.get('REA_ATOMIC_NATIVE_IR_ENGINE_SHA256')
|
||||
if expected is None:
|
||||
return None
|
||||
if not re.fullmatch(r'[0-9a-f]{64}', expected):
|
||||
raise ValueError('Experimental native engine pin must be a SHA-256')
|
||||
path = Path(settings.get('REA_KUNA_COMMAND', ''))
|
||||
if not path.is_absolute() or not path.is_file() or not os.access(path, os.X_OK):
|
||||
raise ValueError('Pinned experimental engine requires an absolute executable file')
|
||||
actual = file_digest(path)
|
||||
if actual != expected:
|
||||
raise ValueError('Experimental engine digest differs from its pin; use a preserved build artifact')
|
||||
return {'path': str(path), 'sha256': actual}
|
||||
|
||||
|
||||
def experimental_engine_unchanged(receipt):
|
||||
if receipt is None:
|
||||
return None
|
||||
try:
|
||||
return file_digest(Path(receipt['path'])) == receipt['sha256']
|
||||
except OSError:
|
||||
return False
|
||||
|
||||
|
||||
def runtime_fingerprint(root, entrypoint=None):
|
||||
def runtime_fingerprint(root):
|
||||
files = []
|
||||
for name in ['dist', 'bridge']:
|
||||
for path in sorted((root / name).rglob('*')):
|
||||
@@ -55,18 +30,6 @@ def runtime_fingerprint(root, entrypoint=None):
|
||||
raise ValueError('Runtime fingerprint does not follow symlinks')
|
||||
if path.is_file():
|
||||
files.append({'path': path.relative_to(root).as_posix(), 'sha256': file_digest(path)})
|
||||
if entrypoint is not None:
|
||||
# Include the experimental modules and their installed QuickJS JS/WASM,
|
||||
# plus the shared fact owner; the entrypoint alone does not pin them.
|
||||
for directory in [entrypoint.parent, root / 'experiments/atomic-re-tools']:
|
||||
for path in sorted(directory.rglob('*')):
|
||||
if path.suffix not in {'.mjs', '.cjs', '.js', '.wasm'} and path.name not in {'package.json', 'package-lock.json'}:
|
||||
continue
|
||||
if path.is_symlink():
|
||||
raise ValueError('Experimental runtime fingerprint does not follow symlinks')
|
||||
if path.is_file():
|
||||
label = path.relative_to(root).as_posix() if path.is_relative_to(root) else path.as_posix()
|
||||
files.append({'path': label, 'sha256': file_digest(path)})
|
||||
files.sort(key=lambda row: row['path'])
|
||||
text = json.dumps(files, separators=(',', ':'), ensure_ascii=False)
|
||||
return files, hashlib.sha256(text.encode()).hexdigest()
|
||||
@@ -76,12 +39,6 @@ def main():
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument('--mode', choices=['direct', 'client-code'], default='direct')
|
||||
parser.add_argument('--runtime-root', type=Path, required=True)
|
||||
parser.add_argument('--mcp-entrypoint', type=Path,
|
||||
help='Caller-selected experimental stdio MCP entrypoint; leaves production dispatcher unchanged')
|
||||
parser.add_argument('--server-name', default='rea')
|
||||
parser.add_argument('--protocol-profile', choices=['canonical', 'atomic', 'atomic-direct'], default='canonical')
|
||||
parser.add_argument('--profile-parent', type=Path,
|
||||
help='Absolute parent outside the repository for the isolated model workspace')
|
||||
parser.add_argument('--revision', required=True)
|
||||
parser.add_argument('--prompt', type=Path, required=True)
|
||||
parser.add_argument('--artifact-path', help='Replace {{artifact_path}} in a question template')
|
||||
@@ -97,10 +54,6 @@ def main():
|
||||
help='Retain owned tool/usage items; excludes account/ambient state')
|
||||
parser.add_argument('--guide-enabled-tools', action='store_true',
|
||||
help='Opt-in exact-name discovery guidance for selected REA tools')
|
||||
parser.add_argument('--normalize-mcp-guidance', action='store_true',
|
||||
help='Opt-in common envelope/stopping guidance, identical across protocol lanes')
|
||||
parser.add_argument('--guide-long-waits', action='store_true',
|
||||
help='Opt-in common client wait guidance; does not change analysis contracts or deadlines')
|
||||
args = parser.parse_args()
|
||||
if args.guide_enabled_tools and args.enabled_tools is None:
|
||||
parser.error('--guide-enabled-tools requires --enabled-tools')
|
||||
@@ -141,19 +94,13 @@ def main():
|
||||
if undeclared:
|
||||
raise ValueError('Declare inherited provider settings explicitly in --providers: '
|
||||
+ ', '.join(undeclared) + '. Values are not logged.')
|
||||
# Fail before copying credentials or spending model tokens. Cargo test/build
|
||||
# can replace a mutable target/release path with a different feature build.
|
||||
experimental_engine = pinned_experimental_engine(provider_env)
|
||||
source_auth_home = (Path(os.environ['CODEX_HOME']) if 'CODEX_HOME' in os.environ
|
||||
else Path.home() / '.codex')
|
||||
source_auth = source_auth_home / 'auth.json'
|
||||
args.output.mkdir(parents=True, exist_ok=False, mode=0o700)
|
||||
runtime_files, runtime_digest = runtime_fingerprint(repository, args.mcp_entrypoint)
|
||||
runtime_files, runtime_digest = runtime_fingerprint(repository)
|
||||
(args.output / 'runtime-files.json').write_text(json.dumps(runtime_files))
|
||||
if args.profile_parent and (not args.profile_parent.is_absolute()
|
||||
or args.profile_parent.resolve().is_relative_to(repository.resolve())):
|
||||
raise ValueError('profile-parent must be an absolute directory outside runtime-root')
|
||||
profile = Path(tempfile.mkdtemp(prefix='rea-model-profile-', dir=args.profile_parent or args.output))
|
||||
profile = Path(tempfile.mkdtemp(prefix='profile-', dir=args.output))
|
||||
trial_codex_home = profile / 'codex'
|
||||
account = profile / 'account'
|
||||
workspace = profile / 'workspace'
|
||||
@@ -172,12 +119,7 @@ def main():
|
||||
'NODE_OPTIONS': '--max-old-space-size=512',
|
||||
})
|
||||
target_digest = file_digest(Path(args.artifact_path)) if args.artifact_path else None
|
||||
if not args.server_name.replace('-', '').replace('_', '').isalnum():
|
||||
raise ValueError('server-name must contain only letters, digits, hyphens or underscores')
|
||||
if args.mcp_entrypoint and not args.mcp_entrypoint.is_absolute():
|
||||
raise ValueError('mcp-entrypoint must be absolute')
|
||||
command_args = ([str(args.mcp_entrypoint)] if args.mcp_entrypoint
|
||||
else [str(repository / 'scripts/rea.mjs'), 'mcp'])
|
||||
command_args = [str(repository / 'scripts/rea.mjs'), 'mcp']
|
||||
instruction = (
|
||||
'Use the registered REA MCP server for this investigation. Static analysis '
|
||||
'only: do not execute target code, fetch URLs, launch a GUI, or modify targets. '
|
||||
@@ -190,23 +132,6 @@ def main():
|
||||
'Prefer summary/focused views where appropriate. Do not invent successful facts '
|
||||
'after a tool error. No other agents. Stop once the requested facts are supported.'
|
||||
)
|
||||
if args.protocol_profile in {'atomic', 'atomic-direct'}:
|
||||
instruction = (
|
||||
'Use the registered ' + args.server_name + ' MCP server for this static binary investigation. '
|
||||
'Call reference to inspect API contracts, then compose the supported analysis calls. '
|
||||
'Do not execute target code, fetch URLs, launch a GUI, modify targets, use shell/file tools, '
|
||||
'or use other agents. API methods return {ok,context,data} or {ok:false,error}; '
|
||||
'execute_javascript returns the script value, with script-derived authority. '
|
||||
'Preserve real evidence_id values, artifact digest, uncertainty and traversal boundaries. '
|
||||
'Check failures before reading data. Do not infer absence from partial coverage. '
|
||||
'Do not return complete inventories or source reports when a focused answer suffices. '
|
||||
'Close the binary when finished. No precomputed answer is available to you.'
|
||||
)
|
||||
if args.protocol_profile == 'atomic-direct':
|
||||
instruction = instruction.replace(
|
||||
'execute_javascript returns the script value, with script-derived authority. ',
|
||||
'Call the registered facts_* tools directly; no JavaScript execution tool is registered. '
|
||||
)
|
||||
if args.authority == 'browser':
|
||||
instruction = (
|
||||
'Use the registered REA tools for the explicitly selected public browser scenario. '
|
||||
@@ -222,13 +147,6 @@ def main():
|
||||
'possible and select the requested normalized_result fields before returning '
|
||||
'intermediate output to model context. Keep real Evidence IDs and limitations.'
|
||||
)
|
||||
if args.normalize_mcp_guidance:
|
||||
instruction += (
|
||||
' MCP calls return transport envelopes: prefer structuredContent when present; '
|
||||
'otherwise parse the text content JSON. Do not print both duplicate encodings. '
|
||||
'Apply the same rule to every tool. Stop once the requested facts are supported, '
|
||||
'state missing coverage explicitly, and close the binary when finished.'
|
||||
)
|
||||
if args.guide_enabled_tools:
|
||||
instruction += (
|
||||
' Configured REA tool function names for this trial: '
|
||||
@@ -237,15 +155,6 @@ def main():
|
||||
'return only their metadata. Preserve the original analyst question, '
|
||||
'dependency order and source Evidence.'
|
||||
)
|
||||
if args.guide_long_waits:
|
||||
instruction += (
|
||||
' When the client executor reports a running cell during a long analysis, '
|
||||
'resume that same cell with wait and yield_time_ms:30000, within the '
|
||||
'advertised tool schema, rather than repeated one-second polls. '
|
||||
'The wait may return early when the result is ready. Do not sleep or '
|
||||
'start duplicate analysis calls just to check completion. Keep the '
|
||||
'original analysis dependencies, evidence and stopping requirements.'
|
||||
)
|
||||
quote = json.dumps
|
||||
config = '\n'.join([
|
||||
'approval_policy = "never"', 'sandbox_mode = "read-only"',
|
||||
@@ -257,16 +166,16 @@ def main():
|
||||
'code_mode = ' + ('true' if args.mode == 'client-code' else 'false'),
|
||||
'code_mode_only = ' + ('true' if args.mode == 'client-code' else 'false'),
|
||||
'[history]', 'persistence = "none"',
|
||||
'[mcp_servers.' + args.server_name + ']', 'command = ' + quote(node),
|
||||
'[mcp_servers.rea]', 'command = ' + quote(node),
|
||||
'args = ' + quote(command_args), 'startup_timeout_sec = 60',
|
||||
'tool_timeout_sec = 600',
|
||||
*([] if enabled_tools is None else ['enabled_tools = ' + quote(enabled_tools)]),
|
||||
'[mcp_servers.' + args.server_name + '.env]',
|
||||
'[mcp_servers.rea.env]',
|
||||
*[key + ' = ' + quote(value) for key, value in provider_env.items()],
|
||||
]) + '\n'
|
||||
# MCP read-only hints cover normal static tools. Explicitly authorize only
|
||||
# the session-opening primitive needed by the requested native tasks.
|
||||
config += '[mcp_servers.' + args.server_name + '.tools.open_binary]\napproval_mode = "approve"\n'
|
||||
config += '[mcp_servers.rea.tools.open_binary]\napproval_mode = "approve"\n'
|
||||
for tool in ['analyze_javascript_application', 'inspect_analysis_view',
|
||||
'inspect_android_package', 'search_android_classes',
|
||||
'inspect_android_class', 'inspect_android_method',
|
||||
@@ -280,7 +189,7 @@ def main():
|
||||
'inspect_evm_interface', 'get_evidence_bundle', 'export_evidence_bundle',
|
||||
'capture_browser_scenario',
|
||||
'resolve_containing_procedure', 'analyze_function', 'close_binary']:
|
||||
config += '[mcp_servers.' + args.server_name + '.tools.' + tool + ']\napproval_mode = "approve"\n'
|
||||
config += '[mcp_servers.rea.tools.' + tool + ']\napproval_mode = "approve"\n'
|
||||
try:
|
||||
if source_auth.exists():
|
||||
shutil.copyfile(source_auth, auth)
|
||||
@@ -300,17 +209,10 @@ def main():
|
||||
'mode': args.mode, 'model': args.model, 'reasoning': args.reasoning,
|
||||
'capture_tool_context': args.capture_tool_context,
|
||||
'guide_enabled_tools': args.guide_enabled_tools,
|
||||
'normalize_mcp_guidance': args.normalize_mcp_guidance,
|
||||
'guide_long_waits': args.guide_long_waits,
|
||||
'driver_commit': subprocess.check_output(
|
||||
['git', 'rev-parse', 'HEAD'], cwd=driver_repository, text=True).strip(),
|
||||
'runtime_revision': args.revision,
|
||||
'runtime_root': str(repository),
|
||||
'mcp_server_name': args.server_name,
|
||||
'protocol_profile': args.protocol_profile,
|
||||
'isolated_profile_parent': str(args.profile_parent) if args.profile_parent else str(args.output),
|
||||
'mcp_entrypoint': str(args.mcp_entrypoint) if args.mcp_entrypoint else None,
|
||||
'mcp_entrypoint_sha256': file_digest(args.mcp_entrypoint) if args.mcp_entrypoint else None,
|
||||
'selected_artifact_path': args.artifact_path,
|
||||
'selected_artifact_sha256_before': target_digest,
|
||||
'enabled_tools': enabled_tools,
|
||||
@@ -328,7 +230,6 @@ def main():
|
||||
'question_template_sha256': template_digest,
|
||||
'provider_settings_sha256': hashlib.sha256(
|
||||
json.dumps(provider_env, sort_keys=True).encode()).hexdigest(),
|
||||
'experimental_engine': experimental_engine,
|
||||
'provider_search_path_sha256': hashlib.sha256(
|
||||
provider_env.get('PATH', os.environ.get('PATH', '')).encode()).hexdigest(),
|
||||
'provider_environment_policy_sha256': hashlib.sha256(
|
||||
@@ -381,9 +282,6 @@ def main():
|
||||
summary = {
|
||||
'mode': args.mode, 'requested_model': args.model, 'reasoning': args.reasoning,
|
||||
'runtime_revision': args.revision,
|
||||
'runtime_inputs_unchanged': runtime_fingerprint(repository, args.mcp_entrypoint)[1] == runtime_digest,
|
||||
'experimental_engine': experimental_engine,
|
||||
'experimental_engine_unchanged': experimental_engine_unchanged(experimental_engine),
|
||||
'capture_tool_context_requested': args.capture_tool_context,
|
||||
'selected_artifact_sha256_before': target_digest,
|
||||
'selected_artifact_sha256_after': final_target_digest,
|
||||
@@ -398,10 +296,6 @@ def main():
|
||||
}
|
||||
(args.output / 'summary.json').write_text(json.dumps(summary, indent=2) + '\n')
|
||||
print(json.dumps(summary))
|
||||
if not summary['runtime_inputs_unchanged']:
|
||||
raise RuntimeError('Runtime inputs changed during the trial; retain costs but reject the comparison')
|
||||
if summary['experimental_engine_unchanged'] is False:
|
||||
raise RuntimeError('Experimental engine changed during the trial; retain costs but reject the comparison')
|
||||
finally:
|
||||
try:
|
||||
if args.capture_tool_context:
|
||||
|
||||
@@ -1,46 +0,0 @@
|
||||
import hashlib
|
||||
import importlib.util
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
|
||||
spec = importlib.util.spec_from_file_location('agent_trial', Path(__file__).with_name('agent-trial.py'))
|
||||
trial = importlib.util.module_from_spec(spec)
|
||||
spec.loader.exec_module(trial)
|
||||
|
||||
|
||||
class ExperimentalEnginePinTests(unittest.TestCase):
|
||||
def test_missing_pin_preserves_other_provider_trials(self):
|
||||
self.assertIsNone(trial.pinned_experimental_engine({}))
|
||||
self.assertIsNone(trial.experimental_engine_unchanged(None))
|
||||
|
||||
def test_feature_build_replacement_and_removal_invalidate_pinned_engine(self):
|
||||
with tempfile.TemporaryDirectory() as directory:
|
||||
engine = Path(directory) / 'kuna'
|
||||
engine.write_bytes(b'embedded-spec-build')
|
||||
engine.chmod(0o755)
|
||||
settings = {'REA_KUNA_COMMAND': str(engine),
|
||||
'REA_ATOMIC_NATIVE_IR_ENGINE_SHA256': hashlib.sha256(engine.read_bytes()).hexdigest()}
|
||||
receipt = trial.pinned_experimental_engine(settings)
|
||||
self.assertTrue(trial.experimental_engine_unchanged(receipt))
|
||||
engine.write_bytes(b'non-embedded-test-build')
|
||||
with self.assertRaisesRegex(ValueError, 'differs from its pin'):
|
||||
trial.pinned_experimental_engine(settings)
|
||||
self.assertFalse(trial.experimental_engine_unchanged(receipt))
|
||||
engine.unlink()
|
||||
self.assertFalse(trial.experimental_engine_unchanged(receipt))
|
||||
with self.assertRaisesRegex(ValueError, 'absolute executable'):
|
||||
trial.pinned_experimental_engine(settings)
|
||||
|
||||
def test_malformed_pin_and_relative_command_are_rejected(self):
|
||||
for settings in [
|
||||
{'REA_ATOMIC_NATIVE_IR_ENGINE_SHA256': 'bad'},
|
||||
{'REA_ATOMIC_NATIVE_IR_ENGINE_SHA256': '0' * 64, 'REA_KUNA_COMMAND': './kuna'},
|
||||
]:
|
||||
with self.assertRaises(ValueError):
|
||||
trial.pinned_experimental_engine(settings)
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
unittest.main()
|
||||
@@ -1,301 +0,0 @@
|
||||
# Atomic RE through Code Mode
|
||||
|
||||
This is an **in-progress experiment**, not an integrated provider release. Branch
|
||||
`experiment/atomic-codemode-20261016` starts at main
|
||||
`42fc6c53ae188978deb3bedfe88287b6b316ba22` and carries the previous atomic-facts
|
||||
prototype. Historical measurements in `../atomic-re-tools/` retain their original
|
||||
baselines. No PR or merge is intended.
|
||||
|
||||
The goal is a real model completing reverse-engineering questions through
|
||||
composable facts, with Code Mode as the calling adapter. A configured feature flag,
|
||||
a scripted verifier or a small response does not prove that goal.
|
||||
|
||||
The reference is [Hex-Rays IDA MCP](https://github.com/HexRaysSA/ida-mcp/blob/df6de7985b534a964c125d5a86e41b26b17b14f6/ida_mcp/mcp.py):
|
||||
an on-demand `reference` lookup and an execution entrypoint over a richer analysis
|
||||
API. This experiment uses JavaScript and a QuickJS WASM environment rather than
|
||||
Python in the host process. It exposes only declared static analysis capabilities.
|
||||
It does not add target execution, patching or GUI control.
|
||||
|
||||
Three boundaries own the implementation:
|
||||
|
||||
- `provider.mjs`, `native-ir.mjs`: existing REA native target/process owners and parsed Kuna
|
||||
string/reference/SSA facts, retained sources, freshness and cleanup.
|
||||
- `api.mjs`: named, validated fact/relationship contracts and session admission.
|
||||
SDK callers, CLI scripts and both MCP adapters use these same methods.
|
||||
- `executor.mjs`, `server.mjs`, `cli.mjs`: bounded JavaScript composition and
|
||||
presentation. No analysis is reimplemented in the execution environment.
|
||||
|
||||
The complete Code Mode MCP catalog is `open_binary`, `reference`,
|
||||
`execute_javascript`, `close_binary`. Setting `REA_ATOMIC_ADAPTER=direct` exposes
|
||||
the same fact API as `facts_*` MCP operations for comparison. This is a distinct
|
||||
experimental server; the production REA catalog is not filtered or replaced.
|
||||
|
||||
Use the pinned Node, an existing REA build and a caller-supplied Kuna 1.833:
|
||||
|
||||
```sh
|
||||
npm ci --prefix experiments/atomic-codemode --ignore-scripts
|
||||
node --test experiments/atomic-codemode/*.test.mjs
|
||||
REA_KUNA_COMMAND=/absolute/path/to/kuna node experiments/atomic-codemode/server.mjs
|
||||
```
|
||||
|
||||
The local MCP registration for real model trials is named `rea-codemode`.
|
||||
`agent-trial.py --mcp-entrypoint /absolute/path/to/server.mjs
|
||||
--server-name rea-codemode --protocol-profile atomic --profile-parent /tmp` creates an isolated
|
||||
configuration, home and empty working directory for the model. Shell/file tools,
|
||||
web search and delegation are disabled. No repository or oracle is placed in that
|
||||
workspace. The copied authentication file is removed with the isolated profile.
|
||||
Model output, actual MCP calls, original errors and usage remain in the trial
|
||||
directory. `REA_ATOMIC_TRACE` and `REA_ATOMIC_EVIDENCE_DIR` retain actual API calls
|
||||
and complete sources outside the model workspace when explicitly selected by the
|
||||
evaluator. The ambient environment and credentials are not logged.
|
||||
|
||||
Current bounds: 64 MiB guest heap, 32 KiB code and script value/error payloads, 64 host calls per execution,
|
||||
120 seconds wall deadline and 2 seconds of guest synchronous execution; native requests have existing supervision, an 8 MiB
|
||||
capture budget and 1 GiB function-memory limit. Retained source storage is 64 MiB
|
||||
per target, with two resident indexes (each derived from at most 2 MiB diagnostic
|
||||
or 8 MiB native wire source). Compressed native graphs admit at most 32 MiB decoded
|
||||
JSON plus independent row/text/relationship bounds. One decoded Evidence text view
|
||||
up to 32 MiB is also retained per lease; these source bounds are not parsed-heap
|
||||
size guarantees. Sources are
|
||||
content-digested and retained until close; explicit evaluator copies survive
|
||||
close. Capacity failures are explicit, not clipped facts. Leases and snapshot
|
||||
handles prevent cross-session joins.
|
||||
|
||||
Invocation context and observed source IDs are additional bounded receipt
|
||||
metadata; the script payload bound is not a 32 KiB whole-MCP-wire limit. A real
|
||||
MCP probe found that a 40,000-byte guest exception bypassed that payload policy
|
||||
and produced a 40,183-byte response. Oversized errors now return explicit
|
||||
`output_budget` with omission metadata. Escaping expansion is checked too;
|
||||
ordinary smaller errors remain intact. `verify-executor.mjs` confirms real MCP
|
||||
delivery and continued access to source Evidence collected before the exception.
|
||||
It measures response bytes, not model tokens or an analysis speedup. Targeted JS
|
||||
tests at this stage totaled 42, alongside the separate real-provider and native-writer checks.
|
||||
|
||||
The diagnostic SSA format still has incomplete CFG and phi predecessor coverage.
|
||||
The opt-in [native recovered IR experiment](native-engine.md) adds full recovered
|
||||
CFG edges, exact value/operand identities, phi incoming-edge bindings, selected
|
||||
`values`/`blocks` reads and derived control dependence. It requires an explicitly
|
||||
pinned patched engine; stock Kuna does not gain those capabilities from a wrapper.
|
||||
Runtime successors, memory aliases and machine-semantic equivalence remain
|
||||
unknown. Other providers are not yet wired to this experimental service.
|
||||
|
||||
The initial [actual model comparison](trials.md) favors the focused traditional
|
||||
workflow. The seeded value-origin task has lower cumulative input and wall time
|
||||
through Code Mode, but initially exposes an IR-to-machine-path error. The revised
|
||||
trial correctly explains that machine path with lower observed task costs;
|
||||
broader claims still require more than this single seeded task.
|
||||
|
||||
The large SQLite VM diagnostic task also completes when its native graph exceeds
|
||||
capacity. Atomic direct beats Code Mode on cumulative input and wall time in
|
||||
that sample, and a traditional reused-project run is fastest. The actual CLI
|
||||
client already exposes JavaScript dispatch in every lane, regardless of requested
|
||||
feature flags. MCP request counts therefore do not establish fewer model rounds;
|
||||
the ledger now records actual model responses and client dispatch/wait calls.
|
||||
|
||||
`instructions` reads selected machine byte/decode windows through the same
|
||||
owner. Recovered operations use `origin_address` and blocks `origin_start`;
|
||||
these locations cannot be substituted for machine branch destinations. The
|
||||
value-origin trial demonstrates why this boundary matters. The experiment
|
||||
now has fourteen fact methods, still behind the same four-tool Code Mode
|
||||
catalog.
|
||||
|
||||
`calls` connects recovered call operations to native target specifications and
|
||||
separates argument SSA values from passing storage. `values(is_input:true)`
|
||||
selects retained callee inputs for exact storage comparisons. These candidate
|
||||
links preserve snapshot identities, missing specifications, unknown storage and
|
||||
runtime-binding limits. The [cross-function task and oracle](call-oracle.md)
|
||||
are fixed before its real-model trials; the graph wire format is version 3, with independent version-1 call facts.
|
||||
|
||||
`references` now selects incoming exact-address references or outgoing
|
||||
instruction/function references, with native kind filtering, field selection and
|
||||
paging. Source Evidence plus `source_row` identifies an occurrence. Function
|
||||
applicability, excluded/unattributed rows, indirect targets and discovery stay
|
||||
explicit. `fields:[]` requests only the always-retained identity columns across
|
||||
projection contracts. The [reference question and independent oracle](reference-oracle.md)
|
||||
cover real SQLite VM aggregates and a machine call omitted from optimized IR.
|
||||
|
||||
`bytes` reads explicit virtual-address ranges as native loaded-image hex without
|
||||
decoding them as instructions. Batched reads preserve short reads and individual
|
||||
failures, with 64 KiB aggregate selected-byte capacity to bound hex/bridge growth.
|
||||
Complete hexdumps remain in Evidence. The original-file-offset mapping and live
|
||||
runtime memory remain unknown. This closes a direct byte-reading contract gap
|
||||
for data tables; it does not establish arbitrary memory-alias analysis.
|
||||
|
||||
`functions` provides selected name/alias/entry reads. Native snapshots report
|
||||
code and recovered-IR facets separately, preserving readable code when graph
|
||||
extraction is unavailable. Real capacity and cold/warm MCP checks are documented
|
||||
in [native-engine.md](native-engine.md). Whole-model costs and unequal-depth
|
||||
comparisons remain separate in the trial report.
|
||||
|
||||
The corrected SQLite comparison reaches the formatter implementation in all
|
||||
three interfaces. Code Mode uses fewer calls than atomic direct in that sample;
|
||||
a traditional run reusing its persistent whole-project cache is faster than
|
||||
Code Mode's fresh lease. Cache state, completed-answer coverage and actual
|
||||
uncached tokens must therefore accompany any cost comparison.
|
||||
|
||||
SDK views share immutable parsed inventories with their owner. A real SDK probe
|
||||
initially changed a returned string and made the next query disappear under the
|
||||
same Evidence ID. Freezing at the cache boundary fixes that producing rule,
|
||||
without copying or parsing the entire inventory for each read. `verify-sdk.mjs`
|
||||
checks the real provider's string/metadata/function views and complete retained
|
||||
source digest. The fixed probe passes under the bounded runner (1.328 seconds,
|
||||
1.144 family CPU seconds, 110.6 MiB kernel family peak, zero OOM kills and verified
|
||||
payload cleanup); it is a correctness check, not a model-cost measurement.
|
||||
|
||||
Encoded size is another real coverage boundary: two SQLite functions initially
|
||||
lose even their C because pretty JSON exceeds transport capacity. The native
|
||||
patch now uses bounded compact serialization and a graph-only failure when
|
||||
needed. Parser's full graph fits; Pragma retains identical C with an explicit
|
||||
missing graph. The 8 MiB bound remains. See the exact sizes, rebuilt engine digest
|
||||
and verification scope in [native-engine.md](native-engine.md).
|
||||
|
||||
Completion requires actual model Code Mode use, independently checked task
|
||||
answers, shared atomic/direct/Code Mode semantics, needed relation coverage,
|
||||
failure/cancellation/identity tests, and equivalent completed-task measurements
|
||||
including discovery/retries, actual model usage, cold/warm analysis, CPU and
|
||||
process-family peak memory. Negative cases such as the small SQLite function
|
||||
must remain in the comparison.
|
||||
|
||||
The outgoing-reference real-model pair now completes with identical required
|
||||
facts and explicit coverage limits. In its single sample, Code Mode uses 264,680
|
||||
cumulative input / 93.837 seconds versus direct's 285,086 / 106.183. The shared
|
||||
CLI/API also verifies the VM aggregates on unpatched stock Kuna 1.833. These are
|
||||
scoped positive results alongside the earlier mixed adapter comparisons, not
|
||||
proof of general superiority. Details and retained failed/partial attempts are
|
||||
in [trials.md](trials.md).
|
||||
|
||||
Recovered calls and their boundary values now survive full-graph capacity
|
||||
failure. The real VM capture preserves 416 call operations and 1,294 boundary
|
||||
values while its graph remains unavailable. Values clearly state their partial
|
||||
scope; definition IDs do not imply readable defining operations. See the
|
||||
[boundary question and oracle](boundary-oracle.md) and its completed real-model
|
||||
pair in [trials.md](trials.md). Both answers pass, using one VM capture each.
|
||||
Code Mode's single-sample latency is lower, with mixed context costs; broader
|
||||
efficiency remains unproven. The focused JS suite at that stage passes 49 tests.
|
||||
|
||||
Available snapshot call facets now include `by_opcode` known/unknown-target
|
||||
counts and their own coverage. The owner derives them once from the same
|
||||
validated immutable calls used by selected reads, including dead operations;
|
||||
target knowledge is independent of opcode. Empty available facets report zeros,
|
||||
while unavailable facets omit counts. This small aggregate addresses an observed
|
||||
cost in the real boundary pair: paging the entire call set just to count it.
|
||||
Both subsequent real-model answers satisfy the unchanged question/oracle without
|
||||
that paging. In the new sample direct is faster and lower on cumulative/uncached
|
||||
input than Code Mode, reversing the preceding latency ordering. Whole-task
|
||||
effects remain mixed; see [the complete results](trials.md). This aggregate is a
|
||||
shared fact-layer improvement, available to either adapter.
|
||||
|
||||
The shared owner also checks the native cleanup `Result` before releasing its
|
||||
lease or clearing retained state. A tagged cleanup failure now stays retryable;
|
||||
MCP errors retain the lease context and reported resources. The real SDK
|
||||
failure-injection verifier first reproduces the old false success, then verifies
|
||||
both close and target replacement retain the staged file/owner until a successful
|
||||
retry. Both public MCP adapters preserve that error and retry successfully.
|
||||
These use an injected tagged failure, not an observed engine cleanup incident;
|
||||
native reads and subsequent cleanup use the real provider. See the recorded
|
||||
scope and resource receipts in [native-engine.md](native-engine.md).
|
||||
|
||||
A [subsequent diagnostic probe](projection-experiment.md) measures the VM's full
|
||||
17 MB recovered graph and much smaller selected definition neighborhoods. It
|
||||
supports trying a bounded operation/value facet, while preserving unresolved
|
||||
definitions and unavailable CFG. The following implementation now makes that
|
||||
facet callable; the original diagnostic remains separate from public results.
|
||||
|
||||
`snapshot.facets.operations` distinguishes a complete recovered graph from a
|
||||
bounded call-rooted definition projection. `operations`, `values`, selected
|
||||
definition/use `relations` and `slice` share the same parsed operation owner in
|
||||
either case. Projection coverage preserves unexpanded definition counts and
|
||||
selected slice frontier reasons; CFG/controller reads remain unavailable.
|
||||
Native operation reads retain `is_dead` and literal operand/value identities.
|
||||
Coverage returns compact counts and bounded issue text, with complete native
|
||||
details retained in Evidence. [The new question/oracle](projection-oracle.md)
|
||||
are fixed before real-model trials of the PHI format alternatives.
|
||||
|
||||
Both first PHI-format model answers pass the full oracle using one VM analysis
|
||||
and a six-operation selected trace, with the global graph/use limits intact.
|
||||
The trace reveals two format sources at each shared call, verified against byte
|
||||
and machine evidence. [Costs and source review](trials.md) also record a direct
|
||||
documentation-lookup miss; named discovery needs fixing before rerunning the
|
||||
same task. No general adapter speedup follows from that first pair.
|
||||
|
||||
Named reference lookup now accepts canonical IDs or exact displayed names,
|
||||
including space/comma-separated lists. One presentation map supplies direct
|
||||
registration and documentation names; unmatched terms remain explicit. Selected
|
||||
schemas come from named contracts, use input semantics (defaulted fields stay
|
||||
optional), and are cached immutably by Zod schema identity for the process.
|
||||
Validation stays in the API registry; documentation has its own module. Both
|
||||
actual MCP catalogs and all fourteen reference schemas validate. The unchanged
|
||||
PHI question/oracle were rerun: both answers pass again, but direct has a native
|
||||
timeout followed by a successful retry. Complete costs and the unknown timeout
|
||||
cause remain in [trials.md](trials.md); this does not isolate an adapter speedup.
|
||||
|
||||
The next observed atom boundary is a wide PHI. One real VM operation has 156
|
||||
inputs; its old default single-node slice is about 44 KB and exceeds the guest's
|
||||
32 KiB result budget. Native operation/slice defaults now return opcode, output
|
||||
and exact input/definition/constant counts. Explicit detailed fields still work.
|
||||
`relations({kind:"operands",seeds:[id],input_indices:[0,105,155]})` reads selected
|
||||
slots; omitting the selector pages all slots. Each slot retains value/definition
|
||||
and phi predecessor/edge identity, including null values and definition-free
|
||||
constants/inputs. Counts concern recovered facts, not source function arity.
|
||||
|
||||
Full and projected native operations use the same owner. Relationship paging
|
||||
counts existing groups before materializing only the requested page; identity
|
||||
reads deduplicate requested IDs in caller order. Diagnostic SSA explicitly lacks
|
||||
native operand relations. The discriminated relation schema advertises the same
|
||||
seed types and selector applicability enforced at runtime. The fourteen-method
|
||||
API and both adapters remain shared; no new engine or output-budget increase is
|
||||
needed. The [seeded wide-operand task/oracle](operands-oracle.md) is separate from
|
||||
the earlier PHI-format question and is pinned before model trials.
|
||||
|
||||
Both actual wide-operand model answers now pass that fixed oracle, including
|
||||
distinct edges sharing a value and absent-slot versus unretained-seed semantics.
|
||||
Each performs one native analysis and reads only selected slots, distinct values
|
||||
and their immediate definitions. Code Mode takes 95.699 seconds versus direct's
|
||||
91.817, with nearly equal cumulative input and lower uncached input. Its four
|
||||
discovery lookups prevent fewer MCP requests from becoming fewer model responses.
|
||||
These [mixed costs](trials.md#selected-wide-phi-input-relationships) validate the
|
||||
fact boundary while leaving broad efficiency unproven. The current shared JS
|
||||
suite passes 57 tests; the Python cost-ledger suite passes three.
|
||||
|
||||
Discovery now also offers `reference({view:"index"})`: the complete fourteen
|
||||
method names and descriptions, without schemas or return contracts. Follow with
|
||||
`view:"contracts"` and exact names for only the needed inputs. The default
|
||||
contract view and named/topic lookup stay compatible. Both views use the same
|
||||
registry and presentation map; index access never constructs unrequested schemas.
|
||||
The index describes declared methods, while `snapshot` remains the live coverage
|
||||
authority. This addresses observed blind discovery without inventing aliases or
|
||||
changing analysis facts; its whole-task effect must be measured on the unchanged
|
||||
wide-operand question.
|
||||
|
||||
The identical question now completes correctly through both indexed adapters.
|
||||
Code Mode uses the intended index plus one named lookup; discovery drops from
|
||||
four calls to two, without changing the seven fact reads. The new pair takes
|
||||
about 98.8 seconds each; cumulative input favors direct, uncached input favors
|
||||
Code Mode. All four wide-operand answers pass, with no retry or OOM. Costs remain
|
||||
mixed. A selected-slot read still incurs whole-function native decompilation on
|
||||
a fresh lease; phase profiling is the next computation question, separate from
|
||||
the now finer fact/query boundary.
|
||||
|
||||
The [native phase study](native-compute-study.md) now measures the VM drive at
|
||||
about 34 seconds, C rendering at 0.22 seconds and projection at 0.03 seconds.
|
||||
Repeated new native processes do not gain the API owner's warm result reuse.
|
||||
Its subsequent [bounded codec implementation](native-graph-transport-study.md)
|
||||
now returns the complete 21,349-operation VM graph in about 3.79 MB, inside the
|
||||
unchanged 8 MiB wire budget. Selected definitions/CFG reads and decoded Evidence
|
||||
work through both real MCP adapters; all graph fields and ordinary C/calls match
|
||||
the prior complete golden. Same-lease reuse is about 67 ms after a 39.7-second
|
||||
fresh capture, with resource limits and no OOM. The [next question/oracle](savepoint-control-oracle.md)
|
||||
start from names/error messages and exercise the new condition/CFG coverage.
|
||||
Two [real-model pairs](savepoint-control-study.md), including a discovery-guided
|
||||
repeat, answer both local guards correctly without a whole-report fetch. Code
|
||||
Mode uses fewer MCP requests, but whole-task costs are mixed and a large capture
|
||||
time variation remains unexplained. Both clients already provide JavaScript
|
||||
orchestration. Source fidelity is not machine equivalence, provider parity or a
|
||||
general speedup. The [native switch experiment](switch-study.md) now exports and independently checks
|
||||
182 labels over those 149 targets, exposes selected case-to-edge relations and
|
||||
keeps them readable when a 2 MiB native output budget omits the full graph.
|
||||
Both real models now answer the selected-case/default question correctly using
|
||||
one retained capture and focused case/byte reads. Code Mode uses fewer tokens
|
||||
in this pair, but unequal native capture time and extra client polling confound
|
||||
the comparison. Whole-task costs and limits are in the same study; a general
|
||||
efficiency advantage remains unproved.
|
||||
@@ -1,420 +0,0 @@
|
||||
// One named contract registry shared by SDK, CLI, direct MCP and Code Mode.
|
||||
import { z } from "zod";
|
||||
import { KunaFactOwner } from "./provider.mjs";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { MAX_BYTE_BATCH_BYTES } from "./limits.mjs";
|
||||
import { createReference } from "./reference.mjs";
|
||||
|
||||
const hex = z.string().regex(/^0x(?:0|[1-9a-f][0-9a-f]{0,15})$/u);
|
||||
const snapshot = z.string().startsWith("snap_");
|
||||
const byteWindow = z.strictObject({
|
||||
address: hex,
|
||||
length: z.number().int().min(1).max(MAX_BYTE_BATCH_BYTES).default(256),
|
||||
});
|
||||
const page = {
|
||||
offset: z.number().int().nonnegative().default(0),
|
||||
limit: z.number().int().min(1).max(128).default(20),
|
||||
};
|
||||
const kinds = z.enum([
|
||||
"value",
|
||||
"load",
|
||||
"store",
|
||||
"phi",
|
||||
"call",
|
||||
"branch",
|
||||
"switch",
|
||||
"return",
|
||||
]);
|
||||
const fields = z
|
||||
.array(
|
||||
z.enum([
|
||||
"kind",
|
||||
"statement",
|
||||
"definitions",
|
||||
"constants",
|
||||
"opcode",
|
||||
"inputs",
|
||||
"output",
|
||||
"is_dead",
|
||||
"input_count",
|
||||
"definition_count",
|
||||
"constant_count",
|
||||
]),
|
||||
)
|
||||
.optional();
|
||||
const definition = (description, input, method, example) => ({
|
||||
description,
|
||||
input: z.strictObject(input),
|
||||
method,
|
||||
example,
|
||||
});
|
||||
const operationSeeds = z.array(z.string()).min(1).max(128);
|
||||
const relationInput = z.discriminatedUnion("kind", [
|
||||
z.strictObject({
|
||||
snapshot,
|
||||
kind: z.literal("cases"),
|
||||
seeds: operationSeeds,
|
||||
case_values: z.array(hex).max(128).optional(),
|
||||
case_indices: z.array(z.number().int().nonnegative()).max(128).optional(),
|
||||
...page,
|
||||
}),
|
||||
z.strictObject({
|
||||
snapshot,
|
||||
kind: z.literal("operands"),
|
||||
seeds: operationSeeds,
|
||||
input_indices: z.array(z.number().int().nonnegative()).max(128).optional(),
|
||||
...page,
|
||||
}),
|
||||
z.strictObject({
|
||||
snapshot,
|
||||
kind: z.enum(["definitions", "uses"]),
|
||||
seeds: operationSeeds,
|
||||
...page,
|
||||
}),
|
||||
z.strictObject({
|
||||
snapshot,
|
||||
kind: z.enum(["incoming", "outgoing", "controllers"]),
|
||||
seeds: z.array(z.number().int().nonnegative()).min(1).max(128),
|
||||
...page,
|
||||
}),
|
||||
]);
|
||||
export const methods = {
|
||||
switches: definition(
|
||||
"Inspect recovered indirect branches and their independent native jump-table metadata, without returning all case rows. Selector is the current recovered BRANCHIND input, which may be normalized rather than the machine jump address. Native consumed mask and signed rendering hint are not complete runtime/signedness proofs. Null tables/defaults stay unknown. The native model default out-edge is separate from any machine out-of-range guard destination. Use relations kind:cases for selected raw table rows, labels and their native mapping to recovered edges; row/edge ordinals are not case values. Available even when the full graph is unavailable if this independent facet was captured.",
|
||||
{
|
||||
snapshot,
|
||||
ids: z.array(z.string()).max(128).optional(),
|
||||
origin_addresses: z.array(hex).max(128).optional(),
|
||||
fields: z.array(z.enum(["selector", "table"])).optional(),
|
||||
...page,
|
||||
},
|
||||
"switches",
|
||||
'await rea.switches({snapshot:s,origin_addresses:["0x1000"]})',
|
||||
),
|
||||
functions: definition(
|
||||
"Search retained native callable entries by literal name/alias, exact name, or entry addresses. Names can match several entries, including PLT/import stubs and implementations; preserve all matches and select by address. A name does not establish runtime symbol binding.",
|
||||
{
|
||||
query: z.string().default(""),
|
||||
exact: z.boolean().default(false),
|
||||
case_sensitive: z.boolean().default(true),
|
||||
addresses: z.array(hex).max(128).optional(),
|
||||
fields: z.array(z.enum(["name", "aliases", "size"])).optional(),
|
||||
...page,
|
||||
},
|
||||
"functions",
|
||||
'await rea.functions({query:"format_error",exact:true})',
|
||||
),
|
||||
strings: definition(
|
||||
"Search retained string inventory by literal text. Paging does not establish artifact completeness.",
|
||||
{ query: z.string(), ...page },
|
||||
"strings",
|
||||
'await rea.strings({query:"incorrect header check"})',
|
||||
),
|
||||
references: definition(
|
||||
"Batch incoming or outgoing lifted-code references without decompilation. Address scope selects exact target (incoming) or source instruction (outgoing); function scope selects outgoing native owning-function references and requires an entry address. Unknown function applicability stays unknown. Kind filtering is pushed to the native query; source rows, endpoints and Evidence identity remain. Recovered IR calls are a separate representation and may omit machine calls. Interior string references require a separate extent query.",
|
||||
{
|
||||
addresses: z.array(hex).min(1).max(16),
|
||||
direction: z.enum(["incoming", "outgoing"]).default("incoming"),
|
||||
scope: z.enum(["address", "function"]).default("address"),
|
||||
kinds: z
|
||||
.array(z.enum(["call", "jump", "data", "read", "write"]))
|
||||
.min(1)
|
||||
.optional(),
|
||||
fields: z
|
||||
.array(z.enum(["fromFunction", "toFunction", "instruction"]))
|
||||
.optional(),
|
||||
...page,
|
||||
},
|
||||
"references",
|
||||
'await rea.references({addresses:["0x2000"]})',
|
||||
),
|
||||
snapshot: definition(
|
||||
"Analyze an explicit function entry once; return a retained capture handle, facets and coverage. Reuses the source within this target lease. Code, recovered calls and values remain independently available when full recovered IR exceeds capacity. A bounded call-rooted operation/definition projection may also be available; inspect facets.operations and coverage before reads. CFG queries still require the complete recovered graph. Available calls include compact by_opcode counts of known/unknown targets over the whole retained call facet; no paging is needed for these counts. Selected calls and complete Evidence remain readable.",
|
||||
{ entry: hex },
|
||||
"snapshot",
|
||||
'await rea.snapshot({entry:"0x1000"})',
|
||||
),
|
||||
bytes: definition(
|
||||
"Read selected native loaded-image bytes without instruction decoding. Batch up to 16 exact virtual-address windows and 64 KiB total selected bytes; hex uses twice that space before provenance. Short reads remain partial. These are not live process memory or an original-file-offset mapping. Global code discovery is not performed.",
|
||||
{
|
||||
windows: z.array(byteWindow).min(1).max(16),
|
||||
},
|
||||
"bytes",
|
||||
'await rea.bytes({windows:[{address:"0x2000",length:32}]})',
|
||||
),
|
||||
instructions: definition(
|
||||
"Batch selected machine instruction windows. Read exact bytes and native disassembly to verify actual control transfers; optimized recovered CFG edges can bypass machine blocks. Address+size with the artifact context identifies the byte range. A caller-selected decode start does not prove execution/alignment.",
|
||||
{
|
||||
windows: z
|
||||
.array(
|
||||
z.strictObject({
|
||||
address: hex,
|
||||
count: z.number().int().min(1).max(4096).default(16),
|
||||
}),
|
||||
)
|
||||
.min(1)
|
||||
.max(16),
|
||||
fields: z
|
||||
.array(z.enum(["bytes", "text", "mnemonic", "operands"]))
|
||||
.optional(),
|
||||
},
|
||||
"instructions",
|
||||
'await rea.instructions({windows:[{address:"0x1000",count:8}],fields:["bytes","text"]})',
|
||||
),
|
||||
operations: definition(
|
||||
"Select SSA operations by IDs, source instruction origins, block IDs, constant literal or kind. Native default rows are compact summaries: opcode/kind/output and input/definition/constant counts; request detailed fields explicitly or page input slots via relations kind:operands. Filters intersect; identities and native is_dead are retained. Availability may be a complete recovered graph or a bounded call-rooted projection; inspect coverage before inferring absence. A literal match is not object identity. origin_addresses selects optimized recovered IR origins; use instructions for actual machine control transfers.",
|
||||
{
|
||||
snapshot,
|
||||
ids: z.array(z.string()).max(128).optional(),
|
||||
origin_addresses: z.array(hex).max(128).optional(),
|
||||
block_ids: z.array(z.number().int().nonnegative()).max(128).optional(),
|
||||
constant: hex.optional(),
|
||||
kind: kinds.optional(),
|
||||
fields,
|
||||
...page,
|
||||
},
|
||||
"operations",
|
||||
'await rea.operations({snapshot:s,constant:"0x2000",fields:["statement"]})',
|
||||
),
|
||||
relations: {
|
||||
description:
|
||||
"Read native operands, definitions/uses or case rows for operation IDs; incoming/outgoing CFG edges or derived controllers for block IDs. cases pages one relation per case/out-edge, with a stable case_id; an unmapped case has one explicit unknown-edge row. total counts relations, not table rows. cases filters optional case_values (native labels tagged value) and case_indices (original row ordinals); both filters intersect and paging spans deduplicated seeds in caller order. Rows preserve missing/no-label status, original recovered target, and separate recovered edges. Partial/unavailable tables and unmapped rows are explicit; no runtime feasibility claim. operands pages every selected input slot, including constants, function inputs and null slots; input_indices optionally selects slots in caller order. Identity is snapshot + operation + input_index; value_id/definition and exact phi predecessor/edge bindings remain separate. Out-of-range slots of retained operations are absent; unretained_seeds remain unknown. A bounded call-rooted projection supports operands/definitions/uses but no CFG; uses cover retained selected operations only. Diagnostic SSA lacks native operands and has partial CFG/printed occurrences. controllers requires a complete recovered graph and concerns finite paths to recovered exits, not runtime feasibility.",
|
||||
input: relationInput,
|
||||
method: "relations",
|
||||
example:
|
||||
'await rea.relations({snapshot:s,kind:"operands",seeds:[op.id],input_indices:[0,2]})',
|
||||
},
|
||||
values: definition(
|
||||
"Read identified native SSA values in a batch; scope may be a full graph, bounded call-rooted definition projection, or call-boundary values only; is_input filters retained operand/output values, not an original source signature. Offset is an address-space offset, not a computed arbitrary-width value. Width, constants, function inputs, annotation status and defining operation are separate facts. Requires the pinned native exporter.",
|
||||
{
|
||||
snapshot,
|
||||
ids: z.array(z.string()).max(128).optional(),
|
||||
is_input: z.boolean().optional(),
|
||||
fields: z
|
||||
.array(
|
||||
z.enum([
|
||||
"space",
|
||||
"offset",
|
||||
"size_bytes",
|
||||
"is_constant",
|
||||
"is_input",
|
||||
"is_annotation",
|
||||
"definition",
|
||||
"type_name",
|
||||
]),
|
||||
)
|
||||
.optional(),
|
||||
...page,
|
||||
},
|
||||
"values",
|
||||
"await rea.values({snapshot:s,ids:op.inputs.map(i=>i.value_id).filter(Boolean)})",
|
||||
),
|
||||
calls: definition(
|
||||
"Select structured recovered call sites and argument bindings. Target comes from native call specifications, never the fspec annotation offset or display text. Each argument separates its SSA value from its passing storage. Calls and their boundary values survive full graph capacity failure. target_value_id retains input zero independently of target resolution; it may be an fspec annotation or an unresolved indirect target value. A null output describes this recovered operation, not proof of a void source signature. Input slots are not original source parameter ordinals; missing storage and indirect/runtime targets remain unknown. To investigate across functions, snapshot the selected callee and select values with is_input:true, comparing exact space/offset/width. Requires the pinned native exporter.",
|
||||
{
|
||||
snapshot,
|
||||
ids: z.array(z.string()).max(128).optional(),
|
||||
origin_addresses: z.array(hex).max(128).optional(),
|
||||
fields: z
|
||||
.array(
|
||||
z.enum([
|
||||
"target",
|
||||
"target_name",
|
||||
"arguments",
|
||||
"output",
|
||||
"prototype",
|
||||
"stack_offset",
|
||||
"target_value_id",
|
||||
"is_dead",
|
||||
]),
|
||||
)
|
||||
.optional(),
|
||||
...page,
|
||||
},
|
||||
"calls",
|
||||
'await rea.calls({snapshot:s,origin_addresses:["0x1100"]})',
|
||||
),
|
||||
blocks: definition(
|
||||
"Read native recovered blocks by stable snapshot-local ID, with ordered predecessors, successor edges and member operation IDs. origin_start describes optimized IR provenance, not a machine block extent. Requires an available native recovered IR graph. Runtime successors remain unknown.",
|
||||
{
|
||||
snapshot,
|
||||
ids: z.array(z.number().int().nonnegative()).max(128).optional(),
|
||||
fields: z
|
||||
.array(
|
||||
z.enum(["origin_start", "operations", "predecessors", "successors"]),
|
||||
)
|
||||
.optional(),
|
||||
...page,
|
||||
},
|
||||
"blocks",
|
||||
"await rea.blocks({snapshot:s,ids:[op.block]})",
|
||||
),
|
||||
slice: definition(
|
||||
"Compose a bounded backward definition traversal from explicit operation IDs. Native default operation rows are compact summaries; request detailed fields or use paged operand relations for wide nodes. Available over a complete recovered graph or bounded call-rooted projection. Missing defining bodies retain frontier reasons; complete use sets and CFG are unavailable for a projection. Default load/phi cutoffs keep boundaries visible; choose stop_kinds:[] to traverse beyond them. Complete reachability and memory aliases remain unknown.",
|
||||
{
|
||||
snapshot,
|
||||
seeds: z.array(z.string()).min(1).max(64),
|
||||
max_operations: z.number().int().min(1).max(256).default(48),
|
||||
stop_at: z.array(z.string()).max(128).default([]),
|
||||
stop_kinds: z.array(kinds).default(["phi", "load"]),
|
||||
fields,
|
||||
},
|
||||
"slice",
|
||||
'await rea.slice({snapshot:s,seeds:callIds,stop_kinds:["load"],fields:["opcode","inputs","output","constants"]})',
|
||||
),
|
||||
code: definition(
|
||||
"Read reconstructed code from the same retained function capture, even when its graph facet is unavailable. offset is zero-based. A large limit reads the complete code inside Code Mode for local filtering. The bridge byte budget bounds the result. No additional decompilation.",
|
||||
{
|
||||
snapshot,
|
||||
offset: page.offset,
|
||||
limit: z.number().int().positive().default(64),
|
||||
},
|
||||
"code",
|
||||
"await rea.code({snapshot:s,offset:100,limit:64})",
|
||||
),
|
||||
evidence: definition(
|
||||
"Read complete retained native source in UTF-16-code-unit pages by evidence_id; no reanalysis. component:source preserves exact original wire JSON. component:recovered_ir reads the decoded graph source, including lossless compressed graphs; source hash remains the original capture and view has its own hash/byte count. This is source inspection, not a claim that malformed graph facts are valid. Available until close_binary.",
|
||||
{
|
||||
evidence_id: z.string().startsWith("ev_"),
|
||||
component: z.enum(["source", "recovered_ir"]).default("source"),
|
||||
offset: page.offset,
|
||||
limit: z.number().int().min(1).max(16384).default(4096),
|
||||
},
|
||||
"evidence",
|
||||
"await rea.evidence({evidence_id:id,offset:0,limit:4096})",
|
||||
),
|
||||
};
|
||||
|
||||
methods.bytes.input = methods.bytes.input.superRefine((value, context) => {
|
||||
// Pattern failures need not prevent object refinements from running in Zod.
|
||||
// Only the named window contract may admit values to exact u64 arithmetic.
|
||||
const windows = value.windows.flatMap((window) => {
|
||||
const parsed = byteWindow.safeParse(window);
|
||||
return parsed.success ? [parsed.data] : [];
|
||||
});
|
||||
if (
|
||||
windows.reduce((sum, window) => sum + window.length, 0) >
|
||||
MAX_BYTE_BATCH_BYTES
|
||||
)
|
||||
context.addIssue({
|
||||
code: "custom",
|
||||
path: ["windows"],
|
||||
message:
|
||||
"Select at most 64 KiB of bytes per batch; use additional windows/pages in later calls",
|
||||
});
|
||||
if (
|
||||
windows.some(
|
||||
(window) =>
|
||||
BigInt(window.address) + BigInt(window.length) > 0xffffffffffffffffn,
|
||||
)
|
||||
)
|
||||
context.addIssue({
|
||||
code: "custom",
|
||||
path: ["windows"],
|
||||
message: "Byte range end exceeds the native u64 address format",
|
||||
});
|
||||
});
|
||||
methods.references.input = methods.references.input.superRefine(
|
||||
(value, context) => {
|
||||
if (value.direction === "incoming" && value.scope !== "address")
|
||||
context.addIssue({
|
||||
code: "custom",
|
||||
path: ["scope"],
|
||||
message: "Incoming references require exact address scope",
|
||||
});
|
||||
},
|
||||
);
|
||||
export class AtomicSession {
|
||||
#owner;
|
||||
#tail = Promise.resolve();
|
||||
constructor(factory = KunaFactOwner.open) {
|
||||
this.factory = factory;
|
||||
}
|
||||
#admit(work) {
|
||||
const pending = this.#tail.then(work, work);
|
||||
this.#tail = pending.catch(() => {});
|
||||
return pending;
|
||||
}
|
||||
get context() {
|
||||
return this.#owner?.context ?? null;
|
||||
}
|
||||
abort() {
|
||||
this.#owner?.cancel();
|
||||
}
|
||||
open(path) {
|
||||
return this.#admit(async () => {
|
||||
if (this.#owner) await this.#owner.close();
|
||||
this.#owner = undefined;
|
||||
this.#owner = await this.factory(path);
|
||||
return {
|
||||
ok: true,
|
||||
context: this.context,
|
||||
methods: Object.keys(methods),
|
||||
evidence_lifetime: "current target lease",
|
||||
};
|
||||
});
|
||||
}
|
||||
close() {
|
||||
return this.#admit(async () => {
|
||||
if (this.#owner) await this.#owner.close();
|
||||
this.#owner = undefined;
|
||||
return { ok: true, closed: true };
|
||||
});
|
||||
}
|
||||
call(
|
||||
name,
|
||||
input,
|
||||
signal = new AbortController().signal,
|
||||
lease = this.context?.lease,
|
||||
) {
|
||||
return this.#admit(async () => {
|
||||
try {
|
||||
if (signal.aborted)
|
||||
throw new FactFailure(
|
||||
"cancelled",
|
||||
"Request was cancelled before admission",
|
||||
);
|
||||
if (!this.#owner)
|
||||
throw new FactFailure(
|
||||
"no_target",
|
||||
"Call open_binary before analysis",
|
||||
);
|
||||
if (lease !== this.context.lease)
|
||||
throw new FactFailure(
|
||||
"stale_context",
|
||||
"Target lease changed; do not join facts across leases",
|
||||
);
|
||||
if (!Object.hasOwn(methods, name))
|
||||
throw new FactFailure(
|
||||
"unknown_method",
|
||||
`Unknown API method: ${name}`,
|
||||
);
|
||||
const contract = methods[name];
|
||||
const parsed = contract.input.parse(input);
|
||||
const data = await this.#owner[contract.method](parsed, signal);
|
||||
if (signal.aborted)
|
||||
throw new FactFailure(
|
||||
"cancelled",
|
||||
"Request was cancelled after observations were collected",
|
||||
{ partial_observation: data },
|
||||
);
|
||||
return { ok: true, context: this.context, data };
|
||||
} catch (error) {
|
||||
return {
|
||||
ok: false,
|
||||
context: this.context,
|
||||
error: {
|
||||
code:
|
||||
error instanceof z.ZodError
|
||||
? "invalid_input"
|
||||
: (error.code ?? "analysis_failed"),
|
||||
message: error.message,
|
||||
details: error.details ?? {},
|
||||
},
|
||||
};
|
||||
}
|
||||
});
|
||||
}
|
||||
}
|
||||
|
||||
export const reference = createReference(methods);
|
||||
@@ -1,145 +0,0 @@
|
||||
import assert from "node:assert/strict";
|
||||
import test from "node:test";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import { AtomicSession, methods, reference } from "./api.mjs";
|
||||
|
||||
const factory = () => {
|
||||
let sequence = 0;
|
||||
const calls = [];
|
||||
const create = async (path) => ({
|
||||
context: { lease: `lease-${++sequence}`, selected_path: path },
|
||||
close: async () => {
|
||||
calls.push("close");
|
||||
},
|
||||
cancel: () => {
|
||||
calls.push("cancel");
|
||||
},
|
||||
strings: async (input) => {
|
||||
calls.push(input);
|
||||
return { items: [], source: { evidence_id: "ev_real-owner" } };
|
||||
},
|
||||
});
|
||||
return { create, calls };
|
||||
};
|
||||
test("relation identity and operand-only selectors agree in runtime and advertised schemas", () => {
|
||||
const validate = new Ajv2020({ strict: false }).compile(
|
||||
reference("relations").methods[0].input_schema,
|
||||
);
|
||||
for (const [request, valid] of [
|
||||
[{ kind: "operands", seeds: ["0x1:0"], input_indices: [0, 155] }, true],
|
||||
[{ kind: "operands", seeds: ["0x1:0"], input_indices: [] }, true],
|
||||
[{ kind: "uses", seeds: ["0x1:0"] }, true],
|
||||
[
|
||||
{
|
||||
kind: "cases",
|
||||
seeds: ["0x1:0"],
|
||||
case_values: ["0xffffffffffffffff"],
|
||||
case_indices: [0, 155],
|
||||
},
|
||||
true,
|
||||
],
|
||||
[{ kind: "cases", seeds: [0] }, false],
|
||||
[{ kind: "cases", seeds: ["0x1:0"], case_values: ["0X1"] }, false],
|
||||
[{ kind: "incoming", seeds: [0] }, true],
|
||||
[{ kind: "incoming", seeds: ["0"] }, false],
|
||||
[{ kind: "operands", seeds: [0] }, false],
|
||||
[{ kind: "uses", seeds: ["0x1:0"], input_indices: [0] }, false],
|
||||
[{ kind: "operands", seeds: ["0x1:0"], input_indices: [-1] }, false],
|
||||
]) {
|
||||
const input = { snapshot: "snap_test", ...request };
|
||||
assert.equal(methods.relations.input.safeParse(input).success, valid);
|
||||
assert.equal(validate(input), valid, JSON.stringify(validate.errors));
|
||||
}
|
||||
});
|
||||
test("named contracts reject extra keys and stale leases before provider admission", async () => {
|
||||
const provider = factory();
|
||||
const session = new AtomicSession(provider.create);
|
||||
await session.open("first");
|
||||
const lease = session.context.lease;
|
||||
const invalid = await session.call("strings", { query: "x", guessed: true });
|
||||
assert.equal(invalid.error.code, "invalid_input");
|
||||
const wrongIdentity = await session.call("relations", {
|
||||
snapshot: "snap_test",
|
||||
kind: "incoming",
|
||||
seeds: ["0"],
|
||||
});
|
||||
assert.equal(wrongIdentity.error.code, "invalid_input");
|
||||
assert.equal(provider.calls.length, 0);
|
||||
await session.open("second");
|
||||
const stale = await session.call("strings", { query: "x" }, undefined, lease);
|
||||
assert.equal(stale.error.code, "stale_context");
|
||||
assert.deepEqual(provider.calls, ["close"]);
|
||||
await session.close();
|
||||
});
|
||||
|
||||
test("cancellation after a completed read retains observations under a cancelled outcome", async () => {
|
||||
const controller = new AbortController();
|
||||
const session = new AtomicSession(async () => ({
|
||||
context: { lease: "late-cancel" },
|
||||
close: async () => {},
|
||||
strings: async () => {
|
||||
controller.abort();
|
||||
return {
|
||||
source: { evidence_id: "ev_observed" },
|
||||
items: [{ value: "retained" }],
|
||||
};
|
||||
},
|
||||
}));
|
||||
await session.open("target");
|
||||
const result = await session.call(
|
||||
"strings",
|
||||
{ query: "x" },
|
||||
controller.signal,
|
||||
);
|
||||
assert.equal(result.ok, false);
|
||||
assert.equal(result.error.code, "cancelled");
|
||||
assert.equal(
|
||||
result.error.details.partial_observation.items[0].value,
|
||||
"retained",
|
||||
);
|
||||
assert.equal(result.context.lease, "late-cancel");
|
||||
await session.close();
|
||||
});
|
||||
test("close drains admitted reads and a failed cleanup retains ownership for retry", async () => {
|
||||
let release;
|
||||
let reads = 0;
|
||||
let closes = 0;
|
||||
const session = new AtomicSession(async () => ({
|
||||
context: { lease: "owned" },
|
||||
strings: () => {
|
||||
reads++;
|
||||
return new Promise((resolve) => {
|
||||
release = resolve;
|
||||
});
|
||||
},
|
||||
close: () => {
|
||||
if (++closes === 1) throw new Error("cleanup failed");
|
||||
},
|
||||
cancel() {},
|
||||
}));
|
||||
await session.open("target");
|
||||
const read = session.call("strings", { query: "x" });
|
||||
await new Promise((resolve) => setImmediate(resolve));
|
||||
const close = session.close();
|
||||
assert.equal(closes, 0);
|
||||
assert.equal(reads, 1);
|
||||
release({ items: [] });
|
||||
assert.equal((await read).ok, true);
|
||||
await assert.rejects(close, /cleanup failed/u);
|
||||
assert.equal(session.context.lease, "owned");
|
||||
await session.close();
|
||||
assert.equal(session.context, null);
|
||||
});
|
||||
test("cancellation before admission makes no provider query; advertised reference schemas compile", async () => {
|
||||
const provider = factory();
|
||||
const session = new AtomicSession(provider.create);
|
||||
await session.open("target");
|
||||
const stop = new AbortController();
|
||||
stop.abort();
|
||||
const cancelled = await session.call("strings", { query: "x" }, stop.signal);
|
||||
assert.equal(cancelled.error.code, "cancelled");
|
||||
assert.deepEqual(provider.calls, []);
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const method of reference().methods) ajv.compile(method.input_schema);
|
||||
await session.close();
|
||||
});
|
||||
@@ -1,53 +0,0 @@
|
||||
# Calls and values surviving full-graph capacity
|
||||
|
||||
The question is `sqlite-boundary-question.txt`. Fix it and this acceptance oracle
|
||||
before either real-model run. Both lanes use the same complete fourteen-method
|
||||
API, preserved engine, model/effort request, common guidance and resource limits.
|
||||
This task requires VM decompilation once per fresh lease; it is different work
|
||||
from the earlier shallow-reference question. Do not compare their task times as
|
||||
an adapter speedup. Actual client JavaScript dispatch is available in both lanes.
|
||||
|
||||
Artifact: Debian `libsqlite3-0:amd64` version `3.40.1-2+deb12u2`,
|
||||
`libsqlite3.so.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
The selected VM implementation entry is `0xea6d0`, not its same-name PLT entry.
|
||||
|
||||
Pre-model public MCP capture with the new exporter establishes:
|
||||
|
||||
- Code is available. Full graph is unavailable with `atomic operation capacity
|
||||
exceeded`. Calls remain available: 416 retained block-member operations,
|
||||
consisting of 395 CALL and 21 CALLIND. The 395 direct targets are known, all
|
||||
21 indirect targets remain unknown. Call-boundary values are available (1,294),
|
||||
covering targets, arguments and outputs only, with 21,349 operations scanned
|
||||
over 1,229 recovered blocks. These do not supply a complete definition/use graph.
|
||||
- At `0xf115c`, recovered call ID `0xf115c:6a84` targets `ram:0x296c0`.
|
||||
Input zero is `v1000118d6`, `fspec:0x102f4`, annotation true, width 8.
|
||||
Three recovered inputs use passing storage `register:0x38/8`, `0x30/8`,
|
||||
`0x10/8` (RDI, RSI, RDX in this verified x86-64 lane).
|
||||
- At `0xf238b`, call ID `0xf238b:7130` has the same resolved target.
|
||||
Input zero is `v100012a80`, `fspec:0x1030c`, annotation true, width 8.
|
||||
Two recovered inputs use passing storage `register:0x38/8`, `0x30/8`.
|
||||
- Both first arguments are the same snapshot-local SSA value `v30000004b`,
|
||||
`register:0x38/8`, `is_input:true`, definition null. Other retained argument
|
||||
values have defining operation IDs outside this boundary view. Exact SSA
|
||||
identity is supported; source parameter ordinals and general aliases are not.
|
||||
|
||||
Independent GNU objdump decoding from the exported VM entry verifies the two
|
||||
machine CALL targets and register setup. The earlier complete direct-call pair
|
||||
oracle contains both. Of the 396 physical direct immediate calls, the only one
|
||||
absent from the recovered 395 is `0xf2493 -> 0x29370` (stack check failure).
|
||||
The question does not require rediscovering this extra call; the recovered-call
|
||||
count must not be claimed as complete machine/runtime coverage.
|
||||
|
||||
Required final facts: correct selected entry and facet availability/scope; both
|
||||
call-count groups and target-unknown distinction; both resolved targets separated
|
||||
from annotation offsets; argument slots/locations; shared first-input identity;
|
||||
selected machine checks; explicit unreadable definition/use and alias/runtime
|
||||
limits; real Evidence and subject digest. Do not require the model to report
|
||||
every additional native metadata counter. Check that its API source arguments
|
||||
show only one VM analysis and all relevant reads reuse its snapshot/Evidence.
|
||||
|
||||
Raw public verification is retained as `facets-v3-mcp.json` and its Evidence
|
||||
directory under `.cache/atomic-codemode/`. The ordinary VM record (C, mappings,
|
||||
variables and other ordinary metadata) equals the prior captured engine output.
|
||||
The independent oracle and precomputed answer are absent from the model workspace.
|
||||
@@ -1,103 +0,0 @@
|
||||
// Native loaded-image data is not instruction decoding or original-file mapping.
|
||||
import { z } from "zod";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { MAX_BYTE_BATCH_BYTES, NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
|
||||
const hex = z.string().regex(/^0x(?:0|[1-9a-f][0-9a-f]{0,15})$/u);
|
||||
const payload = z.string().regex(/^(?:[0-9a-f]{2})*$/u);
|
||||
const header = z.object({
|
||||
binary: z.string(),
|
||||
kind: z.literal("data"),
|
||||
start_hex: hex,
|
||||
end_hex: hex,
|
||||
count: z
|
||||
.number()
|
||||
.int()
|
||||
.nonnegative()
|
||||
.max(MAX_BYTE_BATCH_BYTES / 16),
|
||||
bytes: z.number().int().nonnegative().max(MAX_BYTE_BATCH_BYTES),
|
||||
hex: payload,
|
||||
truncated: z.boolean(),
|
||||
notes: z.array(z.string()),
|
||||
});
|
||||
const row = z.object({
|
||||
address_hex: hex,
|
||||
size: z.number().int().min(1).max(16),
|
||||
bytes: payload,
|
||||
});
|
||||
const requireFact = (valid, message) => {
|
||||
if (!valid) throw new FactFailure("invalid_provider_output", message);
|
||||
};
|
||||
|
||||
export function parseByteFacts(text, inputPath, request) {
|
||||
try {
|
||||
requireFact(
|
||||
Buffer.byteLength(text) <= NATIVE_OUTPUT_BYTES,
|
||||
"Byte source exceeds capture capacity",
|
||||
);
|
||||
const raw = JSON.parse(text);
|
||||
requireFact(
|
||||
Array.isArray(raw?.rows) && raw.rows.length <= MAX_BYTE_BATCH_BYTES / 16,
|
||||
"Byte rows exceed capacity or have invalid shape",
|
||||
);
|
||||
const doc = header.parse(raw);
|
||||
requireFact(
|
||||
doc.binary === inputPath && doc.start_hex === request.address,
|
||||
"Byte source names a different target/window",
|
||||
);
|
||||
requireFact(
|
||||
doc.bytes <= request.length && doc.count === raw.rows.length,
|
||||
"Byte source exceeds the request or disagrees with row count",
|
||||
);
|
||||
let end = BigInt(request.address);
|
||||
const pieces = raw.rows.map((value, index) => {
|
||||
const item = row.parse(value);
|
||||
requireFact(
|
||||
BigInt(item.address_hex) === end && item.bytes.length === item.size * 2,
|
||||
"Byte row address/length is inconsistent",
|
||||
);
|
||||
requireFact(
|
||||
index === raw.rows.length - 1 || item.size === 16,
|
||||
"Nonfinal byte row is not the declared native width",
|
||||
);
|
||||
end += BigInt(item.size);
|
||||
return item.bytes;
|
||||
});
|
||||
requireFact(
|
||||
end === BigInt(doc.end_hex) &&
|
||||
end - BigInt(request.address) === BigInt(doc.bytes) &&
|
||||
pieces.join("") === doc.hex &&
|
||||
doc.hex.length === doc.bytes * 2,
|
||||
"Byte rows, payload and extent disagree",
|
||||
);
|
||||
return {
|
||||
hex: doc.hex,
|
||||
returned_bytes: doc.bytes,
|
||||
end_address: doc.end_hex,
|
||||
coverage: {
|
||||
authority: "native loaded-image bytes",
|
||||
requested_bytes: request.length,
|
||||
status:
|
||||
doc.bytes === request.length
|
||||
? "complete-for-requested-range"
|
||||
: "partial",
|
||||
provider_truncated: doc.truncated,
|
||||
notes: doc.notes,
|
||||
original_file_mapping: "unknown",
|
||||
runtime_memory: "not-observed",
|
||||
},
|
||||
};
|
||||
} catch (error) {
|
||||
if (error instanceof FactFailure) throw error;
|
||||
throw new FactFailure(
|
||||
"invalid_provider_output",
|
||||
"Native byte window does not match the declared format",
|
||||
{
|
||||
reason:
|
||||
error instanceof z.ZodError
|
||||
? error.issues[0]?.message.slice(0, 256)
|
||||
: String(error.message).slice(0, 256),
|
||||
},
|
||||
);
|
||||
}
|
||||
}
|
||||
@@ -1,104 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { parseByteFacts } from "./byte-facts.mjs";
|
||||
import { AtomicSession } from "./api.mjs";
|
||||
|
||||
const make = () => ({
|
||||
binary: "/input",
|
||||
kind: "data",
|
||||
start_hex: "0x20000000000001",
|
||||
end_hex: "0x20000000000005",
|
||||
count: 1,
|
||||
bytes: 4,
|
||||
hex: "00417fff",
|
||||
truncated: false,
|
||||
notes: [],
|
||||
rows: [
|
||||
{
|
||||
address_hex: "0x20000000000001",
|
||||
size: 4,
|
||||
bytes: "00417fff",
|
||||
},
|
||||
],
|
||||
});
|
||||
const request = { address: "0x20000000000001", length: 4 };
|
||||
test("byte facts keep exact u64 extents, null/nontext bytes and loaded-image authority", () => {
|
||||
const facts = parseByteFacts(JSON.stringify(make()), "/input", request);
|
||||
assert.equal(facts.hex, "00417fff");
|
||||
assert.equal(facts.end_address, "0x20000000000005");
|
||||
assert.equal(facts.coverage.status, "complete-for-requested-range");
|
||||
assert.equal(facts.coverage.original_file_mapping, "unknown");
|
||||
assert.equal(facts.coverage.runtime_memory, "not-observed");
|
||||
const clipped = parseByteFacts(JSON.stringify(make()), "/input", {
|
||||
...request,
|
||||
length: 5,
|
||||
});
|
||||
assert.equal(clipped.coverage.status, "partial");
|
||||
const flagged = make();
|
||||
flagged.truncated = true;
|
||||
assert.equal(
|
||||
parseByteFacts(JSON.stringify(flagged), "/input", request).coverage.status,
|
||||
"complete-for-requested-range",
|
||||
);
|
||||
});
|
||||
|
||||
test("wrong byte subject, contradictory duplicated payload, gaps and counts fail before publishing facts", () => {
|
||||
for (const mutate of [
|
||||
(d) => (d.binary = "/other"),
|
||||
(d) => (d.kind = "code"),
|
||||
(d) => (d.hex = "00417f00"),
|
||||
(d) => (d.rows[0].address_hex = "0x20000000000000"),
|
||||
(d) => (d.rows[0].size = 3),
|
||||
(d) => (d.end_hex = "0x20000000000006"),
|
||||
(d) => (d.count = 2),
|
||||
(d) => (d.rows = Array(4097).fill(null)),
|
||||
]) {
|
||||
const document = make();
|
||||
mutate(document);
|
||||
assert.throws(
|
||||
() => parseByteFacts(JSON.stringify(document), "/input", request),
|
||||
{ code: "invalid_provider_output" },
|
||||
);
|
||||
}
|
||||
const empty = make();
|
||||
Object.assign(empty, {
|
||||
hex: "",
|
||||
rows: [],
|
||||
count: 0,
|
||||
bytes: 0,
|
||||
end_hex: request.address,
|
||||
});
|
||||
assert.equal(
|
||||
parseByteFacts(JSON.stringify(empty), "/input", request).coverage.status,
|
||||
"partial",
|
||||
);
|
||||
});
|
||||
|
||||
test("byte input capacity, malformed addresses and u64 end overflow reject before native admission", async () => {
|
||||
let reads = 0;
|
||||
const session = new AtomicSession(async () => ({
|
||||
context: { lease: "bytes" },
|
||||
bytes: () => {
|
||||
reads += 1;
|
||||
},
|
||||
close: () => {},
|
||||
}));
|
||||
await session.open("/input");
|
||||
try {
|
||||
for (const windows of [
|
||||
[{ address: "invalid", length: 1 }],
|
||||
[{ address: "0xffffffffffffffff", length: 1 }],
|
||||
[
|
||||
{ address: "0x1000", length: 65536 },
|
||||
{ address: "0x2000", length: 1 },
|
||||
],
|
||||
[{ address: "0x1000", length: 0 }],
|
||||
]) {
|
||||
const result = await session.call("bytes", { windows });
|
||||
assert.equal(result.error.code, "invalid_input", JSON.stringify(result));
|
||||
}
|
||||
assert.equal(reads, 0);
|
||||
} finally {
|
||||
await session.close();
|
||||
}
|
||||
});
|
||||
@@ -1,64 +0,0 @@
|
||||
# Cross-function context experiment: pinned question and oracle
|
||||
|
||||
This is an experiment acceptance oracle, unavailable to the isolated model.
|
||||
The question is `sqlite-call-question.txt`. Fix this question and the required
|
||||
facts before launching either model lane. Both adapters expose the same complete
|
||||
fourteen-method fact API; Code Mode has four MCP tools, direct has seventeen.
|
||||
Use the same pinned engine, artifact, model request, reasoning effort, common
|
||||
envelope guidance, fresh target leases and resource limits. Do not change runtime
|
||||
sources or HEAD during the model batch. One sample per lane cannot establish a
|
||||
general advantage, and the client itself already offers JavaScript dispatch in
|
||||
both lanes. This comparison isolates the server calling adapter, not all forms of
|
||||
composition or a traditional provider workflow.
|
||||
|
||||
The artifact is the local Debian ELF `libsqlite3.so.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Independent GNU objdump decoding from the function entries covers `0x64060`
|
||||
through `0x65400`; existing earlier captures independently pin the same flow.
|
||||
No target execution or source-code answer is used.
|
||||
|
||||
Required facts for an equal completed answer:
|
||||
|
||||
- Wrapper entry `0x65180`; direct machine CALL `0x65211` targets internal helper
|
||||
`0x64060`. The recovered call operation has an fspec annotation as input zero;
|
||||
that annotation's offset is not the target address.
|
||||
- Caller LEA `0x65201` passes post-prologue `RSP+0x20` in RDI. Structured passing
|
||||
storage is `register:0x38`, width 8. The SSA argument and the callee's retained
|
||||
input are separate snapshot-local identities; matching their storage supplies a
|
||||
candidate connection. The direct machine transfer and pointer setup support
|
||||
this particular connection. Do not globally infer source parameter ordinal,
|
||||
alias completeness or runtime binding from a slot number or a name.
|
||||
- The helper preserves the input context pointer in RBP. Its 32-bit field at
|
||||
context offset `0x30` is incremented at `0x64693`/`0x64696`/`0x64699` for an
|
||||
opening parenthesis and decremented at `0x64c92` for a closing parenthesis.
|
||||
The parser's token/branch context must support the parenthesis interpretation;
|
||||
an ADD/SUB instruction alone does not identify the meaning of a field.
|
||||
- In the wrapper this is `[RSP+0x50]`, initialized to zero with the other context
|
||||
tail fields. After helper return zero, LOAD `0x65268` and TEST/JZ
|
||||
`0x65270`/`0x65272` check it. Zero skips the malformed block; nonzero frees the
|
||||
expression at `0x65274`, clears its output pointer at `0x65279`, then reaches
|
||||
the malformed diagnostic at `0x65281` through formatter call `0x65290`.
|
||||
- Return-code-one errors can also reach the malformed block, via
|
||||
`0x65239`/`0x6523c`; return code 18 selects a separate depth-limit diagnostic
|
||||
at `0x65330`. No demonstrated runtime-feasibility or full memory-alias claim.
|
||||
|
||||
The producer verifier additionally checks two formatter call targets:
|
||||
`0x5f031 -> 0x2adc0` (free PLT) and `0x5f062 -> 0x27e50` (vmprintf PLT), whereas
|
||||
their input-zero fspec offsets are `0x10000`/`0x10001`. Resolving these PLT entries
|
||||
to runtime implementations is outside this task.
|
||||
|
||||
Native producer fidelity is checked against previously retained captures of all
|
||||
three functions: C, ordinary metadata/line mappings, blocks, operations, values
|
||||
and coverage are identical. Only schema version and structured call records are
|
||||
added. Older sources used pretty JSON; their smaller/newer encoded sizes must
|
||||
not be attributed to the call exporter. The earlier compact-writer change is a
|
||||
separate experiment. Unsupported old schema graphs remain unavailable with valid
|
||||
C retained, rather than acquiring guessed call records.
|
||||
|
||||
The first paired question runs at `01e5c5955` both explain the primary state flow
|
||||
correctly but omit the oracle's separate return-1/return-18 distinctions. Because
|
||||
the initial question did not explicitly enumerate those codes, the next round
|
||||
clarifies that requirement in `sqlite-call-question.txt` before either lane
|
||||
starts. The required facts and independent oracle stay unchanged. Earlier
|
||||
questions and costs remain pinned to their original commits; neither incomplete
|
||||
sample is counted as a full-oracle success.
|
||||
@@ -1,20 +0,0 @@
|
||||
import { readFile } from "node:fs/promises";
|
||||
import { AtomicSession } from "./api.mjs";
|
||||
import { executeJavaScript } from "./executor.mjs";
|
||||
|
||||
const [binary, script] = process.argv.slice(2);
|
||||
if (!binary || !script)
|
||||
throw new Error("Usage: node cli.mjs BINARY JAVASCRIPT_FILE");
|
||||
const session = new AtomicSession();
|
||||
try {
|
||||
await session.open(binary);
|
||||
const result = await executeJavaScript(
|
||||
session,
|
||||
await readFile(script, "utf8"),
|
||||
);
|
||||
console.log(JSON.stringify(result));
|
||||
process.exitCode = result.ok ? 0 : 1;
|
||||
} finally {
|
||||
session.abort();
|
||||
await session.close();
|
||||
}
|
||||
@@ -1,51 +0,0 @@
|
||||
# Client waiting and whole-task cost
|
||||
|
||||
The [selected-case pair](switch-study.md) exposes one-second client polls after
|
||||
the snapshot execution yields. Direct makes six waits (three one-second then
|
||||
three ten-second requests); Code Mode makes three one-second requests. Five
|
||||
direct and two Code Mode wait responses still report a running cell. Initial
|
||||
execution yields after about 31 seconds in both lanes. This is observed client
|
||||
behavior, separate from MCP attempts and native source-capture time.
|
||||
|
||||
Native latency remains variable, including in sequential scripted MCP runs
|
||||
without models. The [overlap canary](overlap-study.md) did not establish a cause.
|
||||
No analysis acceleration or cache invalidation change is proposed here.
|
||||
|
||||
## Frozen diagnostic
|
||||
|
||||
The opt-in runner flag `--guide-long-waits` appends exactly the same client wait
|
||||
policy in both lanes: resume an already-running cell with a 30,000 ms wait,
|
||||
respect the advertised schema, and do not sleep or duplicate analysis merely
|
||||
to check progress. It changes neither the initial execution yield policy nor
|
||||
the provider/Code Mode deadlines. It supplies no analysis facts. The policy,
|
||||
flag and instruction digest are recorded in each trial's settings. Older runs
|
||||
without this flag remain a separate stratum.
|
||||
|
||||
Run Code Mode first, then direct, with the unchanged
|
||||
[question](sqlite-switch-question.txt) and [oracle](switch-oracle.md), final
|
||||
`a9ca8a8a...` engine, complete four/eighteen-tool catalogs and the same exact
|
||||
discovery/envelope guidance as the first pair. Use fresh leases and isolated
|
||||
model workspaces, requested `gpt-6.1-sol/high`, Codex 0.161.0 and the same one-CPU,
|
||||
4 GiB hard / 3 GiB early-stop resource policy. No runtime edits during the pair.
|
||||
Local records use `sqlite-switch-{codemode,direct}-v2*`.
|
||||
|
||||
Review complete answers against the frozen independent machine oracle and
|
||||
retained source, including selector widths, all requested cases, signed target
|
||||
arithmetic, case/edge identities, native default versus machine out-of-range
|
||||
guard, provenance and unknown runtime behavior. Require one VM capture and
|
||||
focused reads; a lower token count without a correct answer is not a win.
|
||||
|
||||
Record actual wait arguments and running/completed responses, model responses,
|
||||
all input/cached/uncached/output/reasoning tokens, MCP/API attempts, wall time,
|
||||
family CPU, kernel family peak, native source-capture time, identity/freshness
|
||||
and cleanup/OOM status. Count discovery, guidance and unsuccessful waits in
|
||||
whole-task usage. Fewer polls would establish adherence in this client, not
|
||||
causally lower tokens or native latency. A before/after pair remains subject to
|
||||
model choices, prompt caches, host contention and run order; do not subtract
|
||||
native or polling intervals to manufacture an adapter speedup.
|
||||
|
||||
The [official Codex CLI reference](https://learn.chatgpt.com/docs/cli/reference)
|
||||
documents client configuration but does not establish the `wait` contract used
|
||||
here. The actual captured client calls and subsequent run establish behavior;
|
||||
no configuration feature flag is treated as proof of the available surface.
|
||||
The native/API contracts are unchanged. Results are pending.
|
||||
@@ -1,119 +0,0 @@
|
||||
// Immediate postdominators on the reversed recovered CFG, using linear retained
|
||||
// state. Control dependence is derived for finite paths to recovered exits.
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
export function controlDependencies(
|
||||
blocks,
|
||||
{ workLimit = 2_000_000, dependents } = {},
|
||||
) {
|
||||
const exit = Symbol("recovered-exit");
|
||||
const successors = new Map(
|
||||
blocks.map((block) => [
|
||||
block.id,
|
||||
block.successors.map((edge) => edge.target),
|
||||
]),
|
||||
);
|
||||
const reverse = new Map([
|
||||
[exit, []],
|
||||
...blocks.map((block) => [block.id, []]),
|
||||
]);
|
||||
const tick = (() => {
|
||||
let work = 0;
|
||||
return () => {
|
||||
if (++work > workLimit)
|
||||
throw new FactFailure(
|
||||
"capacity",
|
||||
"Control-dependence work budget exceeded",
|
||||
);
|
||||
};
|
||||
})();
|
||||
for (const [from, targets] of successors) {
|
||||
if (!targets.length) targets.push(exit);
|
||||
for (const to of targets) reverse.get(to).push(from);
|
||||
}
|
||||
const { parent, visited } = postdominatorParents(
|
||||
exit,
|
||||
successors,
|
||||
reverse,
|
||||
tick,
|
||||
);
|
||||
const relations = [];
|
||||
for (const block of blocks) {
|
||||
if (!parent.has(block.id)) continue;
|
||||
for (const edge of block.successors) {
|
||||
let runner = edge.target;
|
||||
while (
|
||||
parent.has(runner) &&
|
||||
runner !== parent.get(block.id) &&
|
||||
runner !== exit
|
||||
) {
|
||||
tick();
|
||||
if (!dependents || dependents.has(runner))
|
||||
relations.push({ controller: block.id, dependent: runner, edge });
|
||||
runner = parent.get(runner);
|
||||
}
|
||||
}
|
||||
}
|
||||
return {
|
||||
relations,
|
||||
coverage: {
|
||||
scope:
|
||||
"derived control dependence in the recovered graph; zero-successor blocks treated as exits; finite paths only",
|
||||
status:
|
||||
visited.size === blocks.length + 1
|
||||
? "complete-for-recovered-finite-exit-paths"
|
||||
: "partial",
|
||||
exit_unreachable_blocks: blocks
|
||||
.filter((block) => !visited.has(block.id))
|
||||
.map((block) => block.id),
|
||||
nontermination: "unknown",
|
||||
runtime_reachability: "unknown",
|
||||
},
|
||||
};
|
||||
}
|
||||
|
||||
function postdominatorParents(exit, successors, reverse, tick) {
|
||||
const visited = new Set([exit]);
|
||||
const stack = [[exit, 0]],
|
||||
postorder = [];
|
||||
while (stack.length) {
|
||||
tick();
|
||||
const frame = stack.at(-1),
|
||||
children = reverse.get(frame[0]);
|
||||
if (frame[1] < children.length) {
|
||||
const child = children[frame[1]++];
|
||||
if (!visited.has(child)) {
|
||||
visited.add(child);
|
||||
stack.push([child, 0]);
|
||||
}
|
||||
} else {
|
||||
postorder.push(frame[0]);
|
||||
stack.pop();
|
||||
}
|
||||
}
|
||||
const order = postorder.reverse();
|
||||
const position = new Map(order.map((id, i) => [id, i]));
|
||||
const parent = new Map([[exit, exit]]);
|
||||
const intersect = (a, b) => {
|
||||
while (a !== b) {
|
||||
tick();
|
||||
if (position.get(a) > position.get(b)) a = parent.get(a);
|
||||
else b = parent.get(b);
|
||||
}
|
||||
return a;
|
||||
};
|
||||
let changed = true;
|
||||
while (changed) {
|
||||
changed = false;
|
||||
for (const node of order.slice(1)) {
|
||||
tick();
|
||||
const known = successors.get(node).filter((id) => parent.has(id));
|
||||
if (!known.length) continue;
|
||||
const next = known.slice(1).reduce(intersect, known[0]);
|
||||
if (parent.get(node) !== next) {
|
||||
parent.set(node, next);
|
||||
changed = true;
|
||||
}
|
||||
}
|
||||
}
|
||||
return { parent, visited };
|
||||
}
|
||||
@@ -1,261 +0,0 @@
|
||||
# Task coverage before cost comparison
|
||||
|
||||
The callable projection now also has a completed [PHI-format model pair](trials.md).
|
||||
Both answers reconstruct all four alternatives with exact retained identities,
|
||||
independent machine/string checks and partial-scope limits. An observed reference
|
||||
lookup mismatch adds work in the direct lane; its costs remain recorded and that
|
||||
discovery boundary needs correction before another same-question comparison.
|
||||
|
||||
Code Mode can only compose facts the underlying API actually supplies. A task
|
||||
cost is comparable after reviewing the completed answer and its evidence. A
|
||||
successful tool call, fewer response bytes or more graph records is insufficient.
|
||||
No claim of complete reverse-engineering capability follows from these tasks.
|
||||
|
||||
The [boundary-facet question/oracle](boundary-oracle.md) now also has a completed
|
||||
real-model pair. Both answers correctly use calls and boundary values despite
|
||||
full-graph capacity failure, reuse one VM analysis and preserve definition/use
|
||||
and runtime uncertainty. Code Mode is faster in this sample and uses slightly
|
||||
less cumulative input, while uncached input and maximum response context are
|
||||
higher. See the full costs and independent source checks in [trials.md](trials.md).
|
||||
This verifies the narrower recovered-call capability; it does not supply selected
|
||||
definition bodies when the full graph is unavailable.
|
||||
|
||||
The next completed pair reuses that exact question and oracle after adding compact
|
||||
call counts to the shared snapshot facet. Both adapters receive the same new
|
||||
metadata, derived after native validation from the complete retained call set;
|
||||
the engine, fact methods and selection semantics are unchanged. The hypothesis
|
||||
is that avoiding full-set paging for these counts reduces task costs. Count
|
||||
coverage remains recovered block-member calls, not complete machine calls or
|
||||
runtime binding. Both answers satisfy the same oracle and avoid call-set paging.
|
||||
Cumulative input is lower in both new samples, but elapsed/uncached costs improve
|
||||
only for direct; direct now beats Code Mode within the pair. These observations
|
||||
support useful shared aggregates, without establishing a general efficiency
|
||||
effect. Selected rows/full source Evidence remain available for composition and
|
||||
audit; all native VM function records equal the earlier captures.
|
||||
|
||||
The atomic direct and Code Mode lanes share the same method registry, provider
|
||||
owner, selectors, failures and retained sources. The traditional lane selects
|
||||
relevant production tools, including the discovery route advertised by its
|
||||
errors; its selection and extra discovery guidance are recorded. Use the same
|
||||
target, engine, authority and question. Freeze runtime files for each model batch.
|
||||
Keep full original errors, retries and partial answers in the record. Never show
|
||||
this oracle document or repository sources to a measured model.
|
||||
|
||||
Review task facts separately from answer length. A confident false physical
|
||||
branch target fails the value-origin task even if its reconstructed IR equation
|
||||
is right. An answer stopping at an import stub does not complete a task whose
|
||||
other lane traces the formatter implementation. A bounded missing graph is not
|
||||
an empty graph. Mark unsupported facts unknown and report incompletion when the
|
||||
question actually needs them.
|
||||
|
||||
Cache regimes are distinct: fresh analysis captures, a reused capture in the
|
||||
same lease, a validated persistent project in a new session, OS file cache and
|
||||
the model's reported prompt-cache tokens. These experiments do not clear the OS
|
||||
cache. The prototype has no cross-session result cache. Whole-model cold and
|
||||
reused-project trials therefore remain separate from the API-only same-lease
|
||||
cache benchmark. Report cumulative input, included cached input, uncached input,
|
||||
output and included reasoning; do not invent server compute or price savings.
|
||||
|
||||
All resource trials are serial with one CPU, 4 GiB hard memory, early stop at
|
||||
3 GiB, and an observed cleanup receipt. Kernel family memory includes charged
|
||||
file cache. A single stochastic trial is an observation, not a reliable ranking;
|
||||
API-only replay cannot replace real model trials or establish their token cost.
|
||||
|
||||
The CLI transcripts establish that all lanes have a client JavaScript `exec`
|
||||
surface, even with `code_mode=false` requested. Count actual model responses,
|
||||
client dispatch/wait calls and MCP requests separately. This comparison is not
|
||||
an ablation against a client incapable of code composition. The optional
|
||||
`agent-trial.py --normalize-mcp-guidance` adds identical envelope/stopping guidance
|
||||
to any lane: select `structuredContent` or parsed text once, avoid printing both
|
||||
encodings, stop when supported and close the target. It is recorded explicitly;
|
||||
the initial SQLite and VM rounds do not have that later guidance. Their native
|
||||
analysis commands/profiles also differ (selected functions versus production
|
||||
project and facet workflows); engine equality alone does not isolate Code Mode.
|
||||
|
||||
The pinned SQLite artifact is `/usr/lib/x86_64-linux-gnu/libsqlite3.so.0.8.6`,
|
||||
SHA-256 `2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
The malformed-MATCH oracle and earlier unequal-depth results are retained in
|
||||
[trials.md](trials.md). The next VM task is fixed in `sqlite-vm-question.txt`
|
||||
before model trials. Its oracle is checked with GNU `objdump -d -M intel` over
|
||||
the shipped executable, independently of both model answers and Kuna IR. Dynamic
|
||||
symbol inventory (`nm -D --defined-only`) identifies `sqlite3VdbeExec` at
|
||||
`0xea6d0`. Reconstructed C and printer line mappings help locate checks, but do
|
||||
not replace the physical instruction evidence:
|
||||
|
||||
| Required local fact | Independently checked anchors |
|
||||
| --------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| Open and release strings are distinct | `0x12c500` and `0x12c538`; loads at `0xf2381` and `0xf2437` |
|
||||
| Operation selector zero takes the open path | Load `[op+4]` at `0xedbe5`, test `0xedbf0`, `JNE 0xf0340` at `0xedbf2` |
|
||||
| Open is rejected for a positive signed counter | Load `[db+0xd8]` at `0xedbf8`, test `0xedbff`, `JG 0xf2372` at `0xedc01` |
|
||||
| Release first resolves the named savepoint, then tests counter and selector | Chain `[db+0x2f0]` at `0xf0340`, links `+0x18` at `0xf0358`, name comparison `0xf036f`; signed `JLE` at `0xf0382` bypasses rejection, selector `==1` at `0xf0384` / `0xf0389` selects `0xf242d` |
|
||||
| Both use the same emission/error tail | Open falls to `sqlite3VdbeError@plt` call `0xf238b`; release jumps there at `0xf2446`; local status becomes `5` at `0xf239d`, then jumps to `0xea828` at `0xf23b8` |
|
||||
| Missing name differs | Null chain tests `0xf0347` / `0xf035f` reach `0xf1139`; separate format `"no such savepoint: %s"` and call `0xf115c` |
|
||||
| Static authority and coverage remain explicit | Counter/selector names are interpretations of offsets; runtime feasibility, actual dynamic binding and the final result after shared error processing require additional evidence |
|
||||
|
||||
The VM's real native-IR exporter exceeds its operation capacity. It retains
|
||||
3,654 lines of reconstructed C and complete source Evidence. This task tests
|
||||
whether a model can use those available facts plus selected physical instruction
|
||||
windows without inventing a recovered graph. Its local status `5` must not be
|
||||
silently promoted to an unconditional final function return: shared error
|
||||
processing can still change behavior. No target execution is authorized here.
|
||||
|
||||
The post-serialization-fix Pragma task is fixed in `sqlite-pragma-question.txt`.
|
||||
It tests a previously inaccessible code facet rather than treating a byte-limit
|
||||
failure as a fast completed answer. The two experimental calling modes use the
|
||||
same twelve methods and new engine digest `2290e8b1…`, with common normalization
|
||||
guidance enabled in both. There is no new traditional cost cell for this task.
|
||||
For the same pinned SQLite artifact, independent GNU instruction checks establish:
|
||||
|
||||
- `sqlite3Pragma` entry `0xb90e0`; value-pointer test `0xbae56` / branch `0xbae5c`
|
||||
distinguishes the read path from change handling at `0xb9b8c`.
|
||||
- `0xb9b90` tests connection byte `+0x65`; zero branches at `0xb9b94` to diagnostic
|
||||
selection at `0xbc58a`. This precedes the additional selector-one bypass at
|
||||
`0xb9b9a` / `0xb9b9f`, so their ordering must be preserved.
|
||||
- Allowed local update computes `(value+1)&7`, substitutes one for zero, and
|
||||
stores the selected record's byte `+0x10` at `0xb9bdc`; byte `+0x11` becomes one
|
||||
at `0xb9be3`. The input interpretation and source member names are derivations.
|
||||
- The actual diagnostic call is `0xbc596` through `sqlite3ErrorMsg@plt` at
|
||||
`0x2a1e0`; `0xbc59b` jumps to common cleanup `0xb9293`.
|
||||
- Cleanup releases both temporary strings, checks the stack canary, and returns
|
||||
at `0xb92d9`. Do not invent a fixed numeric return from this reconstructed
|
||||
void routine. No SQL-level runtime trigger is demonstrated.
|
||||
|
||||
Pragma's complete retained code and mappings are checked equal before/after the
|
||||
serialization fix. Its recovered graph remains explicitly unavailable at the
|
||||
encoded-byte capacity; that is a coverage limitation in both experimental lanes.
|
||||
|
||||
The next seeded table-composition task is fixed in `sqlite-table-question.txt`.
|
||||
Both experimental modes have the same thirteen methods including `bytes`; a
|
||||
traditional comparison is not claimed because its selected catalog/available
|
||||
Kuna byte capability has not been established for this task. GNU disassembly of
|
||||
the pinned artifact shows `sqlite3GetToken` at `0xe1480`: unsigned input byte load
|
||||
at `0xe1480`, class-table base `0x12bcc0` at `0xe1484`, bound `0x1e` at `0xe1492`,
|
||||
class reload `0xe1498`, jump-table base `0x12b600` at `0xe149c`, sign-extending
|
||||
32-bit table load at `0xe14a3`, base addition `0xe14a7`, indirect jump `0xe14aa`.
|
||||
The out-of-range path begins `0xe14f8`.
|
||||
|
||||
Independent direct file reads are valid here because `readelf -lW` binds this
|
||||
artifact's selected addresses to the same file offsets in its read-only LOAD
|
||||
segment (`0x11a000..0x154bfc`). A Python `struct.unpack_from('<i', ...)` check of
|
||||
the actual table bytes gives these initial dispatch facts, before model trials:
|
||||
|
||||
| Input byte | Class | Signed displacement | Initial machine destination |
|
||||
| ---------- | ----: | ------------------: | --------------------------- |
|
||||
| `0x2d` | 11 | -302928 | `0xe16b0` |
|
||||
| `0x2f` | 16 | -302144 | `0xe19c0` |
|
||||
| `0x27` | 8 | -302480 | `0xe1870` |
|
||||
| `0x22` | 8 | -302480 | `0xe1870` |
|
||||
| `0x3f` | 6 | -302272 | `0xe1940` |
|
||||
| `0x80` | 27 | -303440 | `0xe14b0` |
|
||||
|
||||
This oracle establishes static table/dispatch arithmetic, not final token
|
||||
semantics, future runtime table contents or a feasible SQL input. Models must
|
||||
retain those boundaries. This document is not supplied to measured models.
|
||||
|
||||
## Cross-function facts work; adapter gains remain mixed
|
||||
|
||||
The structured-call experiment closes an observed native export gap: the call's
|
||||
fspec annotation does not identify its target code address, and an argument's
|
||||
SSA storage is not its passing location. Reusing the native call-spec owner now
|
||||
allows a model to connect a caller value to a callee input without parsing a
|
||||
display string. Real SQLite answers support this connection with machine
|
||||
instructions, trace a shared parenthesis counter across two functions, and
|
||||
preserve the limits of recovered slots, aliases and runtime feasibility.
|
||||
|
||||
The first question pair omitted oracle-required alternative return paths in both
|
||||
answers. Its wording was less explicit than its oracle, so that result stays a
|
||||
partial-coverage record. A new question clarification was committed before the
|
||||
next pair; both then satisfy the task. The failed engine-pin setup launch also
|
||||
remains recorded. No sample is silently converted into a completed task.
|
||||
|
||||
In the correctness-equal pair, Code Mode uses 596,978 cumulative input tokens and
|
||||
156.821 seconds versus direct's 611,321 and 147.424 seconds. Uncached input,
|
||||
output, largest observed response input and family CPU are higher for Code Mode;
|
||||
family memory peak is lower. Both use the same fourteen fact methods and a client
|
||||
with JavaScript dispatch. The evidence supports usable composable call/input
|
||||
facts, without proving that this server adapter is faster or cheaper. Broader
|
||||
complex-software superiority and provider parity remain unproven.
|
||||
|
||||
Two remaining boundaries deserve independent experiments. First, the initial call view lived inside full recovered IR, so graph capacity
|
||||
could hide a small otherwise useful call view. The next exporter revision now
|
||||
captures calls and their boundary values independently; real VM verification
|
||||
retains both while the graph exceeds capacity. Second, shallow outgoing references should serve a callee/reference list before
|
||||
requesting deeper argument/value recovery. The subsequent implementation now
|
||||
exposes Kuna's existing `xrefs --from` through the shared parser, with native kind
|
||||
filters, explicit scope and paging. Real VM known-target references match an
|
||||
independent disassembly pair set, and remain available when the full IR facet
|
||||
exceeds capacity. The next pre-pinned model question measures this use directly. Keeping those costs
|
||||
and coverage distinctions explicit is more useful than treating a smaller tool
|
||||
catalog as proof of better analysis.
|
||||
|
||||
The shallow-reference question now completes through both real-model adapters.
|
||||
Each composes 396 VM reference rows into 171 exact destinations and 16 calls to
|
||||
the selected PLT address, then distinguishes a real stack-check call absent from
|
||||
optimized IR. Both keep unresolved indirect/discovery/runtime limits. Native
|
||||
reference pairs from both model sessions match the complete independent set of
|
||||
396 direct immediate machine calls; the separate 21 indirect instructions are
|
||||
not silently included in those known-target totals.
|
||||
|
||||
In this one pair, Code Mode takes 93.837 seconds and 264,680 cumulative input
|
||||
versus direct's 106.183 and 285,086. Their maximum observed response input
|
||||
contexts are almost equal. This is useful positive evidence for selected facts
|
||||
plus local aggregation, while prior completed tasks show mixed adapter costs.
|
||||
It supports continuing the experiment, not a universal speed/token/provider
|
||||
superiority claim. Tool coverage and representation semantics were prerequisites
|
||||
for this comparison: a recovered-call-only view would have missed a real call,
|
||||
and VM full-IR capacity cannot establish that its shallow references are absent.
|
||||
|
||||
An additional shared CLI/API run with the unpatched stock Kuna 1.833 executable
|
||||
also returns the same 396/171/16 VM aggregates, with fresh owned HOME/cache. The
|
||||
outgoing-reference capability uses native support, independently of the opt-in
|
||||
IR exporter. It takes 3.044 family wall / 2.695 CPU seconds and 182.6 MiB peak,
|
||||
with cleanup verified and OOM 0; these scripted costs are not model-token data or
|
||||
an adapter comparison. Stock external-spec freshness is still outside the engine
|
||||
binary digest. Full repository typechecking remains resource-limited as recorded
|
||||
in `native-engine.md`; source build, focused tests, changed-test typing and full
|
||||
lint/module checks pass.
|
||||
|
||||
The bounded call-rooted definition projection now supports a local VM PHI task
|
||||
that full-graph capacity previously prevented. Four real-model answers over two
|
||||
revisions correctly recover both PHIs, all four alternatives and machine
|
||||
transfers, with source identities and global partial coverage retained. This is
|
||||
selected static-task coverage, not general complex-software superiority.
|
||||
|
||||
The discovery correction removes an observed name-list miss and fixes defaulted
|
||||
input schemas, but the second pair does not show stable cost improvement. Direct
|
||||
has a 120-second native timeout before a successful retry; Code Mode changes its
|
||||
query strategy and costs more than its first run. The complete costs and failure
|
||||
remain in [the ledger](trials.md#phi-alternatives-after-named-discovery-correction).
|
||||
Fairness requires sufficient facts, correct discovery and honest retry reporting.
|
||||
|
||||
The wide-PHI experiment provides a more precise candidate atom: an input slot is
|
||||
an individually selectable relationship, not just an array inside an operation.
|
||||
Its identity includes the operation and slot index; several relationships can
|
||||
share one value while preserving different predecessor edges. A value definition
|
||||
can then be composed with selected operation/value reads. Both measured models
|
||||
recover those distinctions and separate retained-node absence from unretained
|
||||
membership without fetching all 156 slots.
|
||||
|
||||
This validates the shared query boundary, not adapter superiority. Code Mode
|
||||
groups the same seven fact requests, but four discovery lookups erase the
|
||||
MCP-request reduction at the observed model-response/client-dispatch level. Its
|
||||
95.699 seconds versus direct's 91.817, nearly equal cumulative input and lower
|
||||
uncached input remain a mixed single sample. Discovery is itself part of the
|
||||
analyst workflow and deserves a separate test. Broader behavior reconstruction,
|
||||
complete large-function CFG/use coverage, aliases and a universal RE basis are
|
||||
still unproven. See [the complete scoped results](trials.md#selected-wide-phi-input-relationships).
|
||||
|
||||
The same question after adding a static method index again completes correctly
|
||||
through both adapters. Code Mode now uses two discovery calls instead of four,
|
||||
without changing the seven fact reads. Observed model responses decrease by one,
|
||||
but total time increases slightly; direct also takes longer and uses fewer input
|
||||
tokens without changing query counts. The second pair is approximately 98.8
|
||||
seconds each, with more cumulative and less uncached input for Code Mode. This
|
||||
supports the index's discoverability, while leaving whole-task cost mixed.
|
||||
|
||||
There is a further distinction to test: an atomic fact API does not itself make
|
||||
native computation atomic. A fresh selected-slot question still incurs an entire
|
||||
function decompilation and C rendering. Phase costs are unmeasured, so neither
|
||||
removing C nor introducing persistent state is yet justified as a speed fix.
|
||||
Measure that boundary before claiming analysis-time improvement from narrower
|
||||
fact responses.
|
||||
@@ -1,241 +0,0 @@
|
||||
import { getQuickJS } from "quickjs-emscripten";
|
||||
import { methods } from "./api.mjs";
|
||||
import { MAX_BRIDGE_BYTES as MAX_BRIDGE } from "./limits.mjs";
|
||||
|
||||
const MAX_OUTPUT = 32768;
|
||||
const failure = (code, message) => ({ ok: false, error: { code, message } });
|
||||
const projectGuestError = (code, error) => {
|
||||
const message =
|
||||
typeof error?.message === "string" ? error.message : String(error);
|
||||
// Check raw size first: JSON escaping expands each code unit at most sixfold,
|
||||
// so the second check has a 6×32 KiB string bound plus a fixed envelope.
|
||||
if (
|
||||
Buffer.byteLength(message) > MAX_OUTPUT ||
|
||||
Buffer.byteLength(JSON.stringify({ code, message })) > MAX_OUTPUT
|
||||
)
|
||||
return {
|
||||
code: "output_budget",
|
||||
message:
|
||||
"JavaScript error exceeds the 32 KiB output budget; its message is omitted. Native Evidence remains retained.",
|
||||
details: {
|
||||
original_error_code: code,
|
||||
original_message_code_units: message.length,
|
||||
original_message_omitted: true,
|
||||
},
|
||||
};
|
||||
return { code, message };
|
||||
};
|
||||
export const executeJavaScript = async (
|
||||
session,
|
||||
code,
|
||||
{
|
||||
signal = new AbortController().signal,
|
||||
timeoutMs = 120000,
|
||||
cpuBudgetMs = 2000,
|
||||
} = {},
|
||||
) => {
|
||||
if (Buffer.byteLength(code) > 32768)
|
||||
return failure("code_budget", "JavaScript source exceeds 32 KiB");
|
||||
const started = performance.now();
|
||||
const context = session.context;
|
||||
const lease = context?.lease;
|
||||
const stop = new AbortController();
|
||||
const combined = AbortSignal.any([signal, stop.signal]);
|
||||
const deadline = Date.now() + timeoutMs;
|
||||
const timer = setTimeout(() => stop.abort(), timeoutMs);
|
||||
const QuickJS = await getQuickJS();
|
||||
const runtime = QuickJS.newRuntime();
|
||||
runtime.setMemoryLimit(64 * 1024 * 1024);
|
||||
runtime.setMaxStackSize(512 * 1024);
|
||||
const budget = { used: 0, deadline: 0, exhausted: false };
|
||||
const runGuest = (work) => {
|
||||
budget.deadline = Date.now() + Math.max(1, cpuBudgetMs - budget.used);
|
||||
const before = performance.now();
|
||||
try {
|
||||
return work();
|
||||
} finally {
|
||||
budget.used += performance.now() - before;
|
||||
}
|
||||
};
|
||||
runtime.setInterruptHandler(() => {
|
||||
budget.exhausted ||= Date.now() >= budget.deadline;
|
||||
return budget.exhausted || combined.aborted || Date.now() >= deadline;
|
||||
});
|
||||
const vm = runtime.newContext();
|
||||
const pending = new Set();
|
||||
const calls = [];
|
||||
const state = { disposed: false, evidence: new Set() };
|
||||
const pump = () => {
|
||||
if (!state.disposed) {
|
||||
const result = runGuest(() => runtime.executePendingJobs());
|
||||
if (result.error) {
|
||||
result.error.dispose();
|
||||
stop.abort();
|
||||
}
|
||||
}
|
||||
};
|
||||
installBridge(vm, session, { combined, lease, calls, pending, pump, state });
|
||||
let outcome;
|
||||
try {
|
||||
const text = await evaluateGuest(vm, code, { runGuest, pump, combined });
|
||||
if (Buffer.byteLength(text) > MAX_OUTPUT)
|
||||
outcome = {
|
||||
ok: false,
|
||||
error: {
|
||||
code: "output_budget",
|
||||
message:
|
||||
"Return a smaller projection (32 KiB UTF-8 output budget); native Evidence remains retained",
|
||||
},
|
||||
};
|
||||
else
|
||||
outcome = {
|
||||
ok: true,
|
||||
value: text === "undefined" ? null : JSON.parse(text),
|
||||
};
|
||||
} catch (error) {
|
||||
outcome = {
|
||||
ok: false,
|
||||
error: projectGuestError(
|
||||
budget.exhausted
|
||||
? "cpu_budget"
|
||||
: combined.aborted
|
||||
? "cancelled"
|
||||
: "javascript_error",
|
||||
budget.exhausted
|
||||
? "Guest synchronous execution budget exceeded"
|
||||
: error,
|
||||
),
|
||||
};
|
||||
} finally {
|
||||
clearTimeout(timer);
|
||||
stop.abort();
|
||||
await Promise.allSettled(pending);
|
||||
state.disposed = true;
|
||||
vm.dispose();
|
||||
runtime.dispose();
|
||||
}
|
||||
return {
|
||||
...outcome,
|
||||
context,
|
||||
evidence_ids: [...state.evidence],
|
||||
calls,
|
||||
elapsed_ms: performance.now() - started,
|
||||
};
|
||||
};
|
||||
|
||||
const retainEvidence = (response, evidence) => {
|
||||
const observation =
|
||||
response.data ?? response.error?.details?.partial_observation;
|
||||
for (const row of [observation, ...(observation?.results ?? [])]) {
|
||||
const id =
|
||||
row?.source?.evidence_id ??
|
||||
row?.evidence_id ??
|
||||
row?.error?.details?.evidence_id;
|
||||
if (id) evidence.add(id);
|
||||
}
|
||||
if (response.error?.details?.evidence_id)
|
||||
evidence.add(response.error.details.evidence_id);
|
||||
};
|
||||
|
||||
const installBridge = (
|
||||
vm,
|
||||
session,
|
||||
{ combined, lease, calls, pending, pump, state },
|
||||
) => {
|
||||
const bridge = vm.newFunction("analysis", (methodHandle, inputHandle) => {
|
||||
const name = vm.getString(methodHandle);
|
||||
const inputText = vm.getString(inputHandle);
|
||||
if (calls.length >= 64 || Buffer.byteLength(inputText) > MAX_BRIDGE)
|
||||
throw new Error("Host call or argument budget exceeded");
|
||||
const promise = vm.newPromise();
|
||||
const index = calls.length;
|
||||
calls.push({
|
||||
method: name,
|
||||
input: JSON.parse(inputText),
|
||||
status: "pending",
|
||||
});
|
||||
const work = Promise.resolve()
|
||||
.then(async () => {
|
||||
const before = performance.now();
|
||||
const response = await session.call(
|
||||
name,
|
||||
JSON.parse(inputText),
|
||||
combined,
|
||||
lease,
|
||||
);
|
||||
retainEvidence(response, state.evidence);
|
||||
calls[index].status = response.ok ? "ok" : "error";
|
||||
if (!response.ok) calls[index].error = response.error;
|
||||
calls[index].elapsed_ms = performance.now() - before;
|
||||
let serialized = JSON.stringify(response);
|
||||
if (Buffer.byteLength(serialized) > MAX_BRIDGE)
|
||||
serialized = JSON.stringify({
|
||||
ok: false,
|
||||
error: {
|
||||
code: "bridge_budget",
|
||||
message:
|
||||
"Select fewer fields or a smaller page; full source remains available via evidence",
|
||||
},
|
||||
});
|
||||
if (!state.disposed) {
|
||||
const value = vm.newString(serialized);
|
||||
promise.resolve(value);
|
||||
value.dispose();
|
||||
pump();
|
||||
}
|
||||
})
|
||||
.catch((error) => {
|
||||
if (!state.disposed) {
|
||||
const value = vm.newString(
|
||||
JSON.stringify({
|
||||
ok: false,
|
||||
error: { code: "bridge_failed", message: error.message },
|
||||
}),
|
||||
);
|
||||
promise.resolve(value);
|
||||
value.dispose();
|
||||
pump();
|
||||
}
|
||||
})
|
||||
.finally(() => pending.delete(work));
|
||||
pending.add(work);
|
||||
return promise.handle;
|
||||
});
|
||||
vm.setProp(vm.global, "__analysis", bridge);
|
||||
bridge.dispose();
|
||||
};
|
||||
|
||||
const evaluateGuest = async (vm, code, { runGuest, pump, combined }) => {
|
||||
const bootstrap = `globalThis.rea=Object.freeze(Object.fromEntries(${JSON.stringify(Object.keys(methods))}.map(name=>[name,async(input={})=>JSON.parse(await __analysis(name,JSON.stringify(input)))])));`;
|
||||
vm.unwrapResult(runGuest(() => vm.evalCode(bootstrap))).dispose();
|
||||
const promiseHandle = vm.unwrapResult(
|
||||
runGuest(() =>
|
||||
vm.evalCode(
|
||||
`(async()=>JSON.stringify(await(async()=>{\n${code}\n})()))()`,
|
||||
),
|
||||
),
|
||||
);
|
||||
try {
|
||||
const resolved = vm.resolvePromise(promiseHandle);
|
||||
pump();
|
||||
// A never-settling guest Promise must not hang the MCP request forever.
|
||||
const aborted = new Promise((_, reject) => {
|
||||
if (combined.aborted) {
|
||||
reject(new Error("Execution cancelled"));
|
||||
return;
|
||||
}
|
||||
combined.addEventListener(
|
||||
"abort",
|
||||
() => reject(new Error("Execution cancelled or timed out")),
|
||||
{ once: true },
|
||||
);
|
||||
});
|
||||
const result = await Promise.race([resolved, aborted]);
|
||||
const value = vm.unwrapResult(result);
|
||||
const text = vm.getString(value);
|
||||
value.dispose();
|
||||
return text;
|
||||
} finally {
|
||||
promiseHandle.dispose();
|
||||
}
|
||||
};
|
||||
@@ -1,120 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { executeJavaScript } from "./executor.mjs";
|
||||
|
||||
const fake = {
|
||||
context: { lease: "test" },
|
||||
call: async (name, input) => ({ ok: true, data: { name, input } }),
|
||||
};
|
||||
test("real QuickJS composes async API calls and does not expose host globals", async () => {
|
||||
const result = await executeJavaScript(
|
||||
fake,
|
||||
'const a=await rea.strings({query:"x"}); const b=await rea.operations({ids:[a.data.name]}); return {a,b,host:[typeof process,typeof require,typeof fetch]};',
|
||||
);
|
||||
assert.equal(result.ok, true, JSON.stringify(result));
|
||||
assert.deepEqual(result.value.host, ["undefined", "undefined", "undefined"]);
|
||||
assert.equal(result.calls.length, 2);
|
||||
assert.equal(result.value.b.data.input.ids[0], "strings");
|
||||
});
|
||||
test("infinite guest execution and never-resolving promises terminate", async () => {
|
||||
const loop = await executeJavaScript(fake, "while(true){}", {
|
||||
timeoutMs: 30,
|
||||
});
|
||||
assert.equal(loop.ok, false);
|
||||
const stalled = await executeJavaScript(
|
||||
fake,
|
||||
"await new Promise(()=>{});return 1",
|
||||
{ timeoutMs: 30 },
|
||||
);
|
||||
assert.equal(stalled.ok, false);
|
||||
});
|
||||
test("output budgets and failures survive without silently clipping evidence", async () => {
|
||||
const result = await executeJavaScript(fake, 'return "x".repeat(40000)');
|
||||
assert.equal(result.error.code, "output_budget");
|
||||
const failure = await executeJavaScript(
|
||||
fake,
|
||||
'throw new Error("explicit failure")',
|
||||
);
|
||||
assert.equal(failure.ok, false);
|
||||
});
|
||||
|
||||
test("guest CPU budget is independent from the native-analysis wall deadline", async () => {
|
||||
const before = performance.now();
|
||||
const result = await executeJavaScript(fake, "while(true){}", {
|
||||
timeoutMs: 5000,
|
||||
cpuBudgetMs: 25,
|
||||
});
|
||||
assert.equal(result.error.code, "cpu_budget");
|
||||
assert.ok(performance.now() - before < 2000);
|
||||
});
|
||||
|
||||
test("cancelled reads keep partial-observation Evidence in the execution receipt", async () => {
|
||||
const result = await executeJavaScript(
|
||||
{
|
||||
context: { lease: "partial" },
|
||||
call: async () => ({
|
||||
ok: false,
|
||||
error: {
|
||||
code: "cancelled",
|
||||
details: {
|
||||
partial_observation: { source: { evidence_id: "ev_partial" } },
|
||||
},
|
||||
},
|
||||
}),
|
||||
},
|
||||
"return await rea.strings({query:'x'})",
|
||||
);
|
||||
assert.deepEqual(result.evidence_ids, ["ev_partial"]);
|
||||
assert.equal(result.value.error.code, "cancelled");
|
||||
});
|
||||
|
||||
test("failed compositions keep completed facts and their provenance after draining", async () => {
|
||||
const owner = {
|
||||
context: { lease: "retained" },
|
||||
call: async () => ({ ok: true, data: { evidence_id: "ev_retained" } }),
|
||||
};
|
||||
for (const ending of [
|
||||
'throw new Error("after analysis")',
|
||||
'return "x".repeat(40000)',
|
||||
]) {
|
||||
const result = await executeJavaScript(
|
||||
owner,
|
||||
`await rea.strings({}); ${ending}`,
|
||||
);
|
||||
assert.equal(result.ok, false);
|
||||
assert.deepEqual(result.evidence_ids, ["ev_retained"]);
|
||||
assert.deepEqual(result.context, owner.context);
|
||||
assert.equal(result.calls[0].status, "ok");
|
||||
}
|
||||
});
|
||||
|
||||
test("guest exceptions obey output capacity, including JSON expansion, and preserve observations", async () => {
|
||||
for (const code of [
|
||||
'throw new Error("x".repeat(40000))',
|
||||
'throw new Error("\\0".repeat(6000))',
|
||||
]) {
|
||||
const result = await executeJavaScript(
|
||||
{
|
||||
context: { lease: "error-output" },
|
||||
call: async () => ({
|
||||
ok: true,
|
||||
data: { evidence_id: "ev_before_error" },
|
||||
}),
|
||||
},
|
||||
`await rea.strings({}); ${code}`,
|
||||
);
|
||||
assert.equal(result.ok, false);
|
||||
assert.equal(result.error.code, "output_budget");
|
||||
assert.equal(result.error.details.original_error_code, "javascript_error");
|
||||
assert.equal(result.error.details.original_message_omitted, true);
|
||||
assert.ok(Buffer.byteLength(JSON.stringify(result.error)) <= 32768);
|
||||
assert.deepEqual(result.evidence_ids, ["ev_before_error"]);
|
||||
assert.equal(result.context.lease, "error-output");
|
||||
}
|
||||
const ordinary = await executeJavaScript(
|
||||
fake,
|
||||
'throw new Error("x".repeat(10000))',
|
||||
);
|
||||
assert.equal(ordinary.error.code, "javascript_error");
|
||||
assert.ok(ordinary.error.message.includes("x".repeat(10000)));
|
||||
});
|
||||
@@ -1,24 +0,0 @@
|
||||
import { projectAnalysisError } from "../../dist/domain/analysisErrorProjection.js";
|
||||
|
||||
// Shared expected-failure algebra; execution adapters do not depend on native protocols.
|
||||
export class FactFailure extends Error {
|
||||
constructor(code, message, details = {}) {
|
||||
super(message);
|
||||
this.code = code;
|
||||
this.details = details;
|
||||
}
|
||||
}
|
||||
|
||||
export const requireResult = (result, code = "provider_failure") => {
|
||||
if (!result.ok) {
|
||||
const failure = result.error?._tag
|
||||
? projectAnalysisError(result.error)
|
||||
: result.error;
|
||||
throw new FactFailure(
|
||||
failure?.code ?? code,
|
||||
String(failure?.message ?? failure?.reason ?? failure),
|
||||
{ failure },
|
||||
);
|
||||
}
|
||||
return result.value;
|
||||
};
|
||||
@@ -1,69 +0,0 @@
|
||||
// Shared read projections and partial batch semantics; no provider protocols.
|
||||
// Parsed JSON facts are owned by their cache. SDK callers may retain views,
|
||||
// but must not be able to change later reads under the same Evidence identity.
|
||||
export function freezeFacts(value) {
|
||||
if (value && typeof value === "object") {
|
||||
for (const child of Object.values(value)) freezeFacts(child);
|
||||
Object.freeze(value);
|
||||
}
|
||||
return value;
|
||||
}
|
||||
|
||||
export const sourceSummary = ({ evidence_id, sha256, bytes, elapsed_ms }) => ({
|
||||
evidence_id,
|
||||
sha256,
|
||||
bytes,
|
||||
elapsed_ms,
|
||||
});
|
||||
|
||||
export const projectFact = (row, fields, identity) =>
|
||||
fields
|
||||
? Object.fromEntries(
|
||||
[...new Set([...identity, ...fields])]
|
||||
.filter((key) => Object.hasOwn(row, key))
|
||||
.map((key) => [key, row[key]]),
|
||||
)
|
||||
: row;
|
||||
|
||||
export const pageFacts = (rows, input, identity = []) => ({
|
||||
items: rows
|
||||
.slice(input.offset, input.offset + input.limit)
|
||||
.map((row) => projectFact(row, input.fields, identity)),
|
||||
total: rows.length,
|
||||
next_offset:
|
||||
input.offset + input.limit < rows.length
|
||||
? input.offset + input.limit
|
||||
: null,
|
||||
});
|
||||
|
||||
export async function readFactBatch(requests, capture, read) {
|
||||
const results = [];
|
||||
for (const request of requests) {
|
||||
let source;
|
||||
try {
|
||||
const captured = await capture(request);
|
||||
source = captured.source;
|
||||
const facts = await read(captured, request);
|
||||
results.push({
|
||||
...facts,
|
||||
ok: true,
|
||||
address: request.address,
|
||||
source: sourceSummary(source),
|
||||
});
|
||||
} catch (error) {
|
||||
results.push({
|
||||
ok: false,
|
||||
address: request.address,
|
||||
error: {
|
||||
code: error.code ?? "provider_failure",
|
||||
message: error.message,
|
||||
details: {
|
||||
...error.details,
|
||||
...(source ? { evidence_id: source.evidence_id } : {}),
|
||||
},
|
||||
},
|
||||
});
|
||||
}
|
||||
}
|
||||
return { results };
|
||||
}
|
||||
@@ -1,21 +0,0 @@
|
||||
// Read-only kernel counters for the currently bounded diagnostic payload.
|
||||
import assert from "node:assert/strict";
|
||||
import { readFile } from "node:fs/promises";
|
||||
|
||||
export async function familyCounters() {
|
||||
const group = (await readFile("/proc/self/cgroup", "utf8"))
|
||||
.split("\n")
|
||||
.find((row) => row.startsWith("0::"))
|
||||
?.slice(3);
|
||||
assert.ok(group?.includes("rea-efficiency-"), "Run under resource-run.py");
|
||||
const cpu = (await readFile(`/sys/fs/cgroup${group}/cpu.stat`, "utf8")).match(
|
||||
/^usage_usec (\d+)$/mu,
|
||||
);
|
||||
assert.ok(cpu, "Missing cgroup CPU counter");
|
||||
return {
|
||||
cpu: Number(cpu[1]),
|
||||
peak: Number(
|
||||
(await readFile(`/sys/fs/cgroup${group}/memory.peak`, "utf8")).trim(),
|
||||
),
|
||||
};
|
||||
}
|
||||
@@ -1,33 +0,0 @@
|
||||
import { parseKunaInventory } from "../../dist/kuna/KunaProjectFiles.js";
|
||||
import { requireResult } from "./fact-error.mjs";
|
||||
import { freezeFacts, pageFacts } from "./fact-reads.mjs";
|
||||
|
||||
export const parseFunctionFacts = (text) =>
|
||||
freezeFacts(
|
||||
requireResult(parseKunaInventory(text), "invalid_provider_output"),
|
||||
);
|
||||
|
||||
export function selectFunctionFacts(functions, input) {
|
||||
const addresses = input.addresses && new Set(input.addresses);
|
||||
const normalize = (name) =>
|
||||
input.case_sensitive ? name : name.toLowerCase();
|
||||
const query = normalize(input.query);
|
||||
const matches = functions.filter(
|
||||
(entry) =>
|
||||
(!addresses || addresses.has(entry.address)) &&
|
||||
[entry.name, ...entry.aliases].some((name) =>
|
||||
input.exact
|
||||
? normalize(name) === query
|
||||
: normalize(name).includes(query),
|
||||
),
|
||||
);
|
||||
return {
|
||||
...pageFacts(matches, input, ["address"]),
|
||||
coverage: {
|
||||
scope: "native callable-entry inventory, including import stubs",
|
||||
discovery: "unknown",
|
||||
extents: "upper bounds clipped at the next entry; zero means unmeasured",
|
||||
symbol_binding_at_runtime: "unknown",
|
||||
},
|
||||
};
|
||||
}
|
||||
@@ -1,93 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { parseFunctionFacts, selectFunctionFacts } from "./function-facts.mjs";
|
||||
import { readFactBatch } from "./fact-reads.mjs";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
|
||||
test("function names retain multiple entries and aliases instead of choosing a PLT or implementation", () => {
|
||||
const facts = parseFunctionFacts(
|
||||
JSON.stringify({
|
||||
functions: [
|
||||
{
|
||||
name: "format_error",
|
||||
address_hex: "0x1000",
|
||||
aliases: ["format_error@plt"],
|
||||
size: 16,
|
||||
},
|
||||
{
|
||||
name: "format_error",
|
||||
address_hex: "0x2000",
|
||||
aliases: ["FormatError"],
|
||||
size: 0,
|
||||
},
|
||||
],
|
||||
}),
|
||||
);
|
||||
const query = {
|
||||
query: "format_error",
|
||||
exact: true,
|
||||
case_sensitive: true,
|
||||
offset: 0,
|
||||
limit: 20,
|
||||
};
|
||||
assert.deepEqual(
|
||||
selectFunctionFacts(facts, query).items.map((item) => item.address),
|
||||
["0x1000", "0x2000"],
|
||||
);
|
||||
assert.equal(
|
||||
selectFunctionFacts(facts, { ...query, query: "format_error@plt" }).items[0]
|
||||
.address,
|
||||
"0x1000",
|
||||
);
|
||||
assert.equal(
|
||||
selectFunctionFacts(facts, {
|
||||
...query,
|
||||
query: "FORMAT_ERROR",
|
||||
case_sensitive: false,
|
||||
}).total,
|
||||
2,
|
||||
);
|
||||
assert.deepEqual(
|
||||
selectFunctionFacts(facts, {
|
||||
...query,
|
||||
addresses: ["0x2000"],
|
||||
fields: ["size"],
|
||||
}).items,
|
||||
[{ address: "0x2000", size: 0 }],
|
||||
);
|
||||
assert.throws(() => {
|
||||
facts[0].aliases.push("foreign");
|
||||
});
|
||||
});
|
||||
|
||||
test("batch transport and decoder failures preserve successful items and captured source identity", async () => {
|
||||
const attempts = [];
|
||||
const result = await readFactBatch(
|
||||
[{ address: "0x1" }, { address: "0x2" }, { address: "0x3" }],
|
||||
async (request) => {
|
||||
attempts.push(request.address);
|
||||
if (request.address === "0x2")
|
||||
throw new FactFailure("capacity", "transport bound");
|
||||
return {
|
||||
source: {
|
||||
evidence_id: `ev_${request.address}`,
|
||||
sha256: "digest",
|
||||
bytes: 3,
|
||||
elapsed_ms: 1,
|
||||
},
|
||||
};
|
||||
},
|
||||
async (_, request) => {
|
||||
if (request.address === "0x3")
|
||||
throw new FactFailure(
|
||||
"invalid_provider_output",
|
||||
"decoder rejected source",
|
||||
);
|
||||
return { items: [{ value: 1 }] };
|
||||
},
|
||||
);
|
||||
assert.deepEqual(attempts, ["0x1", "0x2", "0x3"]);
|
||||
assert.equal(result.results[0].items[0].value, 1);
|
||||
assert.equal(result.results[1].error.code, "capacity");
|
||||
assert.equal(result.results[2].error.details.evidence_id, "ev_0x3");
|
||||
});
|
||||
@@ -1,109 +0,0 @@
|
||||
// The machine-decoding boundary is separate from optimized recovered IR.
|
||||
import { z } from "zod";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
|
||||
const hex = z.string().regex(/^0x(?:0|[1-9a-f][0-9a-f]{0,15})$/u);
|
||||
const rowSchema = z.object({
|
||||
address_hex: hex,
|
||||
size: z.number().int().positive().max(65536),
|
||||
bytes: z.string().regex(/^(?:[0-9a-f]{2})+$/u),
|
||||
mnemonic: z.string(),
|
||||
operands: z.string(),
|
||||
text: z.string(),
|
||||
});
|
||||
const envelopeSchema = z.object({
|
||||
binary: z.string(),
|
||||
kind: z.literal("code"),
|
||||
start_hex: hex,
|
||||
end_hex: hex,
|
||||
count: z.number().int().nonnegative().max(4096),
|
||||
bytes: z.number().int().nonnegative(),
|
||||
truncated: z.boolean(),
|
||||
notes: z.array(z.string()),
|
||||
});
|
||||
const requireFact = (condition, message) => {
|
||||
if (!condition) throw new FactFailure("invalid_provider_output", message);
|
||||
};
|
||||
|
||||
export function parseInstructions(
|
||||
text,
|
||||
inputPath,
|
||||
request,
|
||||
fields = ["bytes", "text"],
|
||||
) {
|
||||
try {
|
||||
requireFact(
|
||||
Buffer.byteLength(text) <= NATIVE_OUTPUT_BYTES,
|
||||
"Instruction source exceeds transport capacity",
|
||||
);
|
||||
const raw = JSON.parse(text);
|
||||
requireFact(
|
||||
Array.isArray(raw?.instructions) && raw.instructions.length <= 4096,
|
||||
"Instruction window exceeds decoded-row capacity",
|
||||
);
|
||||
const doc = envelopeSchema.parse(raw);
|
||||
requireFact(
|
||||
doc.binary === inputPath && doc.start_hex === request.address,
|
||||
"Instruction source does not match the selected input/window",
|
||||
);
|
||||
requireFact(
|
||||
doc.count === raw.instructions.length && doc.count <= request.count,
|
||||
"Instruction count does not match the requested window",
|
||||
);
|
||||
let next = BigInt(request.address);
|
||||
const items = raw.instructions.map((value) => {
|
||||
const row = rowSchema.parse(value);
|
||||
requireFact(
|
||||
BigInt(row.address_hex) === next && row.bytes.length === row.size * 2,
|
||||
"Instruction bytes/size or sequential address mismatch",
|
||||
);
|
||||
next += BigInt(row.size);
|
||||
const fact = {
|
||||
address: row.address_hex,
|
||||
size: row.size,
|
||||
status: row.mnemonic === ".byte" ? "undecoded" : "decoded",
|
||||
bytes: row.bytes,
|
||||
mnemonic: row.mnemonic,
|
||||
operands: row.operands,
|
||||
text: row.text,
|
||||
};
|
||||
return Object.fromEntries(
|
||||
[...new Set(["address", "size", "status", ...fields])].map((key) => [
|
||||
key,
|
||||
fact[key],
|
||||
]),
|
||||
);
|
||||
});
|
||||
requireFact(
|
||||
next === BigInt(doc.end_hex) &&
|
||||
next - BigInt(request.address) === BigInt(doc.bytes),
|
||||
"Instruction window end/byte coverage mismatch",
|
||||
);
|
||||
return {
|
||||
items,
|
||||
resume_address: doc.end_hex,
|
||||
coverage: {
|
||||
authority: "machine bytes and native linear decoding; not recovered IR",
|
||||
request,
|
||||
listed_count: items.length,
|
||||
truncated: doc.truncated,
|
||||
notes: doc.notes,
|
||||
alignment:
|
||||
"caller-selected start; runtime execution and instruction alignment are not established",
|
||||
},
|
||||
};
|
||||
} catch (error) {
|
||||
if (error instanceof FactFailure) throw error;
|
||||
throw new FactFailure(
|
||||
"invalid_provider_output",
|
||||
"Native instruction window does not match the declared format",
|
||||
{
|
||||
reason:
|
||||
error instanceof z.ZodError
|
||||
? error.issues[0]?.message.slice(0, 256)
|
||||
: String(error.message).slice(0, 256),
|
||||
},
|
||||
);
|
||||
}
|
||||
}
|
||||
@@ -1,92 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { parseInstructions } from "./instruction-facts.mjs";
|
||||
|
||||
const document = (address = "0xd97e") => ({
|
||||
binary: "/private/input",
|
||||
kind: "code",
|
||||
start_hex: address,
|
||||
end_hex: `0x${(BigInt(address) + 6n).toString(16)}`,
|
||||
count: 1,
|
||||
bytes: 6,
|
||||
truncated: true,
|
||||
notes: ["selected window"],
|
||||
instructions: [
|
||||
{
|
||||
address_hex: address,
|
||||
size: 6,
|
||||
bytes: "0f87e9080000",
|
||||
mnemonic: "JA",
|
||||
operands: "0xe26d",
|
||||
text: "JA 0xe26d",
|
||||
},
|
||||
],
|
||||
});
|
||||
const parse = (doc, address = doc.start_hex) =>
|
||||
parseInstructions(JSON.stringify(doc), "/private/input", {
|
||||
address,
|
||||
count: 1,
|
||||
});
|
||||
|
||||
test("instruction facts preserve exact bytes/addresses and do not substitute an optimized CFG target", () => {
|
||||
const result = parse(document());
|
||||
assert.equal(result.items[0].address, "0xd97e");
|
||||
assert.equal(result.items[0].text, "JA 0xe26d");
|
||||
assert.equal(result.items[0].bytes, "0f87e9080000");
|
||||
assert.equal(result.items[0].status, "decoded");
|
||||
assert.deepEqual(result.coverage.notes, ["selected window"]);
|
||||
const high = document("0x20000000000001");
|
||||
high.instructions[0].address = Number(BigInt(high.start_hex));
|
||||
assert.equal(parse(high).items[0].address, "0x20000000000001");
|
||||
});
|
||||
|
||||
test("unknown bytes remain undecoded and projection retains byte-range identity", () => {
|
||||
const doc = document();
|
||||
doc.instructions[0].mnemonic = ".byte";
|
||||
const result = parseInstructions(
|
||||
JSON.stringify(doc),
|
||||
"/private/input",
|
||||
{ address: "0xd97e", count: 1 },
|
||||
["mnemonic"],
|
||||
);
|
||||
assert.deepEqual(result.items[0], {
|
||||
address: "0xd97e",
|
||||
size: 6,
|
||||
status: "undecoded",
|
||||
mnemonic: ".byte",
|
||||
});
|
||||
});
|
||||
|
||||
test("wrong target/window, gaps, extent and byte-size contradictions fail at decoding boundary", () => {
|
||||
for (const mutate of [
|
||||
(doc) => {
|
||||
doc.binary = "/other/input";
|
||||
},
|
||||
(doc) => {
|
||||
doc.instructions[0].address_hex = "0xd97f";
|
||||
},
|
||||
(doc) => {
|
||||
doc.instructions[0].size = 5;
|
||||
},
|
||||
(doc) => {
|
||||
doc.end_hex = "0xd985";
|
||||
},
|
||||
(doc) => {
|
||||
doc.count = 2;
|
||||
},
|
||||
]) {
|
||||
const doc = document();
|
||||
mutate(doc);
|
||||
assert.throws(
|
||||
() => parse(doc),
|
||||
(error) => error.code === "invalid_provider_output",
|
||||
);
|
||||
}
|
||||
assert.throws(() => parse(document(), "0xd970"), /window/u);
|
||||
});
|
||||
|
||||
test("decoded-row capacity rejects oversized malformed arrays before expanding errors", () => {
|
||||
const doc = document();
|
||||
doc.instructions = Array.from({ length: 4097 }, () => ({}));
|
||||
assert.throws(() => parse(doc), /capacity/u);
|
||||
});
|
||||
@@ -1,86 +0,0 @@
|
||||
--- a/decompiler/crates/kuna-console/src/project.rs
|
||||
+++ b/decompiler/crates/kuna-console/src/project.rs
|
||||
@@ -693,6 +693,17 @@
|
||||
flow_ovr.extend(seed.flow_overrides.iter().cloned());
|
||||
let replay = crate::assertions::has_symbol_scoped(prog, &name, single_target);
|
||||
prog.arch_mut().kuna_record_names = replay;
|
||||
+ let mut profile_clock = (want_atomic_facts
|
||||
+ && std::env::var_os("KUNA_EXPERIMENT_ATOMIC_TIMING").as_deref()
|
||||
+ == Some(std::ffi::OsStr::new("1")))
|
||||
+ .then(std::time::Instant::now);
|
||||
+ let mut profile_phase = |phase: &str| {
|
||||
+ if let Some(clock) = profile_clock.as_mut() {
|
||||
+ let now = std::time::Instant::now();
|
||||
+ eprintln!("{{\"kind\":\"rea_atomic_phase\",\"address\":\"0x{byte_address:x}\",\"phase\":\"{phase}\",\"wall_ns\":{}}}", now.duration_since(*clock).as_nanos());
|
||||
+ *clock = std::time::Instant::now();
|
||||
+ }
|
||||
+ };
|
||||
let step = crate::decompile_step::decompile_one(
|
||||
prog.arch_mut(),
|
||||
&name,
|
||||
@@ -751,6 +762,7 @@
|
||||
}
|
||||
other => other,
|
||||
};
|
||||
+ profile_phase("decompile_drive");
|
||||
match result {
|
||||
Ok(mut fd) => {
|
||||
// (kuna `--assert`) The flow overrides the follower REFUSED are only
|
||||
@@ -813,6 +825,7 @@
|
||||
kuna_decomp::kuna_elemptr::record(prog.arch_mut(), &fd);
|
||||
prog.arch_mut().print_mut().set_record_pointer_arguments(want_pointer_arguments);
|
||||
let rendered = std::panic::catch_unwind(std::panic::AssertUnwindSafe(|| {
|
||||
+ profile_phase("pre_render_metadata");
|
||||
// Trim the surrounding newlines the same way `kuna decompile`
|
||||
// does (`decompile.rs::trim_newlines`), so the per-function
|
||||
// `code` is byte-identical to the single-shot path.
|
||||
@@ -826,6 +839,7 @@
|
||||
};
|
||||
let pointerargs = prog.arch_mut().print_mut().take_pointer_arguments(&untrimmed);
|
||||
let code = untrimmed.trim_matches('\n').to_string();
|
||||
+ profile_phase("render_c");
|
||||
let globals = kuna_decomp::decompile_drive::extract_global_objects(prog.arch());
|
||||
let field_calls = kuna_decomp::kuna_calltargettype::field_declarations(prog.arch(), &fd);
|
||||
let mut variables =
|
||||
@@ -848,15 +862,20 @@
|
||||
prog, &fd, &var_refs, &raw_tokens, &untrimmed, &code, &variables,
|
||||
))
|
||||
});
|
||||
+ profile_phase("metadata_extract");
|
||||
if want_atomic_facts {
|
||||
let detail = detail.get_or_insert_with(Default::default);
|
||||
let membership = kuna_decomp::atomic_membership::inspect(&fd);
|
||||
+ profile_phase("membership");
|
||||
detail.atomic_calls = Some(membership.as_ref().map_err(Clone::clone)
|
||||
.and_then(|m| kuna_decomp::atomic_calls::collect_all(&fd, m)));
|
||||
+ profile_phase("calls");
|
||||
detail.atomic_facts = Some(membership.as_ref().map_err(Clone::clone)
|
||||
.and_then(|m| kuna_decomp::atomic_facts::collect(prog.arch(), &fd, m)));
|
||||
+ profile_phase("full_graph");
|
||||
detail.atomic_projection = Some(membership.as_ref().map_err(Clone::clone)
|
||||
.and_then(|m| kuna_decomp::atomic_projection::collect(prog.arch(), &fd, m)));
|
||||
+ profile_phase("projection");
|
||||
}
|
||||
for variable in &mut variables {
|
||||
for address in &mut variable.addresses {
|
||||
@@ -881,6 +900,7 @@
|
||||
.as_ref()
|
||||
.map(|ctx| ctx.scan(&fd, byte_address))
|
||||
.unwrap_or_default();
|
||||
+ profile_phase("metadata_tail");
|
||||
(code, variables, types, (globals, field_calls), proto, line_mappings, callee_hints, detail, pointerargs, stack_objects)
|
||||
}));
|
||||
match rendered {
|
||||
--- a/docs/spec/00-overview.md
|
||||
+++ b/docs/spec/00-overview.md
|
||||
@@ -3496,3 +3496,10 @@
|
||||
runtime reachability. Collection uses the same live function after C rendering;
|
||||
no second decompilation is performed. Complete source Evidence keeps the native
|
||||
frontier while callers may select a compact summary and individual facts.
|
||||
+
|
||||
+An experiment-only timing overlay recognizes `KUNA_EXPERIMENT_ATOMIC_TIMING=1`
|
||||
+only with atomic capture enabled. It records sequential per-function wall times
|
||||
+to stderr for the decompile drive, pre-render metadata, C rendering, metadata
|
||||
+extraction, membership, calls, full graph, projection and metadata tail. It does
|
||||
+not change analysis parameters or JSON facts. Loading, setup and final JSON
|
||||
+encoding are outside these phase intervals; phase CPU costs are not measured.
|
||||
@@ -1,28 +0,0 @@
|
||||
--- a/decompiler/crates/kuna-decomp/src/substrate/atomic_facts.rs
|
||||
+++ b/decompiler/crates/kuna-decomp/src/substrate/atomic_facts.rs
|
||||
@@ -64,10 +64,10 @@
|
||||
pub fn collect(arch: &Architecture, fd: &Funcdata) -> Result<FunctionFacts, String> {
|
||||
// Bounds cover allocations before JSON serialization, independently of the
|
||||
// transport byte budget. Overflow is an unavailable facet, not clipped IR.
|
||||
- const MAX_OPERATIONS: usize = 12_000;
|
||||
- const MAX_VALUES: usize = 36_000;
|
||||
- const MAX_INPUTS: usize = 50_000;
|
||||
- const MAX_DISPLAY_BYTES: usize = 2 * 1024 * 1024;
|
||||
+ const MAX_OPERATIONS: usize = 32_000;
|
||||
+ const MAX_VALUES: usize = 72_000;
|
||||
+ const MAX_INPUTS: usize = 128_000;
|
||||
+ const MAX_DISPLAY_BYTES: usize = 16 * 1024 * 1024;
|
||||
if fd.bblocks_get_size() <= 0 {
|
||||
return Err("recovered basic-block graph is unavailable".into());
|
||||
}
|
||||
--- a/decompiler/crates/kuna-decomp/src/substrate/atomic_values.rs
|
||||
+++ b/decompiler/crates/kuna-decomp/src/substrate/atomic_values.rs
|
||||
@@ -7,7 +7,7 @@
|
||||
use slotmap::Key;
|
||||
use std::collections::BTreeSet;
|
||||
|
||||
-pub const MAX_TEXT_BYTES: usize = 2 * 1024 * 1024;
|
||||
+pub const MAX_TEXT_BYTES: usize = 16 * 1024 * 1024;
|
||||
#[derive(Debug, Clone, Serialize)]
|
||||
pub struct ValueFact {
|
||||
pub id: String,
|
||||
File diff suppressed because it is too large.
Load diff
@@ -1,5 +0,0 @@
|
||||
// Shared application resource policy, independent of native wire formats.
|
||||
export const NATIVE_OUTPUT_BYTES = 8 * 1024 * 1024;
|
||||
export const MAX_BRIDGE_BYTES = 2 * 1024 * 1024;
|
||||
// Aggregate selected bytes keep hex at 128 KiB, with headroom for provenance.
|
||||
export const MAX_BYTE_BATCH_BYTES = 64 * 1024;
|
||||
@@ -1,70 +0,0 @@
|
||||
"""Lossless encoding feasibility on retained native JSON; no provider implementation."""
|
||||
|
||||
import argparse
|
||||
import base64
|
||||
import gzip
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
import time
|
||||
|
||||
|
||||
def measure(source, output):
|
||||
with source.open('rb') as stream:
|
||||
raw = stream.read(32 * 1024 * 1024 + 1)
|
||||
if len(raw) > 32 * 1024 * 1024:
|
||||
raise ValueError('Input exceeds the diagnostic 32 MiB source bound')
|
||||
document = json.loads(raw)
|
||||
if len(document['functions']) != 1:
|
||||
raise ValueError('Select one retained native function')
|
||||
function = document['functions'][0]
|
||||
graph = function['atomic_facts']
|
||||
graph_json = json.dumps(graph, separators=(',', ':'), ensure_ascii=False).encode()
|
||||
wall, cpu = time.perf_counter(), time.process_time()
|
||||
compressed = gzip.compress(graph_json, compresslevel=1, mtime=0)
|
||||
base64_data = base64.b64encode(compressed)
|
||||
encode_cost = {'wall_ms': (time.perf_counter() - wall) * 1000,
|
||||
'cpu_ms': (time.process_time() - cpu) * 1000}
|
||||
wall, cpu = time.perf_counter(), time.process_time()
|
||||
restored = gzip.decompress(base64.b64decode(base64_data, validate=True))
|
||||
if restored != graph_json:
|
||||
raise ValueError('Lossless byte roundtrip failed')
|
||||
decode_cost = {'wall_ms': (time.perf_counter() - wall) * 1000,
|
||||
'cpu_ms': (time.process_time() - cpu) * 1000}
|
||||
results = {}
|
||||
for name, data in [('gzip-base64', base64_data.decode()),
|
||||
('gzip-hex', compressed.hex())]:
|
||||
function['atomic_facts'] = {
|
||||
'schema_version': 4, 'encoding': name,
|
||||
'raw_bytes': len(graph_json), 'data': data,
|
||||
}
|
||||
results[name] = len(json.dumps(document, separators=(',', ':'),
|
||||
ensure_ascii=False).encode())
|
||||
report = {
|
||||
'source': str(source), 'source_sha256': hashlib.sha256(raw).hexdigest(),
|
||||
'original_document_bytes': len(raw), 'graph_json_bytes': len(graph_json),
|
||||
'graph_statement_bytes': sum(len(op['statement'].encode()) for op in graph['operations']),
|
||||
'graph_operations': len(graph['operations']),
|
||||
'graph_values': len(graph['values']), 'graph_blocks': len(graph['blocks']),
|
||||
'gzip_level': 1, 'gzip_bytes': len(compressed),
|
||||
'candidate_document_bytes': results,
|
||||
'public_transport_budget_bytes': 8 * 1024 * 1024,
|
||||
'lossless_graph_byte_roundtrip': True,
|
||||
'python_gzip_and_base64_encoding': encode_cost,
|
||||
'python_base64_and_gzip_decoding': decode_cost,
|
||||
'scope': 'Offline retained-source format feasibility, not a native/provider/model '
|
||||
'implementation. Timings are Python only, not Rust/Node or agent savings. '
|
||||
'Whole-run CPU/peak family memory are separate resource-run receipts.',
|
||||
}
|
||||
with output.open('x') as stream:
|
||||
json.dump(report, stream, indent=2)
|
||||
stream.write('\n')
|
||||
print(json.dumps(report))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument('source', type=Path)
|
||||
parser.add_argument('output', type=Path)
|
||||
args = parser.parse_args()
|
||||
measure(args.source, args.output)
|
||||
@@ -1,186 +0,0 @@
|
||||
"""Size diagnostic SQLite SSA projections; does not implement a provider API."""
|
||||
|
||||
import argparse
|
||||
from collections import deque
|
||||
import hashlib
|
||||
import json
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
COLUMNS = {
|
||||
'blocks': ['id', 'start', 'operations', 'predecessors', 'successors'],
|
||||
'operations': ['id', 'address', 'block', 'opcode', 'statement', 'is_dead', 'inputs', 'output'],
|
||||
'inputs': ['input_index', 'value_id', 'phi_predecessor', 'phi_predecessor_edge'],
|
||||
'successors': ['id', 'target', 'index', 'polarity', 'condition_value', 'condition_test'],
|
||||
'values': ['id', 'space', 'offset', 'size_bytes', 'is_constant', 'is_input',
|
||||
'is_annotation', 'definition', 'type_name'],
|
||||
}
|
||||
|
||||
|
||||
def encoded(value):
|
||||
return json.dumps(value, ensure_ascii=True, separators=(',', ':')).encode()
|
||||
|
||||
|
||||
def bounded_source(path, budget):
|
||||
with path.open('rb') as source:
|
||||
raw = source.read(budget + 1)
|
||||
assert len(raw) <= budget, 'Use a bounded native capture'
|
||||
return raw
|
||||
|
||||
|
||||
def verify_capture(golden, path):
|
||||
raw = bounded_source(path, 8 * 1024 * 1024)
|
||||
document = json.loads(raw)
|
||||
assert len(document['functions']) == 1
|
||||
candidate = document['functions'][0]
|
||||
ordinary = lambda fn: {key: row for key, row in fn.items()
|
||||
if not key.startswith('atomic_')}
|
||||
assert ordinary(candidate) == ordinary(golden)
|
||||
assert candidate['atomic_calls'] == golden['atomic_calls']
|
||||
projection = candidate['atomic_projection']
|
||||
assert projection['schema_version'] == 1
|
||||
assert projection['coverage']['operation_budget'] == 4096
|
||||
operations = {row['id']: row for row in golden['atomic_facts']['operations']}
|
||||
values = {row['id']: row for row in golden['atomic_facts']['values']}
|
||||
assert all(row == operations[row['id']] for row in projection['operations'])
|
||||
assert all(row == values[row['id']] for row in projection['values'])
|
||||
roots = [row['id'] for row in golden['atomic_calls']['items']]
|
||||
assert projection['roots'] == roots
|
||||
queue, queued, expected = deque(roots), set(roots), []
|
||||
while queue and len(expected) < 4096:
|
||||
key = queue.popleft()
|
||||
if key not in operations:
|
||||
continue
|
||||
expected.append(key)
|
||||
for operand in operations[key]['inputs']:
|
||||
value = values.get(operand['value_id'])
|
||||
definition = value['definition'] if value else None
|
||||
if definition is not None and definition not in queued:
|
||||
queued.add(definition)
|
||||
queue.append(definition)
|
||||
actual = [row['id'] for row in projection['operations']]
|
||||
assert actual == expected
|
||||
referenced = {row['output'] for row in projection['operations'] if row['output'] is not None} | {
|
||||
i['value_id'] for row in projection['operations'] for i in row['inputs']
|
||||
if i['value_id'] is not None}
|
||||
assert len(projection['values']) == len(referenced)
|
||||
assert {row['id'] for row in projection['values']} == referenced
|
||||
missing = {row['definition'] for row in projection['values']
|
||||
if row['definition'] is not None} - set(actual)
|
||||
expected_frontier = {key: 'operation-budget' if key in operations
|
||||
else 'outside-recovered-membership' for key in missing}
|
||||
actual_frontier = {row['definition']: row['reason']
|
||||
for row in projection['coverage']['frontier']}
|
||||
assert len(actual_frontier) == len(projection['coverage']['frontier'])
|
||||
assert actual_frontier == expected_frontier
|
||||
return {'scope': 'Retained-source comparison, not independent machine/runtime equivalence',
|
||||
'source': str(path), 'source_sha256': hashlib.sha256(raw).hexdigest(),
|
||||
'ordinary_record_and_call_facet_equal': True,
|
||||
'all_operation_and_value_records_equal': True,
|
||||
'breadth_first_budget_and_frontier_match': True,
|
||||
'operations': len(actual), 'values': len(referenced), 'frontier': len(missing)}
|
||||
|
||||
|
||||
def pack(table, rows):
|
||||
result = []
|
||||
for row in rows:
|
||||
assert set(row) == set(COLUMNS[table]), 'Do not silently omit unknown fields'
|
||||
result.append([pack(key, row[key]) if key in ('inputs', 'successors') else row[key]
|
||||
for key in COLUMNS[table]])
|
||||
return result
|
||||
|
||||
|
||||
def unpack(table, rows):
|
||||
return [{key: unpack(key, value) if key in ('inputs', 'successors') else value
|
||||
for key, value in zip(COLUMNS[table], row, strict=True)} for row in rows]
|
||||
|
||||
|
||||
def projection(operations, values, roots, depth, cap):
|
||||
retained, wave = set(roots), set(roots)
|
||||
exhausted = False
|
||||
for _ in range(depth):
|
||||
following = {values[i['value_id']]['definition'] for key in wave
|
||||
for i in operations[key]['inputs'] if i['value_id'] is not None
|
||||
and values[i['value_id']]['definition'] in operations} - retained
|
||||
if len(retained | following) > cap:
|
||||
exhausted = True
|
||||
break
|
||||
retained |= following
|
||||
wave = following
|
||||
ids = {operations[key]['output'] for key in retained
|
||||
if operations[key]['output'] is not None} | {
|
||||
i['value_id'] for key in retained for i in operations[key]['inputs']
|
||||
if i['value_id'] is not None}
|
||||
frontier = {values[i['value_id']]['definition'] for key in retained
|
||||
for i in operations[key]['inputs'] if i['value_id'] is not None
|
||||
and values[i['value_id']]['definition'] is not None} - retained
|
||||
payload = {'operations': [operations[key] for key in sorted(retained)],
|
||||
'values': [values[key] for key in sorted(ids)]}
|
||||
return {'requested_definition_depth': depth, 'operation_budget': cap,
|
||||
'operation_budget_exhausted': exhausted, 'operations': len(retained),
|
||||
'values': len(ids), 'encoded_bytes': len(encoded(payload)),
|
||||
'unexpanded_definition_frontier': len(frontier)}
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument('native_json', type=Path)
|
||||
parser.add_argument('report_json', type=Path)
|
||||
parser.add_argument('--verify-capture', type=Path,
|
||||
help='Also compare a public projection capture with this full diagnostic')
|
||||
args = parser.parse_args()
|
||||
budget = 32 * 1024 * 1024
|
||||
raw = bounded_source(args.native_json, budget)
|
||||
document = json.loads(raw)
|
||||
assert len(document['functions']) == 1
|
||||
function = document['functions'][0]
|
||||
assert function['address_hex'] == '0xea6d0', 'This is the seeded SQLite VM experiment'
|
||||
graph = function['atomic_facts']
|
||||
assert graph['schema_version'] == 3
|
||||
assert len(graph['operations']) <= 32000 and len(graph['values']) <= 72000
|
||||
tables = ['blocks', 'operations', 'values']
|
||||
assert set(graph) == {'schema_version', 'coverage', *tables}
|
||||
packed = {'encoding': 'diagnostic-tuples-proposal-not-public',
|
||||
'schema_version': graph['schema_version'], 'columns': COLUMNS,
|
||||
'coverage': graph['coverage'],
|
||||
**{key: pack(key, graph[key]) for key in tables}}
|
||||
restored = {'schema_version': packed['schema_version'], 'coverage': packed['coverage'],
|
||||
**{key: unpack(key, packed[key]) for key in tables}}
|
||||
assert restored == graph
|
||||
operations = {row['id']: row for row in graph['operations']}
|
||||
values = {row['id']: row for row in graph['values']}
|
||||
assert len(operations) == len(graph['operations'])
|
||||
assert len(values) == len(graph['values'])
|
||||
calls = function['atomic_calls']['items']
|
||||
selected = [row['id'] for row in calls
|
||||
if row['origin_address'] in ('0xf115c', '0xf238b')]
|
||||
assert len(selected) == 2
|
||||
report = {
|
||||
'scope': 'Offline diagnostic sizes, not public-provider or model results. '
|
||||
'Tuple roundtrip is lossless, but no native codec is implemented. '
|
||||
'Projections follow retained SSA operands only; no CFG, aliases, '
|
||||
'source equivalence or runtime proof. Bytes are not token/time savings.',
|
||||
'source': str(args.native_json), 'source_sha256': hashlib.sha256(raw).hexdigest(),
|
||||
'whole_native_document_bytes': len(raw), 'full_graph_bytes': len(encoded(graph)),
|
||||
'lossless_tuple_graph_bytes': len(encoded(packed)), 'tuple_roundtrip_equal': True,
|
||||
'estimated_whole_tuple_document_bytes': len(raw) - len(encoded(graph)) + len(encoded(packed)),
|
||||
'operations': len(operations), 'values': len(values), 'blocks': len(graph['blocks']),
|
||||
'inputs': sum(len(row['inputs']) for row in operations.values()),
|
||||
'cfg_edges': sum(len(row['successors']) for row in graph['blocks']),
|
||||
'statement_utf8_bytes': sum(len(row['statement'].encode()) for row in operations.values()),
|
||||
'value_type_utf8_bytes': sum(len(row['type_name'].encode()) for row in values.values()),
|
||||
'selected_call_projections': [projection(operations, values, selected, depth, 512)
|
||||
for depth in range(1, 6)],
|
||||
'all_call_projections': [projection(operations, values, [r['id'] for r in calls], depth, 4096)
|
||||
for depth in range(1, 5)],
|
||||
}
|
||||
if args.verify_capture:
|
||||
report['verified_projection'] = verify_capture(function, args.verify_capture)
|
||||
with args.report_json.open('x') as output:
|
||||
json.dump(report, output, indent=2)
|
||||
output.write('\n')
|
||||
print(json.dumps(report, indent=2))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
@@ -1,57 +0,0 @@
|
||||
import { z } from "zod";
|
||||
import { nativeCall } from "./native-calls.mjs";
|
||||
import {
|
||||
ordinal,
|
||||
value,
|
||||
rowParser,
|
||||
readNativeFacet,
|
||||
requireFact,
|
||||
} from "./native-contract.mjs";
|
||||
|
||||
const callFacet = z.strictObject({
|
||||
schema_version: z.literal(1),
|
||||
items: z.array(nativeCall).max(12000),
|
||||
values: z.array(value).max(36000),
|
||||
coverage: z.strictObject({
|
||||
scope: z.string(),
|
||||
blocks_examined: ordinal.min(1),
|
||||
operations_examined: z
|
||||
.number()
|
||||
.int()
|
||||
.nonnegative()
|
||||
.max(Number.MAX_SAFE_INTEGER),
|
||||
value_scope: z.string(),
|
||||
machine_call_equivalence: z.literal("unknown"),
|
||||
}),
|
||||
});
|
||||
const parseCall = rowParser(nativeCall, {
|
||||
arguments: (row) => nativeCall.shape.arguments.element.parse(row),
|
||||
});
|
||||
const parseFacet = rowParser(callFacet, {
|
||||
items: parseCall,
|
||||
values: (row) => value.parse(row),
|
||||
});
|
||||
export const readCallFacet = (fn) =>
|
||||
readNativeFacet(fn, "atomic_calls", "recovered calls", (raw) => {
|
||||
requireFact(
|
||||
raw && typeof raw === "object" && raw.schema_version === 1,
|
||||
"Unsupported native call facet schema; expected 1",
|
||||
);
|
||||
for (const [name, limit] of [
|
||||
["items", 12000],
|
||||
["values", 36000],
|
||||
])
|
||||
requireFact(
|
||||
Array.isArray(raw[name]) && raw[name].length <= limit,
|
||||
`Native call ${name} exceeds capacity or has invalid shape`,
|
||||
);
|
||||
requireFact(
|
||||
raw.items.reduce(
|
||||
(sum, row) =>
|
||||
sum + (Array.isArray(row?.arguments) ? row.arguments.length : 0),
|
||||
0,
|
||||
) <= 50000,
|
||||
"Native call inputs exceed capacity",
|
||||
);
|
||||
return parseFacet(raw);
|
||||
});
|
||||
@@ -1,162 +0,0 @@
|
||||
import { callCoverage } from "./native-calls.mjs";
|
||||
import { requireFact, requireSameNativeValue } from "./native-contract.mjs";
|
||||
import { uniqueNative, readNativeIdentities } from "./native-reads.mjs";
|
||||
import { freezeFacts } from "./fact-reads.mjs";
|
||||
|
||||
export function indexCallFacet(facts) {
|
||||
const calls = uniqueNative(facts.items),
|
||||
values = uniqueNative(facts.values),
|
||||
used = new Set();
|
||||
requireFact(
|
||||
calls.size <= facts.coverage.operations_examined,
|
||||
"Native call count exceeds examined operation coverage",
|
||||
);
|
||||
const reference = (id) => {
|
||||
requireFact(
|
||||
id === null || values.has(id),
|
||||
"Dangling native call-boundary value",
|
||||
);
|
||||
if (id !== null) used.add(id);
|
||||
};
|
||||
for (const row of calls.values()) {
|
||||
requireFact(
|
||||
row.id.split(":")[0] === row.origin_address,
|
||||
"Native call origin/identity mismatch",
|
||||
);
|
||||
reference(row.target_value_id);
|
||||
reference(row.output);
|
||||
requireFact(
|
||||
row.output === null || values.get(row.output).definition === row.id,
|
||||
"Native call output/definition mismatch",
|
||||
);
|
||||
for (const [i, a] of row.arguments.entries()) {
|
||||
reference(a.value_id);
|
||||
requireFact(
|
||||
a.input_index === i + 1 &&
|
||||
(a.storage === null) === (a.storage_authority === null),
|
||||
"Native call input/storage binding mismatch",
|
||||
);
|
||||
}
|
||||
requireFact(
|
||||
row.spec_available
|
||||
? row.prototype !== null
|
||||
: row.target === null &&
|
||||
row.target_name === null &&
|
||||
row.prototype === null &&
|
||||
row.stack_offset === null &&
|
||||
row.arguments.every((a) => a.storage === null),
|
||||
"Native missing call specification has invented facts",
|
||||
);
|
||||
}
|
||||
requireFact(
|
||||
used.size === values.size,
|
||||
"Native call boundary contains unrelated values",
|
||||
);
|
||||
for (const value of values.values())
|
||||
if (value.definition !== null && calls.has(value.definition))
|
||||
requireFact(
|
||||
calls.get(value.definition).output === value.id,
|
||||
"Native boundary value/known-call definition mismatch",
|
||||
);
|
||||
return { calls, values };
|
||||
}
|
||||
export function validateCallGraph(callIndex, graphIndex) {
|
||||
const { calls, values } = callIndex,
|
||||
{ operations } = graphIndex;
|
||||
for (const row of calls.values()) {
|
||||
const op = operations.get(row.id);
|
||||
requireFact(
|
||||
op &&
|
||||
op.address === row.origin_address &&
|
||||
op.block === row.block &&
|
||||
op.opcode === row.opcode &&
|
||||
op.is_dead === row.is_dead &&
|
||||
op.output === row.output,
|
||||
"Native call/graph ownership mismatch",
|
||||
);
|
||||
requireFact(
|
||||
row.target_value_id === (op.inputs[0]?.value_id ?? null) &&
|
||||
row.arguments.length === Math.max(0, op.inputs.length - 1) &&
|
||||
row.arguments.every((a, i) => a.value_id === op.inputs[i + 1].value_id),
|
||||
"Native call/graph operand mismatch",
|
||||
);
|
||||
}
|
||||
for (const value of values.values())
|
||||
requireSameNativeValue(
|
||||
value,
|
||||
graphIndex.values.get(value.id),
|
||||
"Native call/graph value mismatch",
|
||||
);
|
||||
for (const op of operations.values())
|
||||
requireFact(
|
||||
!["CPUI_CALL", "CPUI_CALLIND"].includes(op.opcode) || calls.has(op.id),
|
||||
"Native call operation omitted from call coverage",
|
||||
);
|
||||
}
|
||||
const summarizeCalls = (calls, coverage) => {
|
||||
const counts = {
|
||||
CPUI_CALL: { known_target_count: 0, unknown_target_count: 0 },
|
||||
CPUI_CALLIND: { known_target_count: 0, unknown_target_count: 0 },
|
||||
};
|
||||
for (const row of calls.values())
|
||||
counts[row.opcode][
|
||||
row.target === null ? "unknown_target_count" : "known_target_count"
|
||||
]++;
|
||||
return freezeFacts({
|
||||
count: calls.size,
|
||||
by_opcode: Object.entries(counts).map(([opcode, row]) => ({
|
||||
opcode,
|
||||
...row,
|
||||
})),
|
||||
coverage,
|
||||
});
|
||||
};
|
||||
export function createCallViews(facts, indexes) {
|
||||
const { calls, values } = indexes;
|
||||
freezeFacts(facts);
|
||||
const coverage = freezeFacts({
|
||||
...callCoverage,
|
||||
...facts.coverage,
|
||||
definition_operations: "not collected in call-boundary facet",
|
||||
});
|
||||
const valueCoverage = freezeFacts({
|
||||
...coverage,
|
||||
scope: facts.coverage.value_scope,
|
||||
});
|
||||
return {
|
||||
calls: (input) => ({
|
||||
...readNativeIdentities(
|
||||
calls,
|
||||
{
|
||||
fields: [
|
||||
"spec_available",
|
||||
"opcode",
|
||||
"is_dead",
|
||||
"target",
|
||||
"target_name",
|
||||
"arguments",
|
||||
"output",
|
||||
],
|
||||
...input,
|
||||
...(input.fields && {
|
||||
fields: ["spec_available", "opcode", "is_dead", ...input.fields],
|
||||
}),
|
||||
},
|
||||
(r) =>
|
||||
!input.origin_addresses ||
|
||||
input.origin_addresses.includes(r.origin_address),
|
||||
),
|
||||
coverage,
|
||||
}),
|
||||
values: (input) => ({
|
||||
...readNativeIdentities(
|
||||
values,
|
||||
input,
|
||||
(r) => input.is_input === undefined || r.is_input === input.is_input,
|
||||
),
|
||||
coverage: valueCoverage,
|
||||
}),
|
||||
summary: summarizeCalls(calls, coverage),
|
||||
value_count: values.size,
|
||||
};
|
||||
}
|
||||
@@ -1,51 +0,0 @@
|
||||
// The native owner supplies targets and passing locations; never parse display.
|
||||
import { z } from "zod";
|
||||
|
||||
import { hex, ordinal, valueId, operationId } from "./native-contract.mjs";
|
||||
const location = z.strictObject({ space: z.string().min(1), offset: hex });
|
||||
const storage = location.extend({
|
||||
size_bytes: z.number().int().positive().max(0x7fffffff),
|
||||
});
|
||||
export const nativeCall = z.strictObject({
|
||||
id: operationId,
|
||||
origin_address: hex,
|
||||
block: ordinal,
|
||||
opcode: z.enum(["CPUI_CALL", "CPUI_CALLIND"]),
|
||||
is_dead: z.boolean(),
|
||||
target_value_id: valueId.nullable(),
|
||||
spec_available: z.boolean(),
|
||||
target: location.nullable(),
|
||||
target_name: z.string().min(1).nullable(),
|
||||
arguments: z.array(
|
||||
z.strictObject({
|
||||
input_index: z.number().int().positive().max(0x7fffffff),
|
||||
value_id: valueId.nullable(),
|
||||
storage: storage.nullable(),
|
||||
storage_authority: z
|
||||
.enum(["final-input-storage", "locked-prototype"])
|
||||
.nullable(),
|
||||
}),
|
||||
),
|
||||
output: valueId.nullable(),
|
||||
prototype: z
|
||||
.strictObject({
|
||||
model: z.string().nullable(),
|
||||
model_unknown: z.boolean().nullable(),
|
||||
inputs_locked: z.boolean(),
|
||||
variadic: z.boolean(),
|
||||
parameter_shift: z.number().int().min(-0x80000000).max(0x7fffffff),
|
||||
})
|
||||
.nullable(),
|
||||
stack_offset: hex.nullable(),
|
||||
});
|
||||
|
||||
export const callCoverage = Object.freeze({
|
||||
scope: "all retained CALL/CALLIND operations in this recovered function",
|
||||
target_authority: "native recovered call specification; not runtime binding",
|
||||
argument_authority:
|
||||
"native finalized passing storage or locked prototype; null remains unknown",
|
||||
source_parameter_ordinals: "unknown",
|
||||
source_signature: "unknown",
|
||||
machine_call_equivalence: "unknown",
|
||||
memory_alias_dependencies: "unknown",
|
||||
});
|
||||
@@ -1,351 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
|
||||
test("malformed compressed graph leaves independent call facts and code readable", () => {
|
||||
const doc = fixture();
|
||||
const originalCode = doc.functions[0].code;
|
||||
doc.functions[0].atomic_facts = {
|
||||
schema_version: 4,
|
||||
encoding: "gzip-base64",
|
||||
raw_bytes: 64,
|
||||
data: "AAAA",
|
||||
};
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.code, originalCode);
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(model.facets.calls.availability, "available");
|
||||
assert.equal(model.calls({}).total, 2);
|
||||
});
|
||||
|
||||
const input = (input_index, value_id) => ({
|
||||
input_index,
|
||||
value_id,
|
||||
phi_predecessor: null,
|
||||
phi_predecessor_edge: null,
|
||||
});
|
||||
const value = (id, space, offset, flags = {}) => ({
|
||||
id,
|
||||
space,
|
||||
offset,
|
||||
size_bytes: 8,
|
||||
is_constant: false,
|
||||
is_annotation: false,
|
||||
is_input: false,
|
||||
definition: null,
|
||||
type_name: "",
|
||||
...flags,
|
||||
});
|
||||
const callOperations = () => [
|
||||
{
|
||||
id: "0x1100:0",
|
||||
address: "0x1100",
|
||||
block: 0,
|
||||
opcode: "CPUI_CALL",
|
||||
statement: "deliberately misleading target 0xbad",
|
||||
is_dead: false,
|
||||
inputs: [input(0, "v0"), input(1, "v1"), input(2, "v2")],
|
||||
output: null,
|
||||
},
|
||||
{
|
||||
id: "0x1110:0",
|
||||
address: "0x1110",
|
||||
block: 0,
|
||||
opcode: "CPUI_CALLIND",
|
||||
statement: "not a signature",
|
||||
is_dead: false,
|
||||
inputs: [input(0, "v2")],
|
||||
output: null,
|
||||
},
|
||||
];
|
||||
const fixtureValues = () => [
|
||||
value("v0", "fspec", "0x10000", { is_annotation: true }),
|
||||
value("v1", "const", "0x2a", { is_constant: true }),
|
||||
value("v2", "register", "0x30", { is_input: true }),
|
||||
];
|
||||
function fixture() {
|
||||
const operations = callOperations(),
|
||||
values = fixtureValues();
|
||||
const rows = [
|
||||
{
|
||||
id: "0x1100:0",
|
||||
spec_available: true,
|
||||
target: { space: "ram", offset: "0x2000" },
|
||||
target_name: "callee",
|
||||
arguments: [
|
||||
{
|
||||
input_index: 1,
|
||||
value_id: "v1",
|
||||
storage: { space: "register", offset: "0x38", size_bytes: 8 },
|
||||
storage_authority: "final-input-storage",
|
||||
},
|
||||
{
|
||||
input_index: 2,
|
||||
value_id: "v2",
|
||||
storage: null,
|
||||
storage_authority: null,
|
||||
},
|
||||
],
|
||||
output: null,
|
||||
prototype: {
|
||||
model: "__cdecl",
|
||||
model_unknown: false,
|
||||
inputs_locked: false,
|
||||
variadic: false,
|
||||
parameter_shift: 0,
|
||||
},
|
||||
stack_offset: null,
|
||||
},
|
||||
{
|
||||
id: "0x1110:0",
|
||||
spec_available: false,
|
||||
target: null,
|
||||
target_name: null,
|
||||
arguments: [],
|
||||
output: null,
|
||||
prototype: null,
|
||||
stack_offset: null,
|
||||
},
|
||||
];
|
||||
return {
|
||||
binary: "/input",
|
||||
count: 1,
|
||||
language: "c",
|
||||
error: null,
|
||||
functions: [
|
||||
{
|
||||
address_hex: "0x1000",
|
||||
code: "retained code",
|
||||
error: null,
|
||||
atomic_facts: {
|
||||
schema_version: 3,
|
||||
blocks: [
|
||||
{
|
||||
id: 0,
|
||||
start: "0x1000",
|
||||
operations: operations.map((o) => o.id),
|
||||
predecessors: [],
|
||||
successors: [],
|
||||
},
|
||||
],
|
||||
operations,
|
||||
values,
|
||||
coverage: {
|
||||
scope: "native graph",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
indirect_branch_operations: [],
|
||||
omitted_definition_ids: [],
|
||||
issues: [],
|
||||
},
|
||||
},
|
||||
atomic_calls: {
|
||||
schema_version: 1,
|
||||
items: rows.map((row, i) => ({
|
||||
...row,
|
||||
origin_address: operations[i].address,
|
||||
block: 0,
|
||||
opcode: operations[i].opcode,
|
||||
is_dead: false,
|
||||
target_value_id: operations[i].inputs[0].value_id,
|
||||
})),
|
||||
values: structuredClone(values),
|
||||
coverage: {
|
||||
scope: "all recovered block-member CALL/CALLIND operations",
|
||||
blocks_examined: 1,
|
||||
operations_examined: 2,
|
||||
value_scope:
|
||||
"call inputs including target and outputs only; definitions may be outside this facet",
|
||||
machine_call_equivalence: "unknown",
|
||||
},
|
||||
},
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
test("structured calls separate fspec identity, SSA value and passing location; missing targets stay unknown", () => {
|
||||
const model = parseNativeIR(JSON.stringify(fixture()), "0x1000");
|
||||
const result = model.calls({ limit: 1 });
|
||||
assert.equal(result.total, 2);
|
||||
assert.equal(result.next_offset, 1);
|
||||
const row = result.items[0];
|
||||
assert.deepEqual(row.target, { space: "ram", offset: "0x2000" });
|
||||
assert.equal(row.arguments[0].value_id, "v1");
|
||||
assert.equal(row.arguments[0].storage.offset, "0x38");
|
||||
assert.equal(row.arguments[1].storage, null);
|
||||
assert.equal(model.values({ ids: ["v1"] }).items[0].offset, "0x2a");
|
||||
assert.equal(Object.hasOwn(row, "prototype"), false);
|
||||
assert.throws(() => {
|
||||
row.arguments[0].storage.offset = "0xbad";
|
||||
}, TypeError);
|
||||
const unknown = model.calls({ offset: 1 }).items[0];
|
||||
assert.equal(unknown.target, null);
|
||||
assert.equal(unknown.spec_available, false);
|
||||
assert.equal(result.coverage.machine_call_equivalence, "unknown");
|
||||
assert.equal(result.coverage.source_parameter_ordinals, "unknown");
|
||||
});
|
||||
|
||||
test("selected call pages and function input values compose without full reports", () => {
|
||||
const model = parseNativeIR(JSON.stringify(fixture()), "0x1000");
|
||||
const result = model.calls({
|
||||
ids: ["0x1100:0", "missing"],
|
||||
origin_addresses: ["0x1100"],
|
||||
fields: ["prototype"],
|
||||
});
|
||||
assert.deepEqual(result.missing, ["missing"]);
|
||||
assert.deepEqual(
|
||||
Object.keys(result.items[0]).sort(),
|
||||
[
|
||||
"id",
|
||||
"origin_address",
|
||||
"block",
|
||||
"spec_available",
|
||||
"opcode",
|
||||
"is_dead",
|
||||
"prototype",
|
||||
].sort(),
|
||||
);
|
||||
assert.equal(
|
||||
model.calls({ ids: ["0x1100:0"], origin_addresses: [] }).total,
|
||||
0,
|
||||
);
|
||||
assert.deepEqual(
|
||||
model.values({ is_input: true }).items.map((v) => v.id),
|
||||
["v2"],
|
||||
);
|
||||
assert.equal(model.values({ is_input: true, ids: ["v1"] }).total, 0);
|
||||
});
|
||||
|
||||
test("bad call facets preserve valid graph and code; cross-facet conflicts disable call composition", () => {
|
||||
const changes = [
|
||||
(f) => {
|
||||
f.items.pop();
|
||||
},
|
||||
(f) => {
|
||||
f.items.push(f.items[0]);
|
||||
},
|
||||
(f) => {
|
||||
f.items[0].arguments[0].value_id = "v2";
|
||||
},
|
||||
(f) => {
|
||||
f.items[0].arguments[0].input_index = 0;
|
||||
},
|
||||
(f) => {
|
||||
f.items[0].arguments[0].storage_authority = null;
|
||||
},
|
||||
(f) => {
|
||||
f.items[1].target = { space: "ram", offset: "0x2000" };
|
||||
},
|
||||
(f) => {
|
||||
f.values[1].offset = "0x2b";
|
||||
},
|
||||
(f) => {
|
||||
f.schema_version = 2;
|
||||
},
|
||||
];
|
||||
for (const change of changes) {
|
||||
const doc = fixture();
|
||||
change(doc.functions[0].atomic_calls);
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.code, "retained code");
|
||||
assert.equal(model.facets.recovered_ir.availability, "available");
|
||||
assert.equal(model.facets.calls.availability, "unavailable");
|
||||
assert.throws(
|
||||
() => model.calls({}),
|
||||
(e) => e.code === "invalid_provider_output",
|
||||
);
|
||||
assert.equal(model.values({ ids: ["v1"] }).items[0].offset, "0x2a");
|
||||
}
|
||||
});
|
||||
|
||||
test("graph capacity failure retains calls, boundary values and explicit unreadable definitions", () => {
|
||||
const doc = fixture(),
|
||||
fn = doc.functions[0];
|
||||
delete fn.atomic_facts;
|
||||
fn.atomic_facts_error = "atomic operation capacity exceeded";
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(model.facets.calls.availability, "available");
|
||||
assert.equal(model.facets.values.scope, "call-boundary-only");
|
||||
assert.equal(model.calls({}).total, 2);
|
||||
const values = model.values({ ids: ["v0", "v1", "missing"] });
|
||||
assert.equal(values.items[0].space, "fspec");
|
||||
assert.equal(values.items[1].offset, "0x2a");
|
||||
assert.deepEqual(values.missing, ["missing"]);
|
||||
assert.match(values.coverage.value_scope, /only/);
|
||||
assert.equal(values.coverage.scope, values.coverage.value_scope);
|
||||
assert.throws(
|
||||
() => model.select({}),
|
||||
(e) => e.code === "capability_unavailable",
|
||||
);
|
||||
assert.equal(
|
||||
model.calls({ ids: ["0x1110:0"], fields: ["target_value_id"] }).items[0]
|
||||
.target_value_id,
|
||||
"v2",
|
||||
);
|
||||
});
|
||||
|
||||
test("malformed graph is isolated from valid calls; call-only malformed identities still fail", () => {
|
||||
const doc = fixture();
|
||||
doc.functions[0].atomic_facts.operations[0].output = "vbad";
|
||||
let model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(model.facets.calls.availability, "available");
|
||||
assert.equal(model.values({ is_input: true }).items[0].id, "v2");
|
||||
doc.functions[0].atomic_calls.items[0].target_value_id = "vdead";
|
||||
model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.calls.availability, "unavailable");
|
||||
assert.throws(
|
||||
() => model.values({}),
|
||||
(e) => e.code === "invalid_provider_output",
|
||||
);
|
||||
assert.equal(model.code, "retained code");
|
||||
});
|
||||
|
||||
test("capture call counts distinguish target knowledge from opcode and include dead rows", () => {
|
||||
const doc = fixture(),
|
||||
fn = doc.functions[0];
|
||||
delete fn.atomic_facts;
|
||||
fn.atomic_facts_error = "capacity";
|
||||
const [direct, indirect] = fn.atomic_calls.items;
|
||||
direct.target = null;
|
||||
direct.is_dead = true;
|
||||
indirect.target = { space: "ram", offset: "0x3000" };
|
||||
indirect.spec_available = true;
|
||||
indirect.prototype = direct.prototype;
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.deepEqual(model.facets.calls.by_opcode, [
|
||||
{ opcode: "CPUI_CALL", known_target_count: 0, unknown_target_count: 1 },
|
||||
{ opcode: "CPUI_CALLIND", known_target_count: 1, unknown_target_count: 0 },
|
||||
]);
|
||||
assert.equal(model.facets.calls.count, 2);
|
||||
assert.equal(model.facets.calls.coverage.machine_call_equivalence, "unknown");
|
||||
assert.equal(model.calls({ ids: [direct.id], fields: [] }).total, 1);
|
||||
assert.equal(model.facets.calls.count, 2);
|
||||
assert.throws(() => {
|
||||
model.facets.calls.by_opcode[0].known_target_count = 99;
|
||||
}, TypeError);
|
||||
});
|
||||
|
||||
test("an empty available call facet has zero counts; unavailable counts remain absent", () => {
|
||||
const doc = fixture(),
|
||||
fn = doc.functions[0];
|
||||
delete fn.atomic_facts;
|
||||
fn.atomic_facts_error = "capacity";
|
||||
fn.atomic_calls.items = [];
|
||||
fn.atomic_calls.values = [];
|
||||
let model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.calls.availability, "available");
|
||||
assert.equal(model.facets.calls.count, 0);
|
||||
assert.ok(
|
||||
model.facets.calls.by_opcode.every(
|
||||
(r) => r.known_target_count === 0 && r.unknown_target_count === 0,
|
||||
),
|
||||
);
|
||||
delete fn.atomic_calls;
|
||||
fn.atomic_calls_error = "capacity";
|
||||
model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.calls.availability, "unavailable");
|
||||
assert.equal(Object.hasOwn(model.facets.calls, "count"), false);
|
||||
assert.equal(Object.hasOwn(model.facets.calls, "by_opcode"), false);
|
||||
});
|
||||
@@ -1,122 +0,0 @@
|
||||
# Native computation and retained fact coverage
|
||||
|
||||
This diagnostic separates native computation from fact selection and Code Mode
|
||||
dispatch after the wide-operand model pairs. It does not change those adapters
|
||||
or establish model-token savings.
|
||||
|
||||
## Phase instrument and actual measurements
|
||||
|
||||
Use experiment commit `7d0da075e` and its versions of both patches to reproduce
|
||||
these historical measurements. Apply `kuna-native-ir.patch`, then
|
||||
`kuna-atomic-timing-diagnostic.patch`, to Kuna
|
||||
`fe01d6294eb0d1053ce343c8943646df88e143fc`. Reuse existing compiled specs and the
|
||||
one-job opt-level-1 embed-specs build in `native-engine.md`. Preserved engine:
|
||||
`c755184455b8fe8950a5194788d954bc0476602779b6dd614cb893a230f6fde7`;
|
||||
overlay: `b2ddfd5b6d3fc4a386cfbb07d97fbd0fcbfd8279837ce68327e62baa3e0f3018`.
|
||||
The preceding `a0685486...` public executable remains independently preserved.
|
||||
|
||||
The instrument requires `--atomic-facts` and the exact environment value
|
||||
`KUNA_EXPERIMENT_ATOMIC_TIMING=1`. It emits nine sequential per-function wall
|
||||
intervals to stderr, excluding logging time from the next interval. Disabled
|
||||
instrumentation emits none. Analysis parameters and JSON facts are unchanged.
|
||||
These are phase wall times, not phase CPU. Loading, preparation before the drive
|
||||
and final output encoding are outside the intervals; the residual cannot all be
|
||||
attributed to one phase.
|
||||
|
||||
`profile-native.mjs` uses production target/installation/process owners and a
|
||||
private runtime/cache root. Its diagnostic launcher adds only the timing flag;
|
||||
ambient environment is not persisted. It measures two new native processes:
|
||||
fresh private cache, then the same cache/staged input. This second process is
|
||||
distinct from the API owner's warm result-cache hit. Complete stdout/stderr and
|
||||
digests are retained before validation. OS cache remains uncontrolled.
|
||||
|
||||
```sh
|
||||
python3 experiments/analysis-efficiency/resource-run.py --observe-payload-exit -- \
|
||||
node experiments/atomic-codemode/profile-native.mjs \
|
||||
/absolute/diagnostic-kuna /usr/lib/x86_64-linux-gnu/libsqlite3.so.0 \
|
||||
0xea6d0 /absolute/new-profile.json /absolute/prior-public-capture.json
|
||||
```
|
||||
|
||||
Add `--without-timing` for the disabled control. Target: SQLite Debian
|
||||
`3.40.1-2+deb12u2` amd64, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
VM entry is `0xea6d0`; parser-helper entry is `0x64060`.
|
||||
|
||||
| Wall milliseconds | VM fresh | VM same cache/new process | Helper fresh | Helper same cache/new process |
|
||||
| --------------------- | ---------: | ------------------------: | -----------: | ----------------------------: |
|
||||
| Entire owned request | 36,336.306 | 36,353.240 | 2,339.393 | 2,193.311 |
|
||||
| Decompile drive | 33,681.369 | 33,855.870 | 256.200 | 266.629 |
|
||||
| C rendering | 220.052 | 215.235 | 9.399 | 7.416 |
|
||||
| Metadata extraction | 64.434 | 128.247 | 1.576 | 1.564 |
|
||||
| Membership scan | 3.259 | 3.496 | 0.137 | 0.093 |
|
||||
| Call facts | 2.889 | 3.328 | 0.245 | 0.133 |
|
||||
| Full graph collection | 0.006* | 0.007* | 6.993 | 5.473 |
|
||||
| Projection collection | 31.383 | 31.711 | 1.955 | 1.770 |
|
||||
|
||||
Pre-render/tail metadata also appear in the nine-interval raw receipts. `*` VM
|
||||
full-graph collection fails its early operation-capacity check; these numbers
|
||||
do not time a complete graph export. Whole-function analysis nevertheless
|
||||
produces the live IR. Each function record deeply equals its prior public golden
|
||||
and its repeat. The disabled helper control also agrees, with zero timing rows.
|
||||
|
||||
| Whole-run resource scope | Family wall seconds | CPU seconds | Kernel peak MiB |
|
||||
| ------------------------------------- | ------------------: | ----------: | --------------: |
|
||||
| Instrument build | 115.279 | 92.082 | 760.7 |
|
||||
| VM, two processes plus validation | 74.218 | 73.867 | 572.2 |
|
||||
| Helper, two processes plus validation | 5.554 | 5.328 | 191.9 |
|
||||
| Disabled helper, two processes | 5.556 | 5.409 | 191.8 |
|
||||
| Final cached build | 0.322 | 0.155 | 26.8 |
|
||||
|
||||
All use one CPU, 4 GiB hard / 3 GiB early-stop bounds, verified cleanup and OOM 0.
|
||||
Kernel peak includes charged file cache, not RSS. Final build keeps the same
|
||||
digest. Patch forward/reverse and lenient `check-spec` pass; full native corpus,
|
||||
stages and workspace gates remain unexecuted. No upstream commit/PR or integration
|
||||
readiness is claimed. Receipts/raw sources are local
|
||||
`phase-timing-{vm,helper,helper-off}.json*` and resource logs under
|
||||
`.cache/atomic-codemode/`.
|
||||
|
||||
C rendering is below 1% of this VM request; skipping it cannot remove the
|
||||
approximately 34-second drive. The helper instead spends most time outside the
|
||||
instrumented phases, requiring further profiling before choosing persistent
|
||||
state. Same-cache new processes reanalyze. Previously measured same-lease API
|
||||
source reuse remains a separate approximately 40 ms path.
|
||||
|
||||
## Lossless full-graph encoding feasibility
|
||||
|
||||
The earlier diagnostic source has 21,349 operations, 35,755 values and 1,229
|
||||
blocks, with no omitted definition IDs. Its whole document is 17,801,972 bytes;
|
||||
the graph is 17,008,223 bytes, including 2,255,616 bytes of operation display and
|
||||
90,013 bytes of value type names. Plain tuple packing was still too large.
|
||||
|
||||
```sh
|
||||
python3 experiments/analysis-efficiency/resource-run.py --observe-payload-exit -- \
|
||||
python3 experiments/atomic-codemode/measure-native-encoding.py \
|
||||
/absolute/diagnostic-native.json /absolute/new-encoding-report.json
|
||||
```
|
||||
|
||||
Gzip level 1 makes the graph 2,020,464 bytes. Hypothetical documents retaining
|
||||
ordinary C/metadata/calls and replacing only the graph with a typed envelope are:
|
||||
|
||||
| Encoding | Entire JSON bytes | Existing transport budget |
|
||||
| ------------- | ----------------: | ------------------------: |
|
||||
| gzip + base64 | 3,487,776 | 8,388,608 |
|
||||
| gzip + hex | 4,834,749 | 8,388,608 |
|
||||
|
||||
The graph byte roundtrip is exact. Python gzip/base64 encode costs 69.037 ms
|
||||
wall / 69.037 ms CPU; decode 34.632 / 34.571 ms. Whole sizing task: 0.826 wall /
|
||||
0.654 CPU seconds, 169.8 MiB peak, cleanup verified and OOM 0. These are offline
|
||||
Python feasibility measurements, not Rust/Node timings, native output, public
|
||||
support or agent savings. Source SHA-256:
|
||||
`0a31513350be440983094710197b155a2ec8c49d28852bb20a524292091637b7`;
|
||||
reproducible receipt: `full-graph-encoding-repro.json` beside its resource log.
|
||||
|
||||
This supports testing bounded lossless transport before introducing live-function
|
||||
ownership or repeating analysis for missing definitions. Required checks include
|
||||
allocation/encoding growth, bounded inflation, exact format/length/row admission,
|
||||
malformed-graph isolation from calls/C, complete Evidence access, actual CPU and
|
||||
memory, fallback behavior and real-model workflows. The shared owner must return
|
||||
selected facts rather than send an encoded graph to the agent. The current
|
||||
public engine at this sizing checkpoint still has a bounded call-rooted
|
||||
projection. The [subsequent implementation](native-graph-transport-study.md)
|
||||
now verifies full recovered-graph access through bounded native/Node encoding
|
||||
and both public MCP adapters; real-model verification remains separate.
|
||||
@@ -1,105 +0,0 @@
|
||||
// Shared native identity/value contracts and bounded row/facet decoding.
|
||||
import { z } from "zod";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { ok, err } from "../../dist/domain/result.js";
|
||||
export const requireFact = (condition, reason) => {
|
||||
if (!condition) throw new FactFailure("invalid_provider_output", reason);
|
||||
};
|
||||
export const hex = z.string().regex(/^0x(?:0|[1-9a-f][0-9a-f]{0,15})$/u);
|
||||
export const edgeId = z.string().regex(/^\d+:\d+$/u);
|
||||
export const ordinal = z.number().int().nonnegative().max(0x7fffffff);
|
||||
export const valueId = z.string().regex(/^v[0-9a-f]{1,16}$/u);
|
||||
export const operationId = z
|
||||
.string()
|
||||
.regex(/^0x[0-9a-f]{1,16}:[0-9a-f]{1,8}$/u);
|
||||
export const value = z.strictObject({
|
||||
id: valueId,
|
||||
space: z.string().nullable(),
|
||||
offset: hex,
|
||||
size_bytes: z.number().int().positive().max(0x7fffffff),
|
||||
is_constant: z.boolean(),
|
||||
is_input: z.boolean(),
|
||||
is_annotation: z.boolean(),
|
||||
definition: operationId.nullable(),
|
||||
type_name: z.string(),
|
||||
});
|
||||
export const requireSameNativeValue = (value, reference, message) =>
|
||||
requireFact(
|
||||
reference &&
|
||||
Object.entries(value).every(([key, v]) => reference[key] === v),
|
||||
message,
|
||||
);
|
||||
export const rowParser = (schema, arrays, fields = {}) => {
|
||||
const metadata = schema.omit(
|
||||
Object.fromEntries(
|
||||
[...Object.keys(arrays), ...Object.keys(fields)].map((key) => [
|
||||
key,
|
||||
true,
|
||||
]),
|
||||
),
|
||||
);
|
||||
return (raw) => {
|
||||
const metadataFields = { ...raw };
|
||||
for (const key of [...Object.keys(arrays), ...Object.keys(fields)])
|
||||
delete metadataFields[key];
|
||||
const result = metadata.parse(metadataFields);
|
||||
for (const [key, parse] of Object.entries(arrays)) {
|
||||
requireFact(Array.isArray(raw[key]), `Native ${key} must be an array`);
|
||||
result[key] = raw[key].map(parse);
|
||||
}
|
||||
for (const [key, parse] of Object.entries(fields))
|
||||
result[key] = parse(raw[key]);
|
||||
return result;
|
||||
};
|
||||
};
|
||||
export const formatFailure = (error) => {
|
||||
if (error instanceof FactFailure) return error;
|
||||
const details =
|
||||
error instanceof z.ZodError
|
||||
? {
|
||||
issues: error.issues.slice(0, 8).map(({ path, code, message }) => ({
|
||||
path,
|
||||
code,
|
||||
message: message.slice(0, 160),
|
||||
})),
|
||||
}
|
||||
: { reason: String(error.message).slice(0, 256) };
|
||||
return new FactFailure(
|
||||
"invalid_provider_output",
|
||||
"Native IR source does not match the declared format",
|
||||
details,
|
||||
);
|
||||
};
|
||||
|
||||
export const readNativeFacet = (fn, name, label, parse) => {
|
||||
try {
|
||||
const reason = fn[`${name}_error`];
|
||||
if (reason !== undefined) {
|
||||
requireFact(
|
||||
fn[name] === undefined,
|
||||
`Native ${label} is both present and unavailable`,
|
||||
);
|
||||
return err(
|
||||
new FactFailure(
|
||||
"capability_unavailable",
|
||||
`Native ${label} facet is unavailable`,
|
||||
{
|
||||
native_error: reason.slice(0, 1024),
|
||||
native_error_truncated: reason.length > 1024,
|
||||
},
|
||||
),
|
||||
);
|
||||
}
|
||||
if (fn[name] === undefined)
|
||||
return err(
|
||||
new FactFailure(
|
||||
"capability_unavailable",
|
||||
`Native ${label} facet was not captured`,
|
||||
{ availability: "not-captured" },
|
||||
),
|
||||
);
|
||||
return ok(parse(fn[name]));
|
||||
} catch (error) {
|
||||
return err(formatFailure(error));
|
||||
}
|
||||
};
|
||||
@@ -1,682 +0,0 @@
|
||||
# Native recovered IR experiment
|
||||
|
||||
Stock Kuna's diagnostic text omits CFG successors and does not encode operand or
|
||||
phi edge identities. `kuna-native-ir.patch` is a read-only exploratory exporter
|
||||
against [Kuna v1.833](https://github.com/Noelo-Lab/kuna/tree/fe01d6294eb0d1053ce343c8943646df88e143fc).
|
||||
It is not an upstream release or an integrated REA capability. No upstream commit,
|
||||
PR or merge is being made.
|
||||
|
||||
`decompile --json --atomic-facts` captures the live function that produced C.
|
||||
The default path does not collect the facet. Graph records include ordered CFG
|
||||
edges, native value identities, indexed operands, outputs and phi input bindings
|
||||
to particular incoming edges. Duplicate instruction origins or predecessor blocks
|
||||
do not collapse identities. Values are local to one snapshot. Constant offsets
|
||||
remain hexadecimal u64 address-space offsets, with width and annotation status
|
||||
separate; they are not arbitrary-width computed bitvectors.
|
||||
|
||||
`native-ir.mjs` owns validation, immutable indexes and selected reads.
|
||||
`control-dependence.mjs` derives control dependence using immediate
|
||||
postdominators of the reversed graph, with linear retained graph state and a
|
||||
bounded work count. It treats zero-successor recovered blocks as exits and
|
||||
considers finite paths only. Exit-unreachable blocks, indirect-branch operations,
|
||||
missing definitions, runtime successors and memory alias limitations stay
|
||||
explicit. This is not a proof of feasible runtime paths.
|
||||
|
||||
The shared API adds `values`, `blocks` and `relations(kind:"controllers")`.
|
||||
Diagnostic SSA sessions report these missing capabilities explicitly. Native
|
||||
definition/use edges include operand and phi bindings; diagnostic printed
|
||||
occurrences retain their original, weaker semantics. CLI, SDK and both MCP
|
||||
adapters use the same owner. Ordinary source summaries are compact; `evidence`
|
||||
retains the exact command argv and separate display argv, complete output and
|
||||
digest. Earlier trials substituted display paths in `arguments`; they must not
|
||||
be retroactively claimed to have captured exact argv.
|
||||
|
||||
To reproduce, check out the pinned upstream revision in a disposable directory
|
||||
and apply this patch with `git apply`. Reuse already compiled `.sla` files from
|
||||
the corresponding stock release under the same relative `specs/` paths; do not
|
||||
recompile specs or replace existing unrelated files. Build from the REA root:
|
||||
|
||||
```sh
|
||||
env CARGO_INCREMENTAL=0 CARGO_PROFILE_RELEASE_DEBUG=0 \
|
||||
CARGO_PROFILE_RELEASE_OPT_LEVEL=1 CARGO_PROFILE_RELEASE_LTO=false \
|
||||
CARGO_PROFILE_RELEASE_CODEGEN_UNITS=16 \
|
||||
python3 experiments/analysis-efficiency/resource-run.py --observe-payload-exit -- \
|
||||
cargo build --manifest-path /absolute/kuna/decompiler/Cargo.toml \
|
||||
--locked --release -j1 -p kuna-cli --features embed-specs
|
||||
```
|
||||
|
||||
Select that executable through `REA_KUNA_COMMAND` and explicitly pin its actual
|
||||
SHA-256 through `REA_ATOMIC_NATIVE_IR_ENGINE_SHA256`. Native IR requires embedded
|
||||
specs and rejects a caller-supplied `--sleighpath`; the provider creates a fresh
|
||||
owned engine cache per lease. The executable digest therefore covers its embedded
|
||||
assets, without reusing an unvalidated external spec cache. The current prototype
|
||||
uses optimization level 1, not the stock release's assumed optimization profile;
|
||||
comparisons must retain this difference or use the same engine in all lanes.
|
||||
|
||||
Native extraction bounds blocks/operations at 12,000, values at 36,000, input and
|
||||
CFG edges at 50,000 and display text at 2 MiB, before the 8 MiB transport bound.
|
||||
Capacity overflow makes this facet unavailable, preserving C and a reason. The
|
||||
provider retains at most two 8 MiB native indexes (two 2 MiB diagnostic indexes)
|
||||
and 64 MiB of complete source captures per lease. The outer observer still
|
||||
enforces one CPU, 4 GiB hard memory and early termination at 3 GiB.
|
||||
|
||||
Real Linux zlib verification produced 3,860 operations, 5,898 values, 472 blocks,
|
||||
779 CFG edges and 790 phi operations. Its 5,533,649-byte native JSON has no omitted
|
||||
definitions or reported producer issues. C is byte-identical to the previously
|
||||
captured stock 1.833 JSON (52,431 characters), and line mappings are identical.
|
||||
The CLI composition independently discovers the string, owning function, store
|
||||
and its two local controllers. Its observed cold session takes 9.191 seconds,
|
||||
8.887 process-family CPU seconds and 198.9 MiB kernel memory peak, with verified
|
||||
cleanup and zero OOM kills. This scripted lane is not a real-model cost result.
|
||||
|
||||
Targeted JS checks cover malformed/foreign identities, dangling relations, phi
|
||||
order, exact hex offsets, immutable facts, loop/exit coverage, resource failure,
|
||||
and all 64 four-node DAGs against an independent path-enumeration oracle.
|
||||
Upstream `check-spec` passes. The full upstream corpus, stages and workspace gates
|
||||
have not run; this patch is not declared upstream-ready. The model comparisons
|
||||
and subsequent boundary findings below do not establish wider provider parity.
|
||||
|
||||
The value-origin model comparison exposed a representation boundary: optimized
|
||||
recovered blocks can absorb a machine predecessor, so an IR edge is not an actual
|
||||
branch instruction target. The experimental API now names operation locations
|
||||
`origin_address`, their selector `origin_addresses`, and block locations
|
||||
`origin_start`; printed CFG targets use `target_origin`. These replace the old
|
||||
ambiguous fields without aliases. Historical captures retain their pinned old
|
||||
contracts. Raw upstream JSON is preserved unchanged in Evidence.
|
||||
|
||||
The added `instructions` method batches selected machine-address windows, using
|
||||
Kuna's existing [disassemble CLI](https://github.com/Noelo-Lab/kuna/blob/fe01d6294eb0d1053ce343c8943646df88e143fc/decompiler/crates/kuna-cli/src/disassemble.rs),
|
||||
with exact bytes, decoded text, extent, coverage and per-window errors. Kuna
|
||||
already has focused agent-oriented read surfaces; its port ancestry does not
|
||||
establish which primitives are missing. No assembly decoding or branch-target
|
||||
parser is duplicated in the MCP adapter.
|
||||
|
||||
Windows are capped at 4,096 decoded rows and 16 windows per call, sharing the
|
||||
8 MiB capture and 64 MiB retained-source budgets. A real maximum-row CLI test
|
||||
returns 4,096 rows in an 842,106-byte source, taking 1.227 family wall seconds,
|
||||
1.042 CPU seconds and 112.0 MiB kernel memory peak. Both real stock and patched
|
||||
MCP tests preserve good windows alongside an unmapped-address failure and report
|
||||
the actual `JA 0xe26d` / `JMP 0xd9c9` bytes. The representation revision had 33
|
||||
targeted tests; its subsequent real-model trial is recorded below.
|
||||
|
||||
That fresh trial now verifies the missing machine path correctly; see
|
||||
[actual task costs](trials.md). The SQLite task then exposes a separate missing
|
||||
primitive: name/alias lookup across distinct callable entries. `functions` now
|
||||
reuses the production inventory parser, preserving every matching entry, import
|
||||
stub ambiguity, original aliases, clipped extents and unknown runtime binding.
|
||||
The inventory is read/validated once per lease, and query/field/page selection
|
||||
returns only the requested view. Batch xref and instruction reads share one
|
||||
partial-outcome owner, retaining good observations and source IDs on failures.
|
||||
|
||||
A real `sqlite3VdbeExec` capture at `0xea6d0` also exceeds the graph export's
|
||||
operation capacity while producing 126,519 characters of valid C. The old adapter
|
||||
incorrectly rejects the entire capture. The revised owner validates the function
|
||||
envelope and input identity independently, then models graph availability as a
|
||||
facet result. Code and its language remain available under the same handle;
|
||||
graph counts stay unknown and graph reads return their reason and Evidence ID.
|
||||
The real MCP regression reads 3,654 code lines after the graph failure and verifies
|
||||
that `sqlite3Fts3ErrMsg` has distinct entries `0x268f0` and `0x5efc0`. Native graph
|
||||
capacity remains a real limitation, not a resolved scalability claim.
|
||||
|
||||
`verify-cache.mjs` measures a real cold/warm pair within one open lease. Cold zlib
|
||||
snapshot reading takes 6,934.4 ms / 6.928 family CPU seconds; warm reading takes
|
||||
67.7 ms / 0.068 CPU seconds and preserves the exact snapshot/Evidence ID. The
|
||||
cumulative kernel memory peak is 275.1 MiB for both phases; the warm number is an
|
||||
upper bound on its isolated phase peak. This is an API-only measurement, with no
|
||||
model token result or cross-session cache claim. Targeted tests now total 38,
|
||||
including late cancellation retaining partial facts and their provenance.
|
||||
|
||||
A subsequent real boundary sweep found a different loss of usable facts:
|
||||
`sqlite3Pragma` (`0xb90e0`) and `sqlite3Parser` (`0xb3d00`) stay below the row
|
||||
limits but overflow the provider's 8 MiB stdout budget. Both public snapshots
|
||||
fail, losing C as well as IR. A separate diagnostic capture with a stated 32 MiB
|
||||
observer limit establishes the exact producer sizes; it is not relabeled as a
|
||||
successful public-provider workflow. The old engine is preserved locally by its
|
||||
`7d357e75…` digest for historical reproduction.
|
||||
|
||||
| Function | Previous pretty JSON bytes | Revised source bytes | C characters | Revised graph |
|
||||
| --------------- | -------------------------: | -------------------: | -----------: | ----------------------------------------------- |
|
||||
| `sqlite3Parser` | 8,492,810 | 4,526,811 | 50,508 | All 5,164 operations retained |
|
||||
| `sqlite3Pragma` | 16,728,407 | 136,501 | 67,300 | Explicitly unavailable at encoded-byte capacity |
|
||||
|
||||
The structural fix is at the producing serialization boundary. Atomic output
|
||||
uses the existing record producer and shared compact ASCII-safe writer.
|
||||
`--atomic-output-bytes` takes an explicit encoded-document budget (default 8 MiB);
|
||||
the provider passes the same bound it supervises. The writer checks each encoded
|
||||
write before growing its bounded buffer, accounting for escaping and newline.
|
||||
If necessary, the document drops only the graph and records an exact facet
|
||||
failure. C, line mappings and metadata survive under the same Evidence. If even
|
||||
the base document cannot fit, the command fails rather than clipping it or
|
||||
repeating the analysis. Ordinary non-atomic JSON keeps its previous formatting.
|
||||
Row limits alone never promise that the whole encoded graph fits.
|
||||
|
||||
The rebuilt local engine is SHA-256
|
||||
`2290e8b13caabe30aab6832ab163d92925b2e652eebf7fc0cdbe65b889cb251b`
|
||||
(same pinned source/release profile plus the revised patch). Real Code Mode MCP
|
||||
verification succeeds for both functions. The preserved C and line mappings are
|
||||
identical to the earlier full diagnostic captures; the Parser graph is also
|
||||
deep-equal, and retained source hashes verify. Pragma's unknown graph counts and
|
||||
failed graph reads retain the correct source ID. No limit was increased.
|
||||
|
||||
The first successful revised MCP check took 13.721 family wall seconds, 13.461
|
||||
CPU seconds and 326.9 MiB kernel family peak, with verified cleanup and no OOM.
|
||||
This is a two-function API regression, not a model-cost comparison or a claimed
|
||||
analysis speedup. Native writer/escaping checks (14) and graph-only fallback (1)
|
||||
pass in the release test binary; the shared JS tests (38), SDK ownership probe,
|
||||
lint and patch apply/reverse checks pass. Full upstream corpus/stages/workspace
|
||||
gates still have not run. The first MCP launch used a stale pre-format-build hash
|
||||
and was correctly refused; the retained successful retry uses the digest above.
|
||||
|
||||
The `bytes` primitive reuses Kuna's existing `read --as data --json` protocol.
|
||||
Unlike instruction decoding, it asks for an exact byte range. It disables the
|
||||
global `listing` / `fast_funcdisc` passes for this data-only query; instructions
|
||||
retain their existing profile. A native probe returns exactly equal JSON with
|
||||
and without those flags. Its 484 / 290 ms timings share a workspace and run in
|
||||
that order, so they do not establish an isolated speedup. Native/source ownership,
|
||||
freshness, decoded-window admission and partial batches remain shared.
|
||||
|
||||
Real public MCP checks pass in native Code Mode, native direct and stock 1.833
|
||||
Code Mode: a known 53-byte string, a 4-of-16-byte read at a mapped-image boundary,
|
||||
an unmapped address, same-source cache reuse and the 64 KiB maximum. Both adapter
|
||||
catalogs validate after SDK conversion. The byte parser rejects wrong subjects,
|
||||
contradictory duplicated hex, invalid extents and rows, and preserves exact u64
|
||||
addresses. Completeness follows examined byte coverage rather than the native
|
||||
truncation flag. It never claims these loaded-image bytes are live memory or
|
||||
proves an original-file mapping. Targeted JS tests now total 41.
|
||||
|
||||
## Structured call relations (wire version 2)
|
||||
|
||||
The next observed gap was input zero of CALL: a native `fspec` annotation is an
|
||||
internal reference, not a code address. The read-only `atomic_calls.rs` exporter
|
||||
now uses the existing `FuncCallSpecs` owner for recovered entry/name, model flags,
|
||||
argument passing storage and stack offset. A single operation-to-spec index
|
||||
avoids a linear lookup for every call. Input slots and SSA IDs remain separate
|
||||
from passing locations. Finalized storage is admitted only at matching arity;
|
||||
locked prototype storage is the alternative. Shifted inputs, placeholders,
|
||||
missing specs, unavailable storage and unresolved indirect entries remain
|
||||
explicit. There is no display-text parser or original-source ordinal claim.
|
||||
|
||||
`calls` provides ID/origin filtering, field projection and paging through the
|
||||
shared SDK/CLI/MCP owner. Callee `values(is_input:true)` exposes retained input
|
||||
values for exact storage candidate matching. Source schemas, unique call
|
||||
ownership, call coverage, operand indexes/values and output identities are
|
||||
validated before exposing an index; a bad or old-version graph preserves valid
|
||||
C. Machine CALL instructions remain the independent check of a recovered edge.
|
||||
|
||||
Engine `64c6b3fe5737056fc29c10b0aad6c7cbd489ad7d17bd1726141b77bdcf5e15b7`
|
||||
uses the same upstream revision, embedded specs and one-job build profile.
|
||||
The rebuild took 317.090 family wall seconds, 315.808 CPU seconds and a 2,145.695
|
||||
MiB kernel process-family memory peak (including charged file cache), OOM 0.
|
||||
The previous compact exporter engine is preserved by its SHA-256 locally.
|
||||
|
||||
Real SQLite MCP verification passes both adapters, including a composed Code
|
||||
Mode join: caller `0x65211 -> 0x64060`, argument passing storage
|
||||
`register:0x38/8`, one matching callee input, and a bounded definition slice from
|
||||
the context updates to that input. Formatter calls `0x5f031`/`0x5f062` point to
|
||||
`0x2adc0`/`0x27e50`; their fspec offsets remain `0x10000`/`0x10001`.
|
||||
All three functions' C, ordinary metadata/line mappings, blocks, operations,
|
||||
values and existing coverage equal retained earlier captures. Those earlier
|
||||
sources were pretty-printed: encoded-size changes are not a call-exporter
|
||||
performance result. The structured call counts are 8, 33 and 2 respectively.
|
||||
|
||||
The Code Mode verifier took 7.872 family wall / 7.573 CPU seconds, 297.1 MiB;
|
||||
the direct verifier took 7.771 / 7.499 seconds, 278.7 MiB. These differ in assertion
|
||||
calls and warm file-cache state, so they are verification receipts, not a
|
||||
controlled adapter speed comparison. Both verified cleanup and OOM 0. The
|
||||
existing real zlib controller/phi/instruction verifier also passes with this
|
||||
engine (10.992 wall / 10.795 CPU seconds, 285.7 MiB, cleanup verified, OOM 0).
|
||||
`check-spec` passes in its default lenient mode; full upstream gates still have
|
||||
not run. The independent cross-function question and acceptance criteria are in
|
||||
[call-oracle.md](call-oracle.md), pinned before model evaluation.
|
||||
|
||||
The strengthened verifier additionally reads all five retained indirect calls in
|
||||
this parser and checks their unknown targets. Machine bytes `ff5328` at
|
||||
`0x644eb` independently describe a register-indirect CALL. The Code Mode rerun
|
||||
passes (16.628 wall / 8.239 CPU seconds, 300.5 MiB); native test compilation was
|
||||
also active on the same CPU, so its wall time is not an isolated performance
|
||||
sample. The source checks pass 45 JS tests across both prototype directories and
|
||||
two Python ledger tests; changed source lint and formatting pass.
|
||||
|
||||
A current coverage limit remains: call facts are part of the recovered IR
|
||||
facet. A function that exceeds graph capacity also loses these call records,
|
||||
even if a selected call-target view could fit. This is a reason to explore
|
||||
independent fact facets; it must not be mistaken for proof of no calls or full
|
||||
coverage on large functions. No full provider parity claim is made.
|
||||
|
||||
The current producer also passes fourteen native JSON-format tests. The initial
|
||||
command compiled unrelated integration targets and a non-embedded CLI build,
|
||||
which replaced Cargo's mutable `target/release/kuna`. Its 181.103 wall / 172.430
|
||||
CPU seconds, 651.6 MiB peak and OOM 0 describe that broader compilation, not just
|
||||
JSON test execution. Future focused commands should select `--bin kuna` and
|
||||
`--features embed-specs`. The original embedded build restores from Cargo cache
|
||||
and has the same `64c6b3f...` digest.
|
||||
|
||||
The first model launch after those tests correctly fails at `open_binary` with
|
||||
`engine_profile_mismatch`; it produces no target Evidence and is not a completed
|
||||
answer. Its rejected-trial costs are retained in `trials.md`. The producing
|
||||
boundary is now fixed by preserving an independent executable copy in a directory
|
||||
named by the full SHA-256, outside Cargo outputs. The model harness checks the
|
||||
explicit experimental engine pin before copying credentials or launching a model,
|
||||
and verifies the same bytes after the task. Changed/missing engines reject the
|
||||
comparison while retaining costs. Other provider trials keep this field unknown.
|
||||
Three regression tests cover missing/malformed pins, replacement and removal;
|
||||
a real driver invocation with a wrong pin rejects before output/profile creation.
|
||||
|
||||
## Shallow outgoing references without decompilation
|
||||
|
||||
Kuna already supplies native `xrefs --from`, including exact-instruction and
|
||||
owning-function queries. The shared `KunaReferenceFiles.ts` boundary now owns
|
||||
both directions, validates direction-dependent peer fields and query ownership,
|
||||
and preserves exact u64 endpoints and source-row occurrences. The old
|
||||
incoming-facts-only parser is removed; both experimental consumers use the same
|
||||
named query contract. Production incoming address-only queries keep their
|
||||
existing projection and bounded cache semantics.
|
||||
|
||||
The experimental `references` method adds direction, explicit address/function
|
||||
scope, native kind filtering and fields. Incoming function scope is rejected;
|
||||
unknown outgoing function applicability, unattributed/excluded rows and unresolved
|
||||
indirect/discovery/runtime coverage remain explicit. Native source cache keys
|
||||
cover result-affecting query choices while scope, fields and pages reuse the same
|
||||
source. `reference-reads.mjs` owns protocol selection/projection; the resource
|
||||
owner and CLI/MCP adapters do not reinterpret native endpoint fields. Empty
|
||||
field selection now requests the existing always-retained identities, rather
|
||||
than being rejected without a format/resource reason.
|
||||
|
||||
Real outgoing CLI probes return 396 VM call-reference rows in 205,432 bytes,
|
||||
versus one row/737 bytes for selected call site `0xf238b`. Four sequential probes
|
||||
(VM function/instruction and wrapper function/instruction) take 6.772 family wall
|
||||
/ 6.613 CPU seconds and 120.0 MiB peak. Each command takes about 1.6–1.7 seconds;
|
||||
these are native metadata-query measurements with warm uncontrolled file caches,
|
||||
not real-model speed/token results.
|
||||
|
||||
Independent GNU objdump decoding over the exported VM symbol (`0xea6d0`, size
|
||||
`0x7e7e`) confirms all 396 `(origin,target)` pairs, 171 distinct exact targets,
|
||||
and 16 calls to `0x296c0`. It also finds 21 indirect calls, outside these
|
||||
known-target counts. In the wrapper, nine lifted machine call references compare
|
||||
with eight retained IR calls: machine `0x6534b -> 0x29370` is an omitted
|
||||
`__stack_chk_fail@plt` call. Its absence from optimized IR/C is not absence from
|
||||
the executable. No native analysis or C behavior is changed to repair that
|
||||
representation difference.
|
||||
|
||||
The real Code Mode MCP verifier passes paging, identity-only fields, scope-cache
|
||||
reuse, exact sites, unknown applicability and local aggregation. It also captures
|
||||
the VM anew: C is available, IR fails with `atomic operation capacity exceeded`,
|
||||
and shallow references still work. This verifier takes 48.142 family wall /
|
||||
47.803 CPU seconds, 610.1 MiB peak; VM decompilation/source capture itself takes
|
||||
35.945 seconds. These are different requested capabilities, not an equivalent
|
||||
workflow speedup claim. The direct verifier passes the same reference contract
|
||||
without that VM decompilation (12.001 / 11.759 seconds, 256.8 MiB). Both verify
|
||||
cleanup and OOM 0. An earlier direct-verifier attempt incorrectly expected target
|
||||
metadata on an SDK input-validation error; the corrected verifier respects the
|
||||
SDK rejection boundary. That failed assertion receipt is retained.
|
||||
|
||||
The exact focused boundary suites pass 38 tests, with a fresh strict production
|
||||
build; changed-test strict typechecking and 45 experimental JS tests also pass.
|
||||
Full lint/module boundaries and changed-file formatting pass. Repository lint
|
||||
warnings remain outside this change. The focused build/test scope peaks at
|
||||
3,030.6 MiB, below the 3 GiB early-stop line, with OOM 0. Two full `check:fast`
|
||||
attempts (including serial execution and Go runtime memory controls) reach that
|
||||
early-stop budget and are terminated before OOM; they are not reported as passed.
|
||||
The changed-test typecheck uses 70.1 MiB and standalone full lint 919.0 MiB.
|
||||
Full-repository typechecking and full upstream Kuna gates remain unverified.
|
||||
The [next question/oracle](reference-oracle.md) are committed before model trials.
|
||||
|
||||
## Independent call and boundary-value facets
|
||||
|
||||
Graph schema 3 removes embedded call records. An independent `atomic_calls`
|
||||
version-1 facet now carries every retained block-member CALL/CALLIND, target-input
|
||||
SSA identity, origin/block/opcode/dead status, call specs and boundary values.
|
||||
`atomic_values.rs` owns common native value interpretation. The call scan walks
|
||||
native block membership without allocating an operation vector for every block;
|
||||
its own call/input/value/text budgets do not depend on the 12,000-op graph cap.
|
||||
The default native path still does not collect either facet and C analysis phases
|
||||
are unchanged. All captures occur after the same C rendering.
|
||||
|
||||
The bounded serializer retains fitting facets independently and preserves ordinary
|
||||
C/metadata. It tries the full document, graph without calls, calls without graph,
|
||||
then the base record with both reasons. This keeps existing graph availability
|
||||
when both facets individually fit but their combination does not. Facet contents
|
||||
are moved for these bounded alternatives rather than cloned; no reanalysis or
|
||||
clipped collection labeled complete is introduced.
|
||||
|
||||
The JS owner validates each facet independently with shared ID/value contracts
|
||||
and row-wise admission. When both parse, call identities, origins, operands,
|
||||
outputs and boundary values must agree with the graph owner. Bad call facts do
|
||||
not erase a valid graph; a bad/oversized graph does not erase valid calls. Code
|
||||
survives either failure. `values` uses the complete retained graph when present,
|
||||
otherwise the explicitly scoped call-boundary set; absent IDs in that set do not
|
||||
prove absence from the native function. Defining operation identities may remain
|
||||
unreadable, and source signatures, source parameter ordinals, aliases and runtime
|
||||
binding stay unknown. `is_dead` is always retained on call rows, like an
|
||||
instruction decode status. CLI, SDK and both MCP adapters share these views.
|
||||
|
||||
The new preserved engine is
|
||||
`e7d3f0cc8d8aae814bd6a2625a567d6aaad96c3fe2405a4f08dc8e065b851588`.
|
||||
Its full rebuild took 315.815 family wall / 315.305 CPU seconds, 2,119.4 MiB
|
||||
kernel process-family peak. Four focused native atomic-JSON tests pass, including
|
||||
both overflow directions, both-overflow C preservation and graph preference when
|
||||
the combined document is too large (95.981 / 95.631 seconds including test-target
|
||||
compilation, 626.8 MiB). The final CLI rebuild after tests is 26.203 / 25.952
|
||||
seconds, 492.7 MiB. It is copied outside Cargo outputs and pinned by full digest.
|
||||
All these runs verify cleanup and OOM 0; full upstream gates remain unexecuted.
|
||||
|
||||
Real public MCP verification keeps VM C available, reports graph operation
|
||||
capacity failure, and now exposes 416 calls plus 1,294 boundary values. The call
|
||||
scan examines 21,349 operations over 1,229 blocks. It preserves 395 direct known
|
||||
targets and 21 unresolved indirect calls; all direct pairs match independent
|
||||
machine pairs except the omitted `0xf2493 -> 0x29370` stack-check call. All 21
|
||||
indirect origin addresses also match independent disassembly. Selected
|
||||
`0xf115c`/`0xf238b` calls target `0x296c0`; their fspec inputs remain separate
|
||||
annotation values. Repeated snapshots reuse the same handle and source, and an
|
||||
explicit unavailable graph read fails without hiding these retained facts.
|
||||
|
||||
The VM source grows from 273,703 to 793,791 bytes to retain this new data; the
|
||||
wrapper grows from 202,206 to 208,083 bytes. Their ordinary records, including C,
|
||||
variables and line mappings, are identical to the prior capture. The wrapper's
|
||||
blocks/operations/values/coverage also remain identical. These are fidelity and
|
||||
source-volume observations, not model-token or speed results.
|
||||
|
||||
The real reference/VM verifier passes in 49.471 wall / 49.130 CPU seconds,
|
||||
615.8 MiB. The existing cross-function call/input verifier passes in 8.377 /
|
||||
8.094 seconds, 292.9 MiB; encoded Pragma/parser C/graph checks pass in 12.403 /
|
||||
12.195 seconds, 335.0 MiB. All verify cleanup and OOM 0. The updated contract
|
||||
suite passes 47 JS tests, including independent facet failure, cross-facet
|
||||
conflicts, input-zero identity and boundary-only value reads. `check-spec` passes
|
||||
in default lenient mode, lint and changed-file formatting pass, and the patch
|
||||
applies forward to the pinned upstream revision and reverses from the clone.
|
||||
Full repository typechecking retains its previously recorded resource limitation.
|
||||
The [boundary question/oracle](boundary-oracle.md) are fixed before model runs.
|
||||
|
||||
The subsequent real-model pair passes that oracle in both calling modes. Each
|
||||
uses one VM analysis and the same retained function record; neither invents a
|
||||
complete graph from the available call-boundary values. Source digests, target
|
||||
and preserved-engine pins, selected machine checks, cleanup and OOM-zero receipts
|
||||
verify. See [whole-task costs](trials.md): the capability survives real model use,
|
||||
but a single mixed-context-cost pair does not establish general efficiency.
|
||||
|
||||
## Compact counts from the validated call owner
|
||||
|
||||
The following API-only change keeps the preserved engine and native source
|
||||
format unchanged. Snapshot call metadata includes counts by recovered opcode
|
||||
and known/unknown native target, plus call coverage. The owner computes these
|
||||
once after validation, before the shared CLI/SDK/MCP views. Counts include dead
|
||||
rows and do not infer target knowledge from CALL versus CALLIND. Available empty
|
||||
sets have zeros; unavailable sets have no counts. Selected call reads do not
|
||||
change this whole-facet summary or require a new analysis.
|
||||
|
||||
The focused suite passes 49 tests, including known indirect/unknown direct
|
||||
targets, dead rows, immutable summaries and empty/unavailable distinction.
|
||||
Both real MCP adapters verify the wrapper's eight direct calls and parser's
|
||||
28 direct/five unresolved indirect calls while preserving the existing
|
||||
cross-function checks and validating advertised schemas after SDK conversion.
|
||||
Code Mode verification takes 29.263 family wall / 21.548 CPU seconds, 283.9 MiB
|
||||
kernel family peak; direct takes 32.893 / 22.781 seconds, 280.6 MiB. Both have
|
||||
verified cleanup and OOM 0. These are validation runs, not a controlled speed
|
||||
comparison with previous verifier timings. An owner replay of the prior
|
||||
digest-verified VM source yields the expected 395/21 distribution; it is not
|
||||
relabeled as a fresh provider or model run. The identical boundary question and
|
||||
oracle are retained for subsequent real-model comparison.
|
||||
|
||||
That comparison now passes in both modes. Both consume the compact counts and
|
||||
read just the two selected calls, reusing one VM analysis. All native VM function
|
||||
records match the preceding pair and all retained sources/digests verify. Direct
|
||||
has lower observed task costs than Code Mode in this sample; the before/after
|
||||
results remain mixed. [trials.md](trials.md) records cumulative/cached/uncached
|
||||
tokens, model responses, elapsed time, CPU and process-family memory, including
|
||||
the slower Code Mode result. No native engine or analysis-speed improvement is
|
||||
claimed for this shared owner metadata change.
|
||||
|
||||
## Tagged cleanup ownership regression
|
||||
|
||||
Source inspection finds that the experimental owner awaits but ignores
|
||||
`KunaTargetCommands.close()`'s tagged `Result`. On error it would clear its state
|
||||
and let `AtomicSession` drop the owner while claiming a successful close. The
|
||||
SDK verifier reproduces this before the fix: an injected `ProviderCleanupError`
|
||||
does not reject close. Its fallback cleanup still removes the real owned state;
|
||||
the original failing result is retained as `cleanup-before.json`.
|
||||
|
||||
The owning boundary now uses the existing `requireResult` projection before
|
||||
closing its cache root or clearing state. The existing session queue therefore
|
||||
retains the owner on failure. Both close and target replacement report
|
||||
`cleanup_incomplete`, preserve the staged file identity and original lease, and
|
||||
succeed on retry; the file is then absent. MCP's open/close error adapter now
|
||||
preserves context and resource details rather than dropping that evidence. The
|
||||
CLI already propagates the thrown error as a failed command. No alternate
|
||||
cleanup owner or downstream retry workaround is added.
|
||||
|
||||
`verify-cleanup.mjs` runs against real provider reads and resources, injecting
|
||||
one tagged failure per close/replacement case. `verify-cleanup-mcp.mjs` injects
|
||||
one explicit test-only failure in an isolated server, then verifies public
|
||||
Code Mode and direct error metadata and successful retry. These are controlled
|
||||
boundary failures, not a claim that the real engine failed to stop. The SDK
|
||||
check takes 1.730 family wall / 1.567 CPU seconds, 130.4 MiB kernel peak; MCP Code
|
||||
Mode takes 1.931 / 1.639 seconds, 200.0 MiB; direct takes 1.730 / 1.489 seconds,
|
||||
192.8 MiB. All verify process cleanup and OOM 0. The original failing verifier
|
||||
takes 1.226 / 0.985 seconds, 122.7 MiB and also verifies fallback cleanup/OOM 0.
|
||||
The 49 focused JS tests, lint and formatting pass. This lifecycle-only change
|
||||
follows the recorded model pairs; no new model token or speed claim is made.
|
||||
|
||||
## Callable bounded definition projection
|
||||
|
||||
The next native exporter retains a version-1 `atomic_projection` independently
|
||||
of graph version 3 and call version 1. One checked native membership scan owns
|
||||
call roots and an optional index of at most 65,536 operation keys; exceeding the
|
||||
index budget disables the projection without discarding calls. Full-graph row
|
||||
capacity is checked before rendering its rows. Full and selected facts share
|
||||
one operation/value normalizer, including exact identities, indexed operands,
|
||||
outputs, phi predecessor edges, flags and display. Default non-atomic analysis
|
||||
and C generation are unchanged.
|
||||
|
||||
All recovered CALL/CALLIND operations are retained as projection roots, followed
|
||||
by backward operand definitions in breadth-first order. The projection bounds
|
||||
4,096 operations, 12,288 values, 16,384 inputs and 2 MiB text. Too many roots or
|
||||
another admission failure makes the facet explicitly unavailable. Omitted
|
||||
definitions retain exact IDs and operation-budget/outside-membership reasons.
|
||||
A fitting full graph omits its redundant projection; an encoded-overflow graph
|
||||
restores the fitting projection. Facets are moved through bounded alternatives,
|
||||
and a valid C/call record survives projection failure. There is no repeated
|
||||
decompilation or clipped collection labeled complete.
|
||||
|
||||
The JS boundary shares operand/value validation and query indexes between full
|
||||
and sparse views. It checks root reachability, exact frontier membership,
|
||||
call/value agreement and phi metadata without inventing a CFG. In the absence
|
||||
of a full graph, conflicting projection facts do not erase independently valid
|
||||
calls. `snapshot.facets.operations` and `facets.values` expose their scopes;
|
||||
selected definitions/uses and slices work over the retained set. Uses remain
|
||||
selected-only, CFG/controller queries retain their graph error and Evidence ID,
|
||||
and slice missing definitions keep their native reasons. A missing projection
|
||||
preserves the original graph failure instead of hiding it behind a generic
|
||||
not-captured error. All adapters use this owner.
|
||||
|
||||
Coverage returns counts instead of global omitted-ID inventories, at most eight
|
||||
160-character issue summaries, full issue counts and omission metadata. Complete
|
||||
native coverage/frontier details remain in source Evidence. Native operation
|
||||
rows retain `is_dead` with their identities; constant summaries add `value_id`
|
||||
and `input_index`, preserving exact offset/width semantics. No original-signature,
|
||||
arbitrary-width computation, memory-alias or runtime proof is introduced.
|
||||
|
||||
Preserved engine SHA-256:
|
||||
`a0685486518e09d3c65f756325a7e9fed8df123a2a988aca2972a72335ce8edc`.
|
||||
The same one-job/embedded-spec build takes 309.805 family wall / 309.523 CPU
|
||||
seconds, 2,099.5 MiB kernel family peak. Six focused native JSON tests pass,
|
||||
including projection restoration and independent overflow (99.752 / 98.662
|
||||
seconds with test compilation, 626.9 MiB). The final CLI build reuses artifacts
|
||||
(0.320 / 0.146 seconds, 28.8 MiB), leaving the preserved digest unchanged.
|
||||
|
||||
Real SQLite VM public reads expose 4,096 operations, 7,469 values and 326
|
||||
unexpanded definitions, while the complete graph remains unavailable. Source is
|
||||
4,008,032 bytes, inside the unchanged 8 MiB public budget. Every selected operation
|
||||
and value equals the earlier 17.8 MB diagnostic capture; ordinary C/metadata and
|
||||
the complete call facet also equal it. An independent BFS over that retained
|
||||
full graph produces exactly the same operation order and frontier. The extended
|
||||
`measure-native-projection.py --verify-capture` reproduces these comparisons.
|
||||
This is producer-fidelity evidence, not independent machine/runtime equivalence.
|
||||
|
||||
Both public MCP adapters read the selected PHI definitions and slices from one
|
||||
VM source. Code Mode verifies unavailable controllers preserve the source ID.
|
||||
The corrected Code Mode verifier takes 52.706 wall / 47.550 CPU seconds,
|
||||
597.4 MiB peak; direct takes 39.696 / 39.418 seconds, 592.5 MiB. The first
|
||||
Code Mode verifier incorrectly expected INDIRECT definitions; independent source
|
||||
inspection shows MULTIEQUAL. Its 37.070 / 36.781 seconds, 593.3 MiB failed
|
||||
assertion is retained, then the verifier is corrected without changing native
|
||||
facts. These scripted runs are validation costs, not an adapter speed comparison.
|
||||
|
||||
Small-function cross-call checks pass (8.476 / 8.204 seconds, 293.6 MiB); all three
|
||||
complete function records equal the prior exporter. Encoded Pragma/Parser checks
|
||||
pass (13.712 / 13.507 seconds, 361.5 MiB): Pragma retains 3,627 projected operations
|
||||
after graph encoding overflow, while Parser keeps its full 5,164-operation graph
|
||||
without a duplicate projection. Zlib control/phi regression passes (10.893 /
|
||||
10.700 seconds, 286.3 MiB). Real CLI reads verify both selected PHIs and repeated
|
||||
snapshot/source identity (37.367 / 37.174 seconds, 527.1 MiB).
|
||||
|
||||
The real VM MCP cache benchmark observes cold snapshot wall 36,448.8 ms / family
|
||||
CPU 36.425 seconds, then warm 40.2 ms / 0.041 seconds with the identical handle
|
||||
and source. Cumulative family peak is 590.1 MiB, an upper bound on isolated warm
|
||||
phase memory. Its complete run is 37.667 / 37.492 seconds. This is same-lease
|
||||
capture reuse, not cross-session persistence or a model-token result; OS caches
|
||||
remain uncontrolled. All resource runs above verify cleanup and OOM 0.
|
||||
|
||||
The focused JS suite passes 53 tests; lint/format, default lenient `check-spec`,
|
||||
Python compilation and patch forward/reverse checks pass. Full upstream corpus,
|
||||
stages and workspace gates remain unexecuted; the patch is not upstream-ready.
|
||||
No whole-repository typecheck pass is claimed beyond the previously recorded
|
||||
resource limitation. The [PHI-format task/oracle](projection-oracle.md) are fixed
|
||||
before real-model launches. No model benefit is assumed from these smaller facts.
|
||||
|
||||
## Named reference discovery and input schemas
|
||||
|
||||
The first PHI model pair completes correctly but exposes a documentation boundary:
|
||||
direct help displays `facts_*` names while lookup only recognizes raw method
|
||||
substrings. A multi-name query returns no matches and causes a full-reference
|
||||
fallback. Registration and help now share one explicit name map, with canonical
|
||||
identity separate from presentation. Exact lists preserve caller order,
|
||||
deduplicate names and retain unmatched terms. Topic/all-contract lookup remain;
|
||||
ambiguous labels fail instead of choosing by catalog position.
|
||||
|
||||
The real client verifier also catches that plain `z.toJSONSchema` defaults to
|
||||
output semantics, making defaulted `offset`/`limit` falsely required in advertised
|
||||
input schemas. Reference generation now explicitly selects `io:input`. Converted
|
||||
schemas are frozen and weakly cached by schema identity; a replacement schema
|
||||
invalidates the conversion, while description/presentation is read separately.
|
||||
The documentation module removes the adapter rename pass. No target analysis,
|
||||
live-result caching or native-engine change is involved.
|
||||
|
||||
The shared suite passes 55 tests. `verify-discovery.mjs` checks both real SDK
|
||||
catalogs and reference responses without opening a target, compiles every catalog
|
||||
and all fourteen reference schemas with Ajv2020, and validates the same
|
||||
omitted-default requests against named reference/direct contracts. It proves the
|
||||
six-name query matches all requested contracts without a miss. The all-schema
|
||||
verifier takes 1.936 family wall / 1.760 CPU seconds, 138.8 MiB kernel peak,
|
||||
cleanup verified and OOM 0. The earlier six-schema probe takes 1.628 / 1.449
|
||||
seconds, 128.9 MiB. Its original 1.730 / 1.487 seconds, 128.1 MiB failed
|
||||
omitted-default check remains recorded. Six contracts occupy 10,835 Code Mode /
|
||||
10,360 direct bytes, versus 21,746 / 20,699 for all fourteen. Bytes do not prove
|
||||
token or time savings. The same PHI question/oracle are retained for the next
|
||||
actual model pair.
|
||||
|
||||
## Wide native operand relations
|
||||
|
||||
A retained-source replay exposes a whole-row cost: VM PHI
|
||||
`0xea998:14649` has 156 inputs and 156 definition edges. Its normalized old
|
||||
operation is 44,447 JSON bytes, and a default one-node slice returns 44,980 data
|
||||
bytes. Real QuickJS returns `output_budget` while preserving the source ID.
|
||||
This replays real producer facts through the shared parser/executor, without
|
||||
a fresh native analysis or model call (`wide-phi-before.json`). Explicitly
|
||||
requesting only `inputs` already fits; the gap is default size and lack of
|
||||
independently selected/paged operands, not inability to obtain the data at all.
|
||||
|
||||
`relations kind:operands` now owns slot-level paging and optional input-index
|
||||
selection. Snapshot + operation + input_index identifies the relationship;
|
||||
value identity, native definition and exact phi edge are separate fields. It
|
||||
retains constants, function inputs and nulls that definition edges cannot cover.
|
||||
Requested out-of-range slots of retained seeds are absent, while unretained
|
||||
seeds stay explicit/unknown. Seeds and selectors deduplicate in caller order;
|
||||
total counts filtered rows before paging. Full graph and projection share this
|
||||
owner; diagnostic SSA rejects the unsupported relation with source context.
|
||||
|
||||
Paging counts existing relation groups before materializing selected rows,
|
||||
avoiding a complete flattened relation copy. Native operation defaults are
|
||||
compact summaries with exact input/definition/constant counts; detailed fields
|
||||
remain selectable. Repeated value/block identity lookups deduplicate at their
|
||||
shared read boundary. A discriminated contract replaces the prior runtime-only
|
||||
relation refinement, so advertised schemas also enforce seed types and prohibit
|
||||
input_indices on other relationship kinds. No native rebuild, higher native
|
||||
capacity, new method or additional authority is involved.
|
||||
|
||||
The shared suite passes 57 tests; changed JS lint has zero warnings and formatting
|
||||
passes. Both real SDK catalogs and all reference schemas compile; selected
|
||||
operand inputs and invalid selector applicability agree with runtime validation.
|
||||
That discovery run takes 7.443 family wall / 4.928 CPU seconds, 141.8 MiB peak
|
||||
including charged file cache. The real Code Mode VM verifier passes ten calls
|
||||
in 54.327 / 47.069 seconds, 597.6 MiB peak. Its default slice fits, selected
|
||||
slots preserve exact native facts, and explicit all-input reads still work.
|
||||
Both runs verify cleanup and OOM 0. This first public run precedes the final
|
||||
identity-deduplication adjustment; subsequent direct/model reads cover it.
|
||||
Scripted bytes and costs are not model-token savings or an adapter speed claim.
|
||||
The subsequent direct verifier passes all ten calls, including duplicate-value
|
||||
deduplication, in 39.385 wall / 39.163 CPU seconds, 591.3 MiB peak, with cleanup
|
||||
verified and OOM 0. The different scripted read paths/host state do not isolate
|
||||
adapter costs. The [wide-operand oracle](operands-oracle.md) fixes the next seeded
|
||||
real-model task before launch; it is not a before/after of the PHI-format task.
|
||||
|
||||
## Static method index
|
||||
|
||||
The wide-operand pair exposes a discovery cost: Code Mode asks four times for
|
||||
help before using the same seven fact requests as direct. `reference view:index`
|
||||
now returns the same fourteen canonical/display names and descriptions without
|
||||
constructing schemas, return contracts or examples. `view:contracts` preserves
|
||||
the existing default and exact-name/topic behavior. Both presentations share one
|
||||
named documentation contract/owner and the API's registry; no analysis change or
|
||||
engine rebuild is involved. An index is static declared support, never live
|
||||
provider capability. Runtime facets remain authoritative.
|
||||
|
||||
The shared suite passes 58 tests, including a getter that proves an index does
|
||||
not access an unrequested input contract. Changed JS lint/format and both real
|
||||
SDK catalogs/reference schemas pass. The first discovery probe takes 5.284 family
|
||||
wall / 4.028 CPU seconds, 137.6 MiB peak, cleanup verified and OOM 0. Code Mode's
|
||||
index is 7,212 JSON bytes versus 24,010 for all contracts; direct is 7,240 versus
|
||||
22,939. These bytes are static discovery output, not demonstrated token/time
|
||||
savings. A subsequent focused check and real probe cover moving the input schema
|
||||
into the documentation owner and unifying presentation construction. The same
|
||||
wide-operand question/oracle are retained for a new real-model pair.
|
||||
|
||||
The final real SDK probe passes in 4.380 wall / 3.807 CPU seconds, 141.4 MiB peak,
|
||||
cleanup verified and OOM 0. Both subsequent real-model answers satisfy the
|
||||
unchanged oracle. Code Mode uses index plus one exact contract query, reducing
|
||||
its discovery lookups from four to two. Whole-task costs remain mixed; all
|
||||
attempts and limits are recorded in [trials.md](trials.md).
|
||||
|
||||
## Fact granularity versus computation granularity
|
||||
|
||||
The eight successful real-model VM captures using the preserved `a0685486...`
|
||||
engine each retain 4,008,032 bytes, including 126,519 bytes of reconstructed C
|
||||
that the selected-operand models never read. Fresh-lease native request wall
|
||||
ranges from 36.702 to 53.124 seconds. A separate failed first attempt hits the
|
||||
120-second timeout before one successful capture; its cause remains unknown.
|
||||
These are fresh private native caches, not controlled OS-cold samples. Same-lease
|
||||
warm source reuse remains separately measured above.
|
||||
|
||||
Source inspection explains the current computation boundary: the project core
|
||||
drives whole-function decompilation, renders C and extracts metadata, then
|
||||
collects independent call/graph/projection facets from the live function. The
|
||||
new slot atom reduces selected read/serialization scope, but does not implement
|
||||
selected-slot native analysis. The complete VM CFG is still unavailable and
|
||||
arbitrary definitions outside the retained call-rooted projection cannot be
|
||||
expanded by a fact read. Provider parity and general complex-software superiority
|
||||
remain unestablished.
|
||||
|
||||
At that checkpoint, pipeline/render/metadata/export costs were not separately
|
||||
profiled. The [subsequent native computation study](native-compute-study.md) now
|
||||
measures them: VM drive about 34 seconds, C rendering 0.22 seconds and projection
|
||||
0.03 seconds. Skipping C is a small opportunity for this workload. An offline
|
||||
lossless encoding probe also puts the earlier full 17.8 MB diagnostic record
|
||||
inside the unchanged 8 MiB transport. The [subsequent codec study](native-graph-transport-study.md)
|
||||
now implements and verifies this through both public experimental MCP adapters,
|
||||
including complete decoded Evidence and shared text-budget ownership. Actual
|
||||
guard/control model behavior is now checked in two real-model pairs. Model
|
||||
comparisons share the same computation/fact owners, avoiding a capability
|
||||
difference between adapters. The [switch extension](switch-study.md) adds an
|
||||
independent native table facet with exact labels and row/edge mapping, selected
|
||||
queries and a verified 2 MiB fallback that retains cases without a full graph.
|
||||
Its final engine is `a9ca8a8ac224936afb08a9d4be9c1225feb69473ad9f35764e8d1c4f04875b5b`;
|
||||
the fixed case/default model question is separate from the earlier guard pairs.
|
||||
@@ -1,49 +0,0 @@
|
||||
// One bounded decoded source view; the source owner revalidates bytes per read.
|
||||
import { createHash } from "node:crypto";
|
||||
import { readNativeGraphEvidence } from "./native-ir-contract.mjs";
|
||||
|
||||
export class NativeEvidenceReads {
|
||||
#graph;
|
||||
constructor(readSource) {
|
||||
this.readSource = readSource;
|
||||
}
|
||||
clear() {
|
||||
this.#graph = undefined;
|
||||
}
|
||||
async read(input) {
|
||||
const { text: original, source } = await this.readSource(input.evidence_id);
|
||||
let text = original,
|
||||
view;
|
||||
if (input.component === "recovered_ir") {
|
||||
const key = `${input.evidence_id}:${source.sha256}`;
|
||||
if (this.#graph?.key !== key) {
|
||||
const decoded = readNativeGraphEvidence(text);
|
||||
this.#graph = {
|
||||
key,
|
||||
...decoded,
|
||||
sha256: createHash("sha256").update(decoded.text).digest("hex"),
|
||||
bytes: Buffer.byteLength(decoded.text),
|
||||
};
|
||||
}
|
||||
const decoded = this.#graph;
|
||||
text = decoded.text;
|
||||
view = {
|
||||
component: "recovered_ir",
|
||||
encoding: decoded.encoding,
|
||||
sha256: decoded.sha256,
|
||||
bytes: decoded.bytes,
|
||||
authority:
|
||||
"uninterpreted retained graph source; fact validity is separate",
|
||||
};
|
||||
}
|
||||
const offset = input.offset ?? 0,
|
||||
limit = input.limit ?? 4096;
|
||||
return {
|
||||
source,
|
||||
...(view ? { view } : {}),
|
||||
text: text.slice(offset, offset + limit),
|
||||
total_code_units: text.length,
|
||||
next_offset: offset + limit < text.length ? offset + limit : null,
|
||||
};
|
||||
}
|
||||
}
|
||||
@@ -1,86 +0,0 @@
|
||||
import assert from "node:assert/strict";
|
||||
import test from "node:test";
|
||||
import { createHash } from "node:crypto";
|
||||
import { gzipSync } from "node:zlib";
|
||||
import { NativeEvidenceReads } from "./native-evidence-reads.mjs";
|
||||
|
||||
const hash = (s) => createHash("sha256").update(s).digest("hex");
|
||||
function captured(text) {
|
||||
const bytes = Buffer.from(text);
|
||||
const wire = JSON.stringify({
|
||||
binary: "/input",
|
||||
count: 1,
|
||||
language: "c",
|
||||
error: null,
|
||||
functions: [
|
||||
{
|
||||
address_hex: "0x1000",
|
||||
code: "retained C",
|
||||
error: null,
|
||||
atomic_facts: {
|
||||
schema_version: 4,
|
||||
encoding: "gzip-base64",
|
||||
raw_bytes: bytes.length,
|
||||
data: gzipSync(bytes).toString("base64"),
|
||||
},
|
||||
},
|
||||
],
|
||||
});
|
||||
return {
|
||||
text: wire,
|
||||
source: {
|
||||
evidence_id: "ev_owned",
|
||||
sha256: hash(wire),
|
||||
bytes: Buffer.byteLength(wire),
|
||||
},
|
||||
};
|
||||
}
|
||||
test("decoded Evidence pages preserve their own byte identity and raw source provenance", async () => {
|
||||
const logical = "{malformed JSON with 中文 and 😀",
|
||||
source = captured(logical);
|
||||
const reader = new NativeEvidenceReads(async () => source);
|
||||
const raw = await reader.read({ evidence_id: "ev_owned", limit: 1e5 });
|
||||
assert.equal(raw.text, source.text);
|
||||
assert.equal(raw.view, undefined);
|
||||
const chunks = [];
|
||||
for (let offset = 0; offset < logical.length; offset += 3) {
|
||||
const page = await reader.read({
|
||||
evidence_id: "ev_owned",
|
||||
component: "recovered_ir",
|
||||
offset,
|
||||
limit: 3,
|
||||
});
|
||||
chunks.push(page.text);
|
||||
assert.equal(page.source.sha256, source.source.sha256);
|
||||
assert.equal(page.view.sha256, hash(logical));
|
||||
assert.equal(page.view.bytes, Buffer.byteLength(logical));
|
||||
assert.equal(page.total_code_units, logical.length);
|
||||
}
|
||||
assert.equal(chunks.join(""), logical);
|
||||
});
|
||||
test("decoded view reuse still requires source-owner validation and follows content identity", async () => {
|
||||
let source = captured('{"schema_version":3}'),
|
||||
failure = false,
|
||||
reads = 0;
|
||||
const reader = new NativeEvidenceReads(async () => {
|
||||
reads++;
|
||||
if (failure) throw new Error("source_changed");
|
||||
return source;
|
||||
});
|
||||
const query = { evidence_id: "ev_owned", component: "recovered_ir" };
|
||||
await reader.read(query);
|
||||
failure = true;
|
||||
await assert.rejects(reader.read(query), /source_changed/);
|
||||
failure = false;
|
||||
source = captured('{"schema_version":3,"new":true}');
|
||||
assert.equal(
|
||||
(await reader.read(query)).text,
|
||||
'{"schema_version":3,"new":true}',
|
||||
);
|
||||
reader.clear();
|
||||
assert.equal(
|
||||
(await reader.read(query)).view.sha256,
|
||||
hash('{"schema_version":3,"new":true}'),
|
||||
);
|
||||
assert.equal(reads, 4);
|
||||
});
|
||||
@@ -1,36 +0,0 @@
|
||||
// The wire owner validates the archive before normal graph-row admission.
|
||||
import { gunzipSync } from "node:zlib";
|
||||
import { z } from "zod";
|
||||
import { requireFact } from "./native-contract.mjs";
|
||||
import { NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
|
||||
export const MAX_GRAPH_JSON_BYTES = 32 * 1024 * 1024;
|
||||
const envelope = z.strictObject({
|
||||
schema_version: z.literal(4),
|
||||
encoding: z.literal("gzip-base64"),
|
||||
raw_bytes: z.number().int().min(1).max(MAX_GRAPH_JSON_BYTES),
|
||||
data: z.string().max(NATIVE_OUTPUT_BYTES),
|
||||
});
|
||||
export function decodeGraphArchive(raw) {
|
||||
if (raw?.schema_version !== 4) return undefined;
|
||||
const encoded = envelope.parse(raw);
|
||||
requireFact(
|
||||
encoded.data.length > 0 &&
|
||||
encoded.data.length % 4 === 0 &&
|
||||
/^[A-Za-z0-9+/]*={0,2}$/u.test(encoded.data),
|
||||
"Native compressed graph has invalid base64",
|
||||
);
|
||||
const bytes = Buffer.from(encoded.data, "base64");
|
||||
requireFact(
|
||||
bytes.toString("base64") === encoded.data,
|
||||
"Native compressed graph has noncanonical base64",
|
||||
);
|
||||
// The native-declared length is admitted before inflation, not checked only
|
||||
// after an unbounded allocation. Node also verifies the gzip framing/CRC.
|
||||
const decoded = gunzipSync(bytes, { maxOutputLength: encoded.raw_bytes });
|
||||
requireFact(
|
||||
decoded.length === encoded.raw_bytes,
|
||||
"Native compressed graph length differs from its declaration",
|
||||
);
|
||||
return new TextDecoder("utf-8", { fatal: true }).decode(decoded);
|
||||
}
|
||||
@@ -1,66 +0,0 @@
|
||||
import assert from "node:assert/strict";
|
||||
import test from "node:test";
|
||||
import { gzipSync } from "node:zlib";
|
||||
import {
|
||||
decodeGraphArchive,
|
||||
MAX_GRAPH_JSON_BYTES,
|
||||
} from "./native-graph-encoding.mjs";
|
||||
|
||||
const encodeBytes = (bytes) => ({
|
||||
schema_version: 4,
|
||||
encoding: "gzip-base64",
|
||||
raw_bytes: bytes.length,
|
||||
data: gzipSync(bytes, { level: 1 }).toString("base64"),
|
||||
});
|
||||
const encode = (value) => encodeBytes(Buffer.from(JSON.stringify(value)));
|
||||
test("bounded gzip preserves logical JSON fields and legacy graphs without a second interpretation", () => {
|
||||
const graph = {
|
||||
schema_version: 3,
|
||||
statement: "中文\u0000😀",
|
||||
offset: "0xffffffffffffffff",
|
||||
};
|
||||
assert.equal(decodeGraphArchive(encode(graph)), JSON.stringify(graph));
|
||||
assert.equal(decodeGraphArchive(graph), undefined);
|
||||
});
|
||||
test("admission rejects unsupported envelopes and malformed base64 before inflation", () => {
|
||||
const good = encode({ schema_version: 3 });
|
||||
for (const change of [
|
||||
{ encoding: "deflate" },
|
||||
{ raw_bytes: 0 },
|
||||
{ raw_bytes: MAX_GRAPH_JSON_BYTES + 1 },
|
||||
{ unexpected: true },
|
||||
{ data: "AAAA\n" },
|
||||
{ data: "A===" },
|
||||
{ data: "?AAA" },
|
||||
{ data: "A".repeat(8 * 1024 * 1024 + 1) },
|
||||
])
|
||||
assert.throws(() => decodeGraphArchive({ ...good, ...change }));
|
||||
assert.throws(
|
||||
() => decodeGraphArchive({ ...good, data: "AB==" }),
|
||||
/noncanonical/,
|
||||
);
|
||||
});
|
||||
test("inflation respects declared size and verifies framing without interpreting logical source", () => {
|
||||
const graph = { schema_version: 3, statement: "x".repeat(1024 * 1024) };
|
||||
const good = encode(graph);
|
||||
assert.throws(() => decodeGraphArchive({ ...good, raw_bytes: 64 }), {
|
||||
code: "ERR_BUFFER_TOO_LARGE",
|
||||
});
|
||||
assert.throws(
|
||||
() => decodeGraphArchive({ ...good, raw_bytes: good.raw_bytes + 1 }),
|
||||
/differs/,
|
||||
);
|
||||
const corrupt = Buffer.from(good.data, "base64");
|
||||
corrupt[corrupt.length - 8] ^= 1;
|
||||
assert.throws(() =>
|
||||
decodeGraphArchive({ ...good, data: corrupt.toString("base64") }),
|
||||
);
|
||||
assert.throws(() =>
|
||||
decodeGraphArchive({
|
||||
...good,
|
||||
data: Buffer.from(good.data, "base64").subarray(0, -5).toString("base64"),
|
||||
}),
|
||||
);
|
||||
assert.throws(() => decodeGraphArchive(encodeBytes(Buffer.from([0xff]))));
|
||||
assert.equal(decodeGraphArchive(encodeBytes(Buffer.from("{"))), "{");
|
||||
});
|
||||
@@ -1,124 +0,0 @@
|
||||
# Bounded full recovered graphs
|
||||
|
||||
This implements the preceding native encoding probe as an opt-in experiment,
|
||||
not stock Kuna support, an integrated release or provider parity.
|
||||
|
||||
## Format and owners
|
||||
|
||||
Logical graph schema 3 stays unchanged. A fitting ordinary record keeps its
|
||||
existing representation. Before dropping facts, the writer can encode only the
|
||||
graph in a schema-4 `{encoding:"gzip-base64",raw_bytes,data}` envelope. Ordinary
|
||||
C/metadata and independent calls keep their fields. Existing graph/call/
|
||||
projection/code fallback and explicit failure still apply if encodings do not
|
||||
fit; no graph is clipped as complete.
|
||||
|
||||
Raw serialization stops at 32 MiB. Envelope overhead and base64 expansion are
|
||||
accounted for before allocation; gzip output and encoded alternatives share the
|
||||
unchanged caller budget of 8 MiB. Alternatives move ownership instead of cloning
|
||||
graphs. Construction separately admits 32,000 operations, 36,000 values, 100,000
|
||||
inputs, 12,000 blocks, 50,000 CFG edges and 4 MiB display/type text. Calls and
|
||||
projections keep their earlier limits.
|
||||
|
||||
Each collector now owns one `TextBudget` for combined display/type charges. The
|
||||
initial larger graph exposed a hidden second 2 MiB limit in value collection;
|
||||
the first real VM probe rejected it with `atomic text capacity exceeded` while
|
||||
preserving calls/C/projection. Passing the collector-owned budget through value
|
||||
extraction removes this competing policy. Final graph text is 2,345,629 bytes.
|
||||
|
||||
`native-graph-encoding.mjs` owns envelope admission, canonical base64, gzip
|
||||
integrity, exact lengths and fatal UTF-8 decoding. Inflation uses Node's
|
||||
[`maxOutputLength`](https://github.com/nodejs/node/blob/main/doc/api/zlib.md#class-options)
|
||||
with an admitted 32 MiB maximum. Logical schema/row/relationship/text admission
|
||||
and existing native identity/CFG/phi checks then operate on the same graph.
|
||||
Nested encodings fail logical admission; bad graphs preserve independent calls/C.
|
||||
|
||||
`evidence` defaults to the exact source wire record and original hash.
|
||||
`component:"recovered_ir"` pages decoded graph source with a separate view hash
|
||||
and byte count. Well-framed malformed logical facts remain inspectable without
|
||||
being accepted as facts. `NativeEvidenceReads` owns one decoded view up to 32 MiB
|
||||
per lease; the source owner revalidates bytes on every read. Content identity
|
||||
keys the view; successful close clears it. All adapters share these owners.
|
||||
|
||||
## Real verification
|
||||
|
||||
These graph-only engine/patch measurements use commit `e2205313a` and its
|
||||
version of `kuna-native-ir.patch`. The subsequent [switch extension](switch-study.md)
|
||||
has its own pinned engine/patch and retains the same logical graph fields.
|
||||
|
||||
Kuna upstream: `fe01d6294eb0d1053ce343c8943646df88e143fc`.
|
||||
Final preserved engine:
|
||||
`677b9795d49a65e459054ecbd3c0df37d603b541d72df9f74b46778947ed3936`.
|
||||
Patch SHA-256:
|
||||
`9f5754a15bd971326d14085f4c515348735d140be3bf36a371e7cf336fc291f3`.
|
||||
Binaries are copied outside mutable Cargo outputs, not hardlinked or committed.
|
||||
|
||||
On the pinned Debian SQLite target, both actual MCP adapters under Node 24.18.0
|
||||
verify 21,349 operations, 35,755 values, 1,229 blocks and zero omitted definitions.
|
||||
The complete graph deeply equals the earlier diagnostic golden, and ordinary
|
||||
C/metadata plus all 416 recovered calls are unchanged. Selected old projection
|
||||
frontiers now resolve, selected CFG/PHI reads work, both complete Evidence views
|
||||
verify and repeated snapshots reuse one source. This establishes producer
|
||||
fidelity, not independent machine/IR equivalence or runtime feasibility.
|
||||
|
||||
The actual wire document is **3,788,881 bytes**; decoded graph is **17,008,223
|
||||
bytes**. The public byte budget remains 8,388,608. Raw receipts/source copies are
|
||||
`full-graph-codec-*` under `.cache/atomic-codemode/`.
|
||||
|
||||
| Executed scope | Family wall seconds | CPU seconds | Kernel peak MiB |
|
||||
| ---------------------------------------- | ------------------: | ----------: | --------------: |
|
||||
| Initial build | 441.295 | 391.223 | 2,194.8 |
|
||||
| Focused native tests, including compile | 98.589 | 98.034 | 632.9 |
|
||||
| Final budget-version native tests | 176.333 | 175.748 | 644.1 |
|
||||
| Formatted cached rebuild | 26.999 | 26.749 | 499.3 |
|
||||
| Collector-budget rebuild | 361.041 | 342.220 | 2,192.9 |
|
||||
| First probe: hidden text-cap failure | 39.279 | 39.010 | 650.9 |
|
||||
| Exploratory success under system Node 20 | 44.858 | 44.480 | 576.6 |
|
||||
| Final Code Mode MCP, pinned Node 24 | 42.422 | 42.111 | 651.0 |
|
||||
| Final direct MCP, pinned Node 24 | 41.496 | 41.252 | 644.2 |
|
||||
| CLI: incorrect unique-name assumption | 1.229 | 1.037 | 132.5 |
|
||||
| CLI: corrected string/xref composition | 43.512 | 43.217 | 584.2 |
|
||||
| Full-graph cache benchmark | 41.912 | 39.955 | 588.4 |
|
||||
| Both catalogs/reference schemas | 3.644 | 1.716 | 142.7 |
|
||||
| Three small functions / call composition | 8.579 | 8.358 | 300.4 |
|
||||
| Legacy engine / current reader | 39.813 | 39.425 | 642.5 |
|
||||
|
||||
All resource runs use one CPU, 4 GiB hard / 3 GiB early-stop bounds, verified
|
||||
cleanup and OOM 0. Kernel peak includes charged file cache, not RSS. The system
|
||||
Node probe is outside pinned-toolchain validation; the verifier now rejects a
|
||||
wrong Node before native analysis. The first failed probe used preserved
|
||||
pre-fix `fb717dde...`. Native focused tests pass ten cases; shared JS passes 65.
|
||||
Changed JS lint/format, lenient native `check-spec`, applied-source whitespace
|
||||
and patch forward/reverse checks pass. Patch-context spaces are format markers.
|
||||
The three small-function records retain their prior ordinary representations;
|
||||
the current reader also accepts the old engine's explicit full-graph refusal,
|
||||
independent calls and 4,096-operation projection through real MCP reads.
|
||||
The final native test rerun also passes the same ten selected cases after the
|
||||
collector-budget change. Other CLI integration tests were compiled but filtered
|
||||
out; they are not additional executed coverage.
|
||||
Native full corpus/stages/workspace and whole-REA gates remain unexecuted or
|
||||
previously resource-limited; no upstream/PR/integration readiness is claimed.
|
||||
|
||||
Fresh-lease snapshot: 39,677.7 ms / 38.726 family CPU seconds. Same-lease warm:
|
||||
67.1 ms / 0.068 seconds, identical handle/source. Cumulative peak is 588.4 MiB,
|
||||
an upper bound on isolated warm memory. OS cache is uncontrolled; no cross-session
|
||||
cache or model-token benefit is inferred from this scripted benchmark.
|
||||
|
||||
## From names/messages to conditions
|
||||
|
||||
The inventory has two identical VDBE names (`0x288f0`, `0xea6d0`) without aliases.
|
||||
The initial CLI assumed uniqueness. Its corrected workflow composes strings,
|
||||
incoming references and owning-function addresses before one VM snapshot; then
|
||||
constant producers, uses, derived controllers and selected values/definitions.
|
||||
It obtains both savepoint guards in sixteen host reads without the C report.
|
||||
The final example also checks pagination before treating selected facts as a
|
||||
complete set. A retained-source replay derives the two controller relations in
|
||||
about 8.9 ms wall; isolated replay CPU/peak memory were not measured.
|
||||
|
||||
The [question/oracle](savepoint-control-oracle.md) are fixed before model trials,
|
||||
with function name/messages but no supplied code addresses. Selected independent
|
||||
machine evidence checks local guards, load/spill/setup and shared call. Recovered
|
||||
CFG/use coverage still does not establish full machine flow, memory aliases,
|
||||
nontermination or feasible execution. The first [real-model pair](savepoint-control-study.md)
|
||||
now passes the frozen question through both adapters; its whole-task costs are
|
||||
mixed and include discovery noise and a Code Mode parameter retry. Public
|
||||
scripted checks alone do not establish those model outcomes.
|
||||
@@ -1,258 +0,0 @@
|
||||
// Native JSON decoding and resource admission; row-wise validation bounds errors.
|
||||
import { z } from "zod";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
import { readCallFacet } from "./native-call-contract.mjs";
|
||||
import { readProjectionFacet } from "./native-projection-contract.mjs";
|
||||
import { readSwitchFacet } from "./native-switch-contract.mjs";
|
||||
import { operation, parseOperation } from "./native-operation-contract.mjs";
|
||||
import {
|
||||
decodeGraphArchive,
|
||||
MAX_GRAPH_JSON_BYTES,
|
||||
} from "./native-graph-encoding.mjs";
|
||||
import {
|
||||
hex,
|
||||
ordinal,
|
||||
valueId,
|
||||
operationId,
|
||||
edgeId,
|
||||
value,
|
||||
rowParser,
|
||||
requireFact,
|
||||
formatFailure,
|
||||
readNativeFacet,
|
||||
} from "./native-contract.mjs";
|
||||
|
||||
const edge = z.strictObject({
|
||||
id: edgeId,
|
||||
target: ordinal,
|
||||
index: ordinal,
|
||||
polarity: z.enum(["true", "false"]).nullable(),
|
||||
condition_value: valueId.nullable(),
|
||||
condition_test: z.enum(["eq-zero", "ne-zero"]).nullable(),
|
||||
});
|
||||
const factsSchema = z.strictObject({
|
||||
schema_version: z.literal(3),
|
||||
blocks: z
|
||||
.array(
|
||||
z.strictObject({
|
||||
id: ordinal,
|
||||
start: hex,
|
||||
operations: z.array(operationId),
|
||||
predecessors: z.array(ordinal),
|
||||
successors: z.array(edge),
|
||||
}),
|
||||
)
|
||||
.min(1)
|
||||
.max(12000),
|
||||
operations: z.array(operation).max(32000),
|
||||
values: z.array(value).max(36000),
|
||||
coverage: z.strictObject({
|
||||
scope: z.string(),
|
||||
machine_semantic_equivalence: z.literal("unknown"),
|
||||
indirect_branch_operations: z.array(operationId),
|
||||
omitted_definition_ids: z.array(operationId),
|
||||
issues: z.array(z.string()),
|
||||
}),
|
||||
});
|
||||
const documentSchema = z.object({
|
||||
binary: z.string(),
|
||||
count: z.literal(1),
|
||||
language: z.string(),
|
||||
error: z.null(),
|
||||
functions: z
|
||||
.array(
|
||||
z.object({
|
||||
address_hex: hex,
|
||||
code: z.string(),
|
||||
error: z.null(),
|
||||
atomic_facts: z.unknown().optional(),
|
||||
atomic_facts_error: z.string().optional(),
|
||||
atomic_calls: z.unknown().optional(),
|
||||
atomic_calls_error: z.string().optional(),
|
||||
atomic_switches: z.unknown().optional(),
|
||||
atomic_switches_error: z.string().optional(),
|
||||
atomic_projection: z.unknown().optional(),
|
||||
atomic_projection_error: z.string().optional(),
|
||||
}),
|
||||
)
|
||||
.length(1),
|
||||
});
|
||||
|
||||
// Parse one row at a time: a bad provider must not expand thousands of malformed
|
||||
// records into a giant Zod issue list before the caller can reject the source.
|
||||
const nativeBlock = factsSchema.shape.blocks.element;
|
||||
const parseBlock = rowParser(nativeBlock, {
|
||||
operations: (id) => operationId.parse(id),
|
||||
predecessors: (id) => ordinal.parse(id),
|
||||
successors: (row) => edge.parse(row),
|
||||
});
|
||||
const graphCoverage = factsSchema.shape.coverage;
|
||||
const parseCoverage = rowParser(graphCoverage, {
|
||||
indirect_branch_operations: (row) => operationId.parse(row),
|
||||
omitted_definition_ids: (row) => operationId.parse(row),
|
||||
issues: (row) => z.string().parse(row),
|
||||
});
|
||||
const parseFacts = rowParser(
|
||||
factsSchema,
|
||||
{
|
||||
blocks: parseBlock,
|
||||
operations: parseOperation,
|
||||
values: (row) => value.parse(row),
|
||||
},
|
||||
{ coverage: parseCoverage },
|
||||
);
|
||||
|
||||
function preflight(facts) {
|
||||
requireFact(
|
||||
facts && typeof facts === "object",
|
||||
"Native atomic facts facet is unavailable; inspect the retained source for its reason and C",
|
||||
);
|
||||
requireFact(
|
||||
facts.schema_version === 3,
|
||||
"Unsupported native atomic facts schema version; expected 3",
|
||||
);
|
||||
requireFact(
|
||||
facts.blocks?.length > 0,
|
||||
"Native recovered graph is unavailable",
|
||||
);
|
||||
for (const [key, limit] of [
|
||||
["blocks", 12000],
|
||||
["operations", 32000],
|
||||
["values", 36000],
|
||||
])
|
||||
requireFact(
|
||||
Array.isArray(facts[key]) && facts[key].length <= limit,
|
||||
`Native ${key} exceeds resource capacity or has invalid shape`,
|
||||
);
|
||||
for (const [rows, limit] of [
|
||||
[facts.coverage?.indirect_branch_operations, 32000],
|
||||
[facts.coverage?.omitted_definition_ids, 36000],
|
||||
[facts.coverage?.issues, 12000],
|
||||
])
|
||||
requireFact(
|
||||
Array.isArray(rows) && rows.length <= limit,
|
||||
"Native graph coverage exceeds capacity or has invalid shape",
|
||||
);
|
||||
const inputs = facts.operations.reduce(
|
||||
(sum, op) => sum + (Array.isArray(op?.inputs) ? op.inputs.length : 0),
|
||||
0,
|
||||
);
|
||||
const members = facts.blocks.reduce(
|
||||
(sum, block) =>
|
||||
sum + (Array.isArray(block?.operations) ? block.operations.length : 0),
|
||||
0,
|
||||
);
|
||||
const edges = facts.blocks.reduce(
|
||||
(sum, block) =>
|
||||
sum +
|
||||
(Array.isArray(block?.successors) ? block.successors.length : 0) +
|
||||
(Array.isArray(block?.predecessors) ? block.predecessors.length : 0),
|
||||
0,
|
||||
);
|
||||
requireFact(
|
||||
inputs <= 100000 && members <= 32000 && edges <= 100000,
|
||||
"Native relationships exceed resource capacity",
|
||||
);
|
||||
const displayBytes =
|
||||
facts.operations.reduce(
|
||||
(sum, op) =>
|
||||
sum +
|
||||
(typeof op?.statement === "string"
|
||||
? Buffer.byteLength(op.statement)
|
||||
: 0),
|
||||
0,
|
||||
) +
|
||||
facts.values.reduce(
|
||||
(sum, value) =>
|
||||
sum +
|
||||
(typeof value?.type_name === "string"
|
||||
? Buffer.byteLength(value.type_name)
|
||||
: 0),
|
||||
0,
|
||||
);
|
||||
requireFact(
|
||||
displayBytes <= 4 * 1024 * 1024,
|
||||
"Native graph display exceeds resource capacity",
|
||||
);
|
||||
}
|
||||
const readGraph = (fn) =>
|
||||
readNativeFacet(fn, "atomic_facts", "recovered IR", (raw) => {
|
||||
const decoded = decodeGraphArchive(raw);
|
||||
const graph = decoded === undefined ? raw : JSON.parse(decoded);
|
||||
preflight(graph);
|
||||
return parseFacts(graph);
|
||||
});
|
||||
|
||||
function readSource(text) {
|
||||
requireFact(
|
||||
Buffer.byteLength(text) <= NATIVE_OUTPUT_BYTES,
|
||||
"Native JSON source exceeds transport budget",
|
||||
);
|
||||
return JSON.parse(text);
|
||||
}
|
||||
function readFunction(raw, entry, expectedInput) {
|
||||
requireFact(
|
||||
Array.isArray(raw?.functions) && raw.functions.length === 1,
|
||||
"Native source must describe exactly one function",
|
||||
);
|
||||
const doc = documentSchema.parse(raw),
|
||||
fn = doc.functions[0];
|
||||
if (expectedInput !== undefined && doc.binary !== expectedInput)
|
||||
throw new FactFailure(
|
||||
"subject_mismatch",
|
||||
"Native source names a different command input",
|
||||
{ expected: expectedInput, observed: doc.binary },
|
||||
);
|
||||
if (entry !== undefined && fn.address_hex !== entry)
|
||||
throw new FactFailure(
|
||||
"subject_mismatch",
|
||||
"Native function entry differs from requested subject",
|
||||
);
|
||||
return { doc, fn };
|
||||
}
|
||||
export function parseNativeFunction(text, entry, expectedInput) {
|
||||
try {
|
||||
const { doc, fn } = readFunction(readSource(text), entry, expectedInput);
|
||||
return {
|
||||
code: fn.code,
|
||||
language: doc.language,
|
||||
facts: readGraph(fn),
|
||||
calls: readCallFacet(fn),
|
||||
switches: readSwitchFacet(fn),
|
||||
projection: readProjectionFacet(fn),
|
||||
};
|
||||
} catch (error) {
|
||||
throw formatFailure(error);
|
||||
}
|
||||
}
|
||||
|
||||
export function readNativeGraphEvidence(text) {
|
||||
try {
|
||||
const raw = readSource(text);
|
||||
if (raw?.functions === undefined)
|
||||
throw new FactFailure(
|
||||
"capability_unavailable",
|
||||
"This retained source is not a native function record",
|
||||
);
|
||||
const { fn } = readFunction(raw);
|
||||
if (fn.atomic_facts === undefined)
|
||||
throw new FactFailure(
|
||||
"capability_unavailable",
|
||||
"Recovered graph source was not captured",
|
||||
{ native_error: fn.atomic_facts_error?.slice(0, 1024) },
|
||||
);
|
||||
const decoded = decodeGraphArchive(fn.atomic_facts);
|
||||
const view = decoded ?? JSON.stringify(fn.atomic_facts);
|
||||
requireFact(
|
||||
Buffer.byteLength(view) <= MAX_GRAPH_JSON_BYTES,
|
||||
"Native graph Evidence view exceeds decoded capacity",
|
||||
);
|
||||
return {
|
||||
text: view,
|
||||
encoding: decoded === undefined ? "json" : "gzip-base64",
|
||||
};
|
||||
} catch (error) {
|
||||
throw formatFailure(error);
|
||||
}
|
||||
}
|
||||
@@ -1,355 +0,0 @@
|
||||
// The native-format boundary owns validation and indexes; consumers never parse
|
||||
// diagnostic statements to recover value identity, operand indexes or CFG edges.
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import {
|
||||
indexSwitchFacet,
|
||||
validateSwitchGraph,
|
||||
validateSwitchValues,
|
||||
createSwitchViews,
|
||||
} from "./native-switch-views.mjs";
|
||||
import { parseNativeFunction } from "./native-ir-contract.mjs";
|
||||
import { controlDependencies } from "./control-dependence.mjs";
|
||||
import { freezeFacts as freeze } from "./fact-reads.mjs";
|
||||
import {
|
||||
indexCallFacet,
|
||||
validateCallGraph,
|
||||
createCallViews,
|
||||
} from "./native-call-views.mjs";
|
||||
import { requireFact } from "./native-contract.mjs";
|
||||
import {
|
||||
indexOperationFacts,
|
||||
createOperationViews,
|
||||
coverageIssues,
|
||||
} from "./native-operation-views.mjs";
|
||||
import {
|
||||
indexProjection,
|
||||
createProjectionViews,
|
||||
} from "./native-projection-views.mjs";
|
||||
import {
|
||||
uniqueNative as unique,
|
||||
nativePage as page,
|
||||
readNativeIdentities as readIdentities,
|
||||
} from "./native-reads.mjs";
|
||||
function validateBlocks(operations, values, blocks) {
|
||||
const membership = new Set();
|
||||
for (const block of blocks.values()) {
|
||||
for (const id of block.operations) {
|
||||
requireFact(
|
||||
operations.get(id)?.block === block.id && !membership.has(id),
|
||||
"Native operation/block ownership mismatch",
|
||||
);
|
||||
membership.add(id);
|
||||
}
|
||||
for (const [index, edge] of block.successors.entries()) {
|
||||
requireFact(
|
||||
edge.index === index &&
|
||||
edge.id === `${block.id}:${index}` &&
|
||||
blocks.get(edge.target)?.predecessors.includes(block.id),
|
||||
"Native CFG successor/predecessor mismatch",
|
||||
);
|
||||
requireFact(
|
||||
edge.condition_value === null || values.has(edge.condition_value),
|
||||
"Dangling native branch condition",
|
||||
);
|
||||
}
|
||||
for (const from of block.predecessors)
|
||||
requireFact(
|
||||
blocks.get(from)?.successors.some((edge) => edge.target === block.id),
|
||||
"Dangling native predecessor",
|
||||
);
|
||||
}
|
||||
requireFact(
|
||||
membership.size === operations.size,
|
||||
"Native operation omitted from block membership",
|
||||
);
|
||||
}
|
||||
|
||||
function validatePhiBindings(operations, blocks) {
|
||||
for (const op of operations.values())
|
||||
for (const [position, input] of op.inputs.entries()) {
|
||||
if (input.phi_predecessor !== null)
|
||||
requireFact(
|
||||
blocks.get(op.block).predecessors[position] === input.phi_predecessor,
|
||||
"Native phi/predecessor mismatch",
|
||||
);
|
||||
if (input.phi_predecessor_edge !== null)
|
||||
requireFact(
|
||||
blocks
|
||||
.get(input.phi_predecessor)
|
||||
?.successors.some(
|
||||
(edge) =>
|
||||
edge.id === input.phi_predecessor_edge &&
|
||||
edge.target === op.block,
|
||||
),
|
||||
"Native phi incoming-edge identity mismatch",
|
||||
);
|
||||
}
|
||||
}
|
||||
|
||||
function indexNative(facts) {
|
||||
const core = indexOperationFacts(
|
||||
facts.operations,
|
||||
facts.values,
|
||||
facts.coverage.omitted_definition_ids,
|
||||
);
|
||||
const blocks = unique(
|
||||
facts.blocks.map(({ start, ...block }) =>
|
||||
freeze({ ...block, origin_start: start }),
|
||||
),
|
||||
);
|
||||
validateBlocks(core.operations, core.values, blocks);
|
||||
validatePhiBindings(core.operations, blocks);
|
||||
const incoming = new Map(),
|
||||
outgoing = new Map();
|
||||
for (const block of blocks.values())
|
||||
for (const edge of block.successors) {
|
||||
const row = freeze({
|
||||
...edge,
|
||||
from: block.id,
|
||||
to: edge.target,
|
||||
operation: block.operations.at(-1) ?? null,
|
||||
kind: edge.polarity ? "conditional" : "recovered",
|
||||
});
|
||||
for (const [map, key] of [
|
||||
[incoming, edge.target],
|
||||
[outgoing, block.id],
|
||||
]) {
|
||||
if (!map.has(key)) map.set(key, []);
|
||||
map.get(key).push(row);
|
||||
}
|
||||
}
|
||||
return { ...core, blocks, incoming, outgoing };
|
||||
}
|
||||
|
||||
function createViews(facts, code, indexes) {
|
||||
const { blocks, incoming, outgoing, normalized } = indexes;
|
||||
const {
|
||||
issues,
|
||||
indirect_branch_operations,
|
||||
omitted_definition_ids,
|
||||
...metadata
|
||||
} = facts.coverage;
|
||||
const coverage = freeze({
|
||||
...metadata,
|
||||
...coverageIssues(issues),
|
||||
indirect_branch_operation_count: indirect_branch_operations.length,
|
||||
omitted_definition_count: omitted_definition_ids.length,
|
||||
operations: normalized.size,
|
||||
values: indexes.values.size,
|
||||
blocks: blocks.size,
|
||||
control_flow: "complete-for-recovered-graph",
|
||||
representation:
|
||||
"optimized recovered IR; origins do not establish machine branch targets or block extents",
|
||||
runtime_successors: "unknown",
|
||||
memory_alias_dependencies: "unknown",
|
||||
});
|
||||
const core = createOperationViews(indexes, coverage);
|
||||
return Object.freeze({
|
||||
...core,
|
||||
code,
|
||||
block: (id) => blocks.get(id) ?? null,
|
||||
incoming: (id) => incoming.get(id) ?? [],
|
||||
blocks: (input) => readIdentities(blocks, input),
|
||||
edges: (input) => {
|
||||
if (["definitions", "uses", "operands"].includes(input.kind))
|
||||
return core.operationRelations(input);
|
||||
if (input.kind === "controllers") {
|
||||
const controls = controlDependencies(facts.blocks, {
|
||||
dependents: new Set(input.seeds),
|
||||
});
|
||||
return {
|
||||
...page(controls.relations, input),
|
||||
relation_coverage: controls.coverage,
|
||||
};
|
||||
}
|
||||
return page(
|
||||
input.seeds.flatMap(
|
||||
(seed) =>
|
||||
(input.kind === "incoming" ? incoming : outgoing).get(seed) ?? [],
|
||||
),
|
||||
input,
|
||||
);
|
||||
},
|
||||
});
|
||||
}
|
||||
|
||||
const projectedFailure = (error) => ({
|
||||
code: error.code ?? "invalid_provider_output",
|
||||
message: error.message,
|
||||
details: error.details ?? {},
|
||||
});
|
||||
|
||||
const operationFailure = (projectionError, graphError) =>
|
||||
projectionError?.details?.availability === "not-captured"
|
||||
? graphError
|
||||
: (projectionError ?? graphError);
|
||||
|
||||
const captureFacets = ({
|
||||
graph,
|
||||
projection,
|
||||
calls,
|
||||
graphError,
|
||||
operationError,
|
||||
callError,
|
||||
switches,
|
||||
switchError,
|
||||
}) =>
|
||||
freeze({
|
||||
code: { availability: "available" },
|
||||
recovered_ir: graph
|
||||
? { availability: "available" }
|
||||
: { availability: "unavailable", error: graphError },
|
||||
operations: graph
|
||||
? {
|
||||
availability: "available",
|
||||
scope: "complete-for-recovered-graph",
|
||||
count: graph.coverage.operations,
|
||||
}
|
||||
: projection
|
||||
? {
|
||||
availability: "available",
|
||||
scope: "bounded-call-rooted-definitions",
|
||||
count: projection.coverage.operations,
|
||||
root_count: projection.coverage.root_count,
|
||||
definition_frontier_count:
|
||||
projection.coverage.definition_frontier_count,
|
||||
}
|
||||
: {
|
||||
availability: "unavailable",
|
||||
error: operationError,
|
||||
},
|
||||
calls: calls
|
||||
? { availability: "available", ...calls.summary }
|
||||
: { availability: "unavailable", error: callError },
|
||||
switches: switches
|
||||
? { availability: "available", ...switches.summary }
|
||||
: { availability: "unavailable", error: switchError },
|
||||
values: graph
|
||||
? {
|
||||
availability: "available",
|
||||
scope: "complete-for-recovered-graph",
|
||||
count: graph.coverage.values,
|
||||
}
|
||||
: projection
|
||||
? {
|
||||
availability: "available",
|
||||
scope: "call-rooted-definitions",
|
||||
count: projection.coverage.values,
|
||||
}
|
||||
: calls
|
||||
? {
|
||||
availability: "available",
|
||||
scope: "call-boundary-only",
|
||||
count: calls.value_count,
|
||||
}
|
||||
: { availability: "unavailable", error: callError },
|
||||
});
|
||||
|
||||
const loadFacet = (result, build) => {
|
||||
if (!result.ok) return { error: projectedFailure(result.error) };
|
||||
try {
|
||||
return build(result.value);
|
||||
} catch (error) {
|
||||
return { error: projectedFailure(error) };
|
||||
}
|
||||
};
|
||||
const unavailable =
|
||||
(facet, error = {}) =>
|
||||
() => {
|
||||
throw new FactFailure(error.code, error.message, {
|
||||
...error.details,
|
||||
facet,
|
||||
});
|
||||
};
|
||||
export function parseNativeIR(text, entry, expectedInput) {
|
||||
const capture = parseNativeFunction(text, entry, expectedInput);
|
||||
const graph = loadFacet(capture.facts, (facts) => {
|
||||
freeze(facts);
|
||||
const index = indexNative(facts);
|
||||
return { index, view: createViews(facts, capture.code, index) };
|
||||
});
|
||||
const calls = loadFacet(capture.calls, (facts) => {
|
||||
const index = indexCallFacet(facts);
|
||||
if (graph.view) validateCallGraph(index, graph.index);
|
||||
return { index, view: createCallViews(facts, index) };
|
||||
});
|
||||
const unavailableGraph = unavailable("recovered_ir", graph.error);
|
||||
const projection = graph.view
|
||||
? {}
|
||||
: loadFacet(capture.projection, (facts) => {
|
||||
const index = indexProjection(facts);
|
||||
// Without a full graph, the independent call owner validates the projection.
|
||||
if (calls.view) validateCallGraph(calls.index, index);
|
||||
return {
|
||||
index,
|
||||
view: createProjectionViews(
|
||||
facts,
|
||||
capture.code,
|
||||
index,
|
||||
unavailableGraph,
|
||||
),
|
||||
};
|
||||
});
|
||||
const unavailableCalls = unavailable("calls", calls.error);
|
||||
const switches = loadFacet(capture.switches, (facts) => {
|
||||
const index = indexSwitchFacet(facts);
|
||||
if (graph.view) validateSwitchGraph(index, graph.index);
|
||||
else {
|
||||
validateSwitchValues(index, calls.index?.values);
|
||||
validateSwitchValues(index, projection.index?.values);
|
||||
}
|
||||
return { view: createSwitchViews(facts, index) };
|
||||
});
|
||||
const unavailableSwitches = unavailable("switches", switches.error);
|
||||
const operationError = operationFailure(projection.error, graph.error);
|
||||
const unavailableOperations = unavailable("operations", operationError);
|
||||
const fallback = {
|
||||
code: capture.code,
|
||||
coverage: freeze({
|
||||
scope: "native reconstructed code; recovered IR facet unavailable",
|
||||
operations: null,
|
||||
blocks: null,
|
||||
control_flow: "unknown",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
issues: [graph.error?.message],
|
||||
}),
|
||||
...Object.fromEntries(
|
||||
["operation", "findConstant", "uses", "select", "slice"].map((name) => [
|
||||
name,
|
||||
unavailableOperations,
|
||||
]),
|
||||
),
|
||||
...Object.fromEntries(
|
||||
["block", "incoming", "blocks"].map((name) => [name, unavailableGraph]),
|
||||
),
|
||||
edges: (input) =>
|
||||
["definitions", "uses"].includes(input.kind)
|
||||
? unavailableOperations()
|
||||
: unavailableGraph(),
|
||||
};
|
||||
return Object.freeze({
|
||||
...(graph.view ?? projection.view ?? fallback),
|
||||
edges: (input) =>
|
||||
input.kind === "cases"
|
||||
? (switches.view?.cases ?? unavailableSwitches)(input)
|
||||
: (graph.view ?? projection.view ?? fallback).edges(input),
|
||||
language: capture.language,
|
||||
calls: calls.view?.calls ?? unavailableCalls,
|
||||
switches: switches.view?.switches ?? unavailableSwitches,
|
||||
values:
|
||||
graph.view?.values ??
|
||||
projection.view?.values ??
|
||||
calls.view?.values ??
|
||||
unavailableCalls,
|
||||
facets: captureFacets({
|
||||
graph: graph.view,
|
||||
projection: projection.view,
|
||||
calls: calls.view,
|
||||
graphError: graph.error,
|
||||
operationError,
|
||||
callError: calls.error,
|
||||
switches: switches.view,
|
||||
switchError: switches.error,
|
||||
}),
|
||||
});
|
||||
}
|
||||
@@ -1,401 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { gzipSync } from "node:zlib";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
import { readNativeGraphEvidence } from "./native-ir-contract.mjs";
|
||||
import { controlDependencies } from "./control-dependence.mjs";
|
||||
|
||||
const id = (n) => `0x${n}:0`;
|
||||
const val = (n, definition = null) => ({
|
||||
id: `v${n}`,
|
||||
space: "register",
|
||||
offset: "0x0",
|
||||
size_bytes: 8,
|
||||
is_constant: false,
|
||||
is_input: definition === null,
|
||||
is_annotation: false,
|
||||
definition,
|
||||
type_name: "uint64",
|
||||
});
|
||||
const input = (n, value_id, phi_predecessor = null) => ({
|
||||
input_index: n,
|
||||
value_id,
|
||||
phi_predecessor,
|
||||
phi_predecessor_edge:
|
||||
phi_predecessor === null ? null : `${phi_predecessor}:0`,
|
||||
});
|
||||
const edge = (from, target, index, polarity = null) => ({
|
||||
id: `${from}:${index}`,
|
||||
target,
|
||||
index,
|
||||
polarity,
|
||||
condition_value: polarity ? "v0" : null,
|
||||
condition_test: polarity ? "ne-zero" : null,
|
||||
});
|
||||
const op = (block, opcode, inputs, output = null) => ({
|
||||
id: id(block),
|
||||
address: `0x${block}`,
|
||||
block,
|
||||
opcode,
|
||||
inputs,
|
||||
output,
|
||||
statement: "display is deliberately unparseable",
|
||||
is_dead: false,
|
||||
});
|
||||
|
||||
function diamond() {
|
||||
return {
|
||||
binary: "/private/input",
|
||||
count: 1,
|
||||
language: "c",
|
||||
error: null,
|
||||
functions: [
|
||||
{
|
||||
address_hex: "0x1000",
|
||||
error: null,
|
||||
code: "reconstructed C",
|
||||
atomic_facts: {
|
||||
schema_version: 3,
|
||||
blocks: [
|
||||
{
|
||||
id: 0,
|
||||
start: "0x1000",
|
||||
operations: [id(0)],
|
||||
predecessors: [],
|
||||
successors: [edge(0, 1, 0, "false"), edge(0, 2, 1, "true")],
|
||||
},
|
||||
{
|
||||
id: 1,
|
||||
start: "0x1001",
|
||||
operations: [id(1)],
|
||||
predecessors: [0],
|
||||
successors: [edge(1, 3, 0)],
|
||||
},
|
||||
{
|
||||
id: 2,
|
||||
start: "0x1002",
|
||||
operations: [id(2)],
|
||||
predecessors: [0],
|
||||
successors: [edge(2, 3, 0)],
|
||||
},
|
||||
{
|
||||
id: 3,
|
||||
start: "0x1003",
|
||||
operations: [id(3)],
|
||||
predecessors: [1, 2],
|
||||
successors: [],
|
||||
},
|
||||
],
|
||||
operations: [
|
||||
op(0, "CPUI_CBRANCH", [input(0, "v0")]),
|
||||
op(1, "CPUI_COPY", [input(0, "v4")], "v1"),
|
||||
op(2, "CPUI_COPY", [input(0, "v4")], "v2"),
|
||||
op(
|
||||
3,
|
||||
"CPUI_MULTIEQUAL",
|
||||
[input(0, "v1", 1), input(1, "v2", 2)],
|
||||
"v3",
|
||||
),
|
||||
],
|
||||
values: [
|
||||
val(0),
|
||||
val(1, id(1)),
|
||||
val(2, id(2)),
|
||||
val(3, id(3)),
|
||||
{
|
||||
...val(4),
|
||||
is_constant: true,
|
||||
is_input: false,
|
||||
space: "const",
|
||||
offset: "0xffffffffffffffff",
|
||||
},
|
||||
],
|
||||
coverage: {
|
||||
scope: "native recovered graph",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
indirect_branch_operations: [],
|
||||
omitted_definition_ids: [],
|
||||
issues: [],
|
||||
},
|
||||
},
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
test("compressed and ordinary graphs have identical parsed facts and malformed archives preserve code", () => {
|
||||
const doc = diamond(),
|
||||
raw = Buffer.from(JSON.stringify(doc.functions[0].atomic_facts));
|
||||
const ordinary = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
doc.functions[0].atomic_facts = {
|
||||
schema_version: 4,
|
||||
encoding: "gzip-base64",
|
||||
raw_bytes: raw.length,
|
||||
data: gzipSync(raw).toString("base64"),
|
||||
};
|
||||
const encoded = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.deepEqual(encoded.coverage, ordinary.coverage);
|
||||
assert.deepEqual(
|
||||
encoded.select({ fields: ["inputs", "output", "definitions"] }),
|
||||
ordinary.select({ fields: ["inputs", "output", "definitions"] }),
|
||||
);
|
||||
assert.deepEqual(encoded.blocks({}), ordinary.blocks({}));
|
||||
assert.deepEqual(readNativeGraphEvidence(JSON.stringify(doc)), {
|
||||
text: raw.toString(),
|
||||
encoding: "gzip-base64",
|
||||
});
|
||||
for (const change of [
|
||||
{ raw_bytes: 2 },
|
||||
{ data: "broken" },
|
||||
{ raw_bytes: 32 * 1024 * 1024 + 1 },
|
||||
]) {
|
||||
const bad = structuredClone(doc);
|
||||
Object.assign(bad.functions[0].atomic_facts, change);
|
||||
const model = parseNativeIR(JSON.stringify(bad), "0x1000");
|
||||
assert.equal(model.code, "reconstructed C");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.throws(() => model.select({}));
|
||||
}
|
||||
for (const text of [
|
||||
"{malformed logical JSON",
|
||||
JSON.stringify({ schema_version: 4, data: "nested" }),
|
||||
]) {
|
||||
const bad = structuredClone(doc),
|
||||
bytes = Buffer.from(text);
|
||||
bad.functions[0].atomic_facts = {
|
||||
schema_version: 4,
|
||||
encoding: "gzip-base64",
|
||||
raw_bytes: bytes.length,
|
||||
data: gzipSync(bytes).toString("base64"),
|
||||
};
|
||||
const source = JSON.stringify(bad);
|
||||
assert.equal(
|
||||
parseNativeIR(source, "0x1000").facets.recovered_ir.availability,
|
||||
"unavailable",
|
||||
);
|
||||
assert.equal(
|
||||
readNativeGraphEvidence(source).text,
|
||||
text,
|
||||
"Malformed logical facts remain inspectable through owned decoded source",
|
||||
);
|
||||
}
|
||||
assert.throws(
|
||||
() => readNativeGraphEvidence('{"hex":"0102"}'),
|
||||
(error) => error.code === "capability_unavailable",
|
||||
);
|
||||
});
|
||||
|
||||
test("native identity, implicit CFG and operand/phi bindings do not depend on display text", () => {
|
||||
const model = parseNativeIR(JSON.stringify(diamond()), "0x1000");
|
||||
assert.equal(model.operation("0x3:0").origin_address, "0x3");
|
||||
assert.equal(Object.hasOwn(model.operation("0x3:0"), "address"), false);
|
||||
assert.equal(model.block(1).origin_start, "0x1001");
|
||||
const operand = model.edges({
|
||||
kind: "operands",
|
||||
seeds: ["0x3:0"],
|
||||
input_indices: [1],
|
||||
}).items[0];
|
||||
assert.deepEqual(operand, {
|
||||
operation: "0x3:0",
|
||||
input_index: 1,
|
||||
value_id: "v2",
|
||||
definition: "0x2:0",
|
||||
phi_predecessor: 2,
|
||||
phi_predecessor_edge: "2:0",
|
||||
});
|
||||
assert.equal(Object.hasOwn(model.block(1), "start"), false);
|
||||
assert.equal(model.select({ origin_addresses: ["0x3"] }).total, 1);
|
||||
assert.deepEqual(model.findConstant("0xffffffffffffffff"), [
|
||||
"0x1:0",
|
||||
"0x2:0",
|
||||
]);
|
||||
assert.equal(model.edges({ kind: "outgoing", seeds: [1] }).items[0].to, 3);
|
||||
assert.deepEqual(
|
||||
model
|
||||
.operation("0x3:0")
|
||||
.definitions.map(({ input_index, phi_predecessor }) => [
|
||||
input_index,
|
||||
phi_predecessor,
|
||||
]),
|
||||
[
|
||||
[0, 1],
|
||||
[1, 2],
|
||||
],
|
||||
);
|
||||
assert.equal(
|
||||
model.values({ ids: ["v4"] }).items[0].offset,
|
||||
"0xffffffffffffffff",
|
||||
);
|
||||
const batch = model.values({ ids: ["v4", "vdead"] });
|
||||
assert.equal(batch.items.length, 1);
|
||||
assert.deepEqual(batch.missing, ["vdead"]);
|
||||
assert.throws(() => {
|
||||
model.operation("0x3:0").inputs[0].value_id = "foreign";
|
||||
});
|
||||
assert.deepEqual(
|
||||
model
|
||||
.edges({ kind: "controllers", seeds: [1] })
|
||||
.items.map((row) => [row.controller, row.edge.polarity]),
|
||||
[[0, "false"]],
|
||||
);
|
||||
assert.equal(model.edges({ kind: "controllers", seeds: [3] }).total, 0);
|
||||
});
|
||||
|
||||
test("malformed identities, dangling relationships, foreign entry and phi reorder fail at the producer boundary", () => {
|
||||
for (const mutate of [
|
||||
(facts) => {
|
||||
facts.values.push(facts.values[0]);
|
||||
},
|
||||
(facts) => {
|
||||
facts.operations[0].inputs[0].value_id = "vdead";
|
||||
},
|
||||
(facts) => {
|
||||
facts.blocks[1].predecessors = [];
|
||||
},
|
||||
(facts) => {
|
||||
facts.operations[3].inputs[0].phi_predecessor = 2;
|
||||
},
|
||||
(facts) => {
|
||||
facts.values[1].definition = "0xf:0";
|
||||
},
|
||||
]) {
|
||||
const doc = diamond();
|
||||
mutate(doc.functions[0].atomic_facts);
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(model.code, "reconstructed C");
|
||||
assert.throws(() => model.select({}));
|
||||
}
|
||||
assert.throws(
|
||||
() => parseNativeIR(JSON.stringify(diamond()), "0x2000"),
|
||||
/subject/u,
|
||||
);
|
||||
});
|
||||
|
||||
test("typed slices retain symbolic input frontiers and precise definition bindings", () => {
|
||||
const model = parseNativeIR(JSON.stringify(diamond()), "0x1000");
|
||||
const result = model.slice({
|
||||
seeds: ["0x3:0", "0x0:0"],
|
||||
max_operations: 10,
|
||||
stop_at: [],
|
||||
stop_kinds: [],
|
||||
});
|
||||
assert.equal(result.operations.length, 4);
|
||||
assert.deepEqual(result.value_frontier, [
|
||||
{ value_id: "v0", reason: "function-input" },
|
||||
]);
|
||||
assert.equal(result.coverage.runtime_reachability, "unknown");
|
||||
});
|
||||
|
||||
test("native facet failure preserves C without fabricating an empty graph; source identity still gates both", () => {
|
||||
const doc = diamond();
|
||||
delete doc.functions[0].atomic_facts;
|
||||
doc.functions[0].atomic_facts_error = "atomic operation capacity exceeded";
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000", "/private/input");
|
||||
assert.equal(model.code, "reconstructed C");
|
||||
assert.equal(model.language, "c");
|
||||
assert.equal(model.coverage.operations, null);
|
||||
assert.equal(model.facets.code.availability, "available");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.throws(
|
||||
() => model.select({}),
|
||||
(error) =>
|
||||
error.code === "capability_unavailable" &&
|
||||
error.details.native_error === doc.functions[0].atomic_facts_error,
|
||||
);
|
||||
assert.throws(
|
||||
() => parseNativeIR(JSON.stringify(doc), "0x1000", "/other/input"),
|
||||
(error) => error.code === "subject_mismatch",
|
||||
);
|
||||
});
|
||||
|
||||
test("oversized or malformed native arrays fail before diagnostic expansion", () => {
|
||||
const capacity = diamond();
|
||||
capacity.functions[0].atomic_facts.operations = Array.from(
|
||||
{ length: 32001 },
|
||||
() => ({}),
|
||||
);
|
||||
assert.throws(
|
||||
() => parseNativeIR(JSON.stringify(capacity), "0x1000").select({}),
|
||||
/capacity/u,
|
||||
);
|
||||
const malformed = diamond();
|
||||
malformed.functions[0].atomic_facts.operations = Array.from(
|
||||
{ length: 12000 },
|
||||
() => ({}),
|
||||
);
|
||||
assert.throws(
|
||||
() => parseNativeIR(JSON.stringify(malformed), "0x1000").select({}),
|
||||
(error) =>
|
||||
error.code === "invalid_provider_output" &&
|
||||
error.details.issues.length <= 8 &&
|
||||
JSON.stringify(error.details).length < 2048,
|
||||
);
|
||||
});
|
||||
|
||||
test("control dependence handles multiple exits, loops, unreachable exits and bounded work", () => {
|
||||
const blocks = [
|
||||
{ id: 0, successors: [{ target: 1 }, { target: 2 }] },
|
||||
{ id: 1, successors: [] },
|
||||
{ id: 2, successors: [] },
|
||||
{ id: 3, successors: [{ target: 3 }] },
|
||||
];
|
||||
const result = controlDependencies(blocks);
|
||||
assert.deepEqual(
|
||||
result.relations.map((r) => [r.controller, r.dependent]),
|
||||
[
|
||||
[0, 1],
|
||||
[0, 2],
|
||||
],
|
||||
);
|
||||
assert.deepEqual(result.coverage.exit_unreachable_blocks, [3]);
|
||||
assert.throws(() => controlDependencies(blocks, { workLimit: 1 }), /budget/u);
|
||||
const linear = Array.from({ length: 12000 }, (_, id) => ({
|
||||
id,
|
||||
successors: id < 11999 ? [{ target: id + 1 }] : [],
|
||||
}));
|
||||
assert.equal(controlDependencies(linear).relations.length, 0);
|
||||
});
|
||||
|
||||
test("postdominator control dependence agrees with independently enumerated paths for every four-node DAG", () => {
|
||||
const possible = [
|
||||
[0, 1],
|
||||
[0, 2],
|
||||
[0, 3],
|
||||
[1, 2],
|
||||
[1, 3],
|
||||
[2, 3],
|
||||
];
|
||||
for (let mask = 0; mask < 64; mask++) {
|
||||
const blocks = Array.from({ length: 4 }, (_, id) => ({
|
||||
id,
|
||||
successors: possible
|
||||
.filter(([from], i) => from === id && mask & (1 << i))
|
||||
.map(([, target]) => ({ target })),
|
||||
}));
|
||||
const paths = (node) =>
|
||||
blocks[node].successors.length
|
||||
? blocks[node].successors.flatMap(({ target }) =>
|
||||
paths(target).map((path) => [node, ...path]),
|
||||
)
|
||||
: [[node]];
|
||||
const postdominators = blocks.map(({ id }) =>
|
||||
paths(id).reduce(
|
||||
(shared, path) =>
|
||||
new Set([...shared].filter((node) => path.includes(node))),
|
||||
new Set([0, 1, 2, 3]),
|
||||
),
|
||||
);
|
||||
const expected = blocks.flatMap((block) =>
|
||||
block.successors.flatMap((edge) =>
|
||||
[...postdominators[edge.target]]
|
||||
.filter((dependent) => !postdominators[block.id].has(dependent))
|
||||
.map((dependent) => `${block.id}:${edge.target}:${dependent}`),
|
||||
),
|
||||
);
|
||||
const actual = controlDependencies(blocks).relations.map(
|
||||
(row) => `${row.controller}:${row.edge.target}:${row.dependent}`,
|
||||
);
|
||||
assert.deepEqual(actual.sort(), expected.sort(), `DAG mask ${mask}`);
|
||||
}
|
||||
});
|
||||
@@ -1,31 +0,0 @@
|
||||
// One native operand/operation schema shared by complete and projected facts.
|
||||
import { z } from "zod";
|
||||
import {
|
||||
hex,
|
||||
ordinal,
|
||||
valueId,
|
||||
operationId,
|
||||
edgeId,
|
||||
rowParser,
|
||||
} from "./native-contract.mjs";
|
||||
export const operation = z.strictObject({
|
||||
id: operationId,
|
||||
address: hex,
|
||||
block: ordinal,
|
||||
opcode: z.string(),
|
||||
statement: z.string(),
|
||||
is_dead: z.boolean(),
|
||||
inputs: z.array(
|
||||
z.strictObject({
|
||||
input_index: ordinal,
|
||||
value_id: valueId.nullable(),
|
||||
phi_predecessor: ordinal.nullable(),
|
||||
phi_predecessor_edge: edgeId.nullable(),
|
||||
}),
|
||||
),
|
||||
output: valueId.nullable(),
|
||||
});
|
||||
|
||||
export const parseOperation = rowParser(operation, {
|
||||
inputs: (row) => operation.shape.inputs.element.parse(row),
|
||||
});
|
||||
@@ -1,312 +0,0 @@
|
||||
// Shared immutable SSA operands and queries; no CFG completeness is inferred.
|
||||
import { backwardSlice } from "../atomic-re-tools/kuna-ssa.mjs";
|
||||
import { freezeFacts as freeze } from "./fact-reads.mjs";
|
||||
import { requireFact } from "./native-contract.mjs";
|
||||
import {
|
||||
uniqueNative as unique,
|
||||
nativePage as page,
|
||||
readNativeIdentities as readIdentities,
|
||||
} from "./native-reads.mjs";
|
||||
export const DEFAULT_NATIVE_OPERATION_FIELDS = Object.freeze([
|
||||
"kind",
|
||||
"opcode",
|
||||
"output",
|
||||
"input_count",
|
||||
"definition_count",
|
||||
"constant_count",
|
||||
]);
|
||||
export const coverageIssues = (issues) => ({
|
||||
issues: issues.slice(0, 8).map((message) => message.slice(0, 160)),
|
||||
issue_count: issues.length,
|
||||
issues_omitted: Math.max(0, issues.length - 8),
|
||||
issue_messages_truncated: issues.slice(0, 8).some((s) => s.length > 160),
|
||||
});
|
||||
const index = (map, key, row) => {
|
||||
if (!map.has(key)) map.set(key, []);
|
||||
map.get(key).push(row);
|
||||
};
|
||||
const kind = (opcode) =>
|
||||
({
|
||||
CPUI_LOAD: "load",
|
||||
CPUI_STORE: "store",
|
||||
CPUI_MULTIEQUAL: "phi",
|
||||
CPUI_CALL: "call",
|
||||
CPUI_CALLIND: "call",
|
||||
CPUI_BRANCH: "branch",
|
||||
CPUI_CBRANCH: "branch",
|
||||
CPUI_BRANCHIND: "switch",
|
||||
CPUI_RETURN: "return",
|
||||
})[opcode] ?? "value";
|
||||
|
||||
function validateOperands(operations, values) {
|
||||
const referenced = new Set();
|
||||
for (const op of operations.values()) {
|
||||
requireFact(
|
||||
op.id.split(":")[0] === op.address,
|
||||
"Native operation origin/identity mismatch",
|
||||
);
|
||||
for (const [position, input] of op.inputs.entries()) {
|
||||
requireFact(
|
||||
input.input_index === position &&
|
||||
(input.value_id === null || values.has(input.value_id)),
|
||||
"Dangling or misindexed native operand",
|
||||
);
|
||||
requireFact(
|
||||
(input.phi_predecessor === null) ===
|
||||
(input.phi_predecessor_edge === null) &&
|
||||
(input.phi_predecessor === null ||
|
||||
(op.opcode === "CPUI_MULTIEQUAL" &&
|
||||
input.phi_predecessor_edge.startsWith(
|
||||
`${input.phi_predecessor}:`,
|
||||
))),
|
||||
"Native phi binding is inconsistent",
|
||||
);
|
||||
if (input.value_id !== null) referenced.add(input.value_id);
|
||||
}
|
||||
requireFact(
|
||||
op.output === null || values.get(op.output)?.definition === op.id,
|
||||
"Native output/definition mismatch",
|
||||
);
|
||||
if (op.output !== null) referenced.add(op.output);
|
||||
}
|
||||
requireFact(
|
||||
referenced.size === values.size,
|
||||
"Native operation set contains unrelated values",
|
||||
);
|
||||
}
|
||||
function validateDefinitions(operations, values, omittedIDs) {
|
||||
const omitted = new Set(omittedIDs),
|
||||
missing = new Set();
|
||||
requireFact(
|
||||
omitted.size === omittedIDs.length,
|
||||
"Duplicate native definition frontier",
|
||||
);
|
||||
for (const value of values.values())
|
||||
if (value.definition !== null) {
|
||||
if (!operations.has(value.definition)) missing.add(value.definition);
|
||||
requireFact(
|
||||
operations.get(value.definition)?.output === value.id ||
|
||||
(!operations.has(value.definition) && omitted.has(value.definition)),
|
||||
"Native value/definition mismatch",
|
||||
);
|
||||
}
|
||||
requireFact(
|
||||
missing.size === omitted.size,
|
||||
"Native definition frontier contains unrelated identities",
|
||||
);
|
||||
}
|
||||
|
||||
export function indexOperationFacts(rows, valueRows, omittedIDs) {
|
||||
const operations = unique(rows),
|
||||
values = unique(valueRows);
|
||||
validateOperands(operations, values);
|
||||
validateDefinitions(operations, values, omittedIDs);
|
||||
const users = new Map(),
|
||||
constants = new Map();
|
||||
const normalized = new Map();
|
||||
for (const op of operations.values()) {
|
||||
const definitions = [],
|
||||
literals = [];
|
||||
for (const input of op.inputs) {
|
||||
const value = values.get(input.value_id);
|
||||
if (value?.definition) {
|
||||
const relation = freeze({
|
||||
definition: value.definition,
|
||||
use: op.id,
|
||||
input_index: input.input_index,
|
||||
value_id: value.id,
|
||||
phi_predecessor: input.phi_predecessor,
|
||||
phi_predecessor_edge: input.phi_predecessor_edge,
|
||||
});
|
||||
definitions.push(relation);
|
||||
index(users, value.definition, relation);
|
||||
}
|
||||
if (value?.is_constant) {
|
||||
literals.push({
|
||||
value_id: value.id,
|
||||
input_index: input.input_index,
|
||||
value: value.offset,
|
||||
printed_size_bytes: value.size_bytes,
|
||||
is_annotation: value.is_annotation,
|
||||
});
|
||||
index(constants, value.offset, op.id);
|
||||
}
|
||||
}
|
||||
const { address: origin_address, ...payload } = op;
|
||||
normalized.set(
|
||||
op.id,
|
||||
freeze({
|
||||
...payload,
|
||||
origin_address,
|
||||
kind: kind(op.opcode),
|
||||
source_line: null,
|
||||
definitions,
|
||||
constants: literals,
|
||||
input_count: op.inputs.length,
|
||||
definition_count: definitions.length,
|
||||
constant_count: literals.length,
|
||||
}),
|
||||
);
|
||||
}
|
||||
return { operations, values, users, constants, normalized };
|
||||
}
|
||||
|
||||
export function createOperationViews(
|
||||
indexes,
|
||||
coverage,
|
||||
frontierReasons = new Map(),
|
||||
) {
|
||||
const { values, users, constants, normalized } = indexes;
|
||||
const model = {
|
||||
coverage,
|
||||
operation: (id) => normalized.get(id) ?? null,
|
||||
findConstant: (value) => [...new Set(constants.get(value) ?? [])],
|
||||
uses: (id) => users.get(id) ?? [],
|
||||
select: (input = {}) =>
|
||||
selectOperations(normalized, constants, {
|
||||
...input,
|
||||
fields: input.fields ?? DEFAULT_NATIVE_OPERATION_FIELDS,
|
||||
}),
|
||||
values: (input) => ({
|
||||
...readIdentities(
|
||||
values,
|
||||
input,
|
||||
(value) =>
|
||||
input.is_input === undefined || value.is_input === input.is_input,
|
||||
),
|
||||
coverage,
|
||||
}),
|
||||
operationRelations: (input) =>
|
||||
readOperationRelations(indexes, coverage, input),
|
||||
slice: (input) => sliceValues(model, values, input, frontierReasons),
|
||||
};
|
||||
return Object.freeze(model);
|
||||
}
|
||||
|
||||
function readOperationRelations(indexes, coverage, input) {
|
||||
const { normalized, values, users } = indexes;
|
||||
const seeds = [...new Set(input.seeds)];
|
||||
const operands = input.kind === "operands";
|
||||
const indices = input.input_indices && [...new Set(input.input_indices)];
|
||||
const groups = seeds.map((seed) => {
|
||||
const op = normalized.get(seed);
|
||||
const rows = operands
|
||||
? indices
|
||||
? indices.flatMap((index) => op?.inputs[index] ?? [])
|
||||
: (op?.inputs ?? [])
|
||||
: input.kind === "definitions"
|
||||
? (op?.definitions ?? [])
|
||||
: (users.get(seed) ?? []);
|
||||
return { seed, rows };
|
||||
});
|
||||
const offset = input.offset ?? 0,
|
||||
limit = input.limit ?? 20;
|
||||
let total = 0;
|
||||
const items = [];
|
||||
// Count groups without flattening every relation or copying unselected rows.
|
||||
for (const { seed, rows } of groups) {
|
||||
const start = Math.max(0, offset - total);
|
||||
const end = Math.min(rows.length, offset + limit - total);
|
||||
for (let i = start; i < end; i++) {
|
||||
const row = rows[i];
|
||||
items.push(
|
||||
operands
|
||||
? {
|
||||
operation: seed,
|
||||
...row,
|
||||
definition: values.get(row.value_id)?.definition ?? null,
|
||||
}
|
||||
: row,
|
||||
);
|
||||
}
|
||||
total += rows.length;
|
||||
}
|
||||
return {
|
||||
items,
|
||||
total,
|
||||
next_offset: offset + limit < total ? offset + limit : null,
|
||||
unretained_seeds: seeds.filter((id) => !normalized.has(id)),
|
||||
relation_coverage: {
|
||||
scope: operands
|
||||
? "all selected native input slots of retained operations; null values remain explicit"
|
||||
: input.kind === "uses"
|
||||
? (coverage.uses_scope ??
|
||||
"complete-for-retained-recovered-operations")
|
||||
: "definitions of retained operands; missing defining bodies remain unknown",
|
||||
...(operands
|
||||
? {
|
||||
slot_presence:
|
||||
"out-of-range slots of retained seeds are absent; unretained seeds remain unknown",
|
||||
order:
|
||||
"distinct caller seed order, then distinct requested input-index order or native slot order",
|
||||
}
|
||||
: {}),
|
||||
},
|
||||
};
|
||||
}
|
||||
|
||||
function selectOperations(normalized, constants, input = {}) {
|
||||
const ids = input.ids && new Set(input.ids),
|
||||
addresses = input.origin_addresses && new Set(input.origin_addresses),
|
||||
blockIds = input.block_ids && new Set(input.block_ids);
|
||||
const candidates = ids
|
||||
? [...ids].flatMap((id) => normalized.get(id) ?? [])
|
||||
: input.constant
|
||||
? [...new Set(constants.get(input.constant) ?? [])].map((id) =>
|
||||
normalized.get(id),
|
||||
)
|
||||
: normalized.values();
|
||||
return page(
|
||||
[...candidates].filter(
|
||||
(op) =>
|
||||
(!addresses || addresses.has(op.origin_address)) &&
|
||||
(!blockIds || blockIds.has(op.block)) &&
|
||||
(!input.constant ||
|
||||
op.constants.some((literal) => literal.value === input.constant)) &&
|
||||
(!input.kind || op.kind === input.kind),
|
||||
),
|
||||
input,
|
||||
["id", "origin_address", "block", "is_dead"],
|
||||
);
|
||||
}
|
||||
|
||||
function sliceValues(model, values, input, frontierReasons) {
|
||||
const result = backwardSlice(model, input.seeds, {
|
||||
maxOperations: input.max_operations,
|
||||
stopAt: input.stop_at,
|
||||
stopKinds: input.stop_kinds,
|
||||
});
|
||||
const frontier = new Map();
|
||||
for (const op of result.operations)
|
||||
for (const input of op.inputs) {
|
||||
const value = values.get(input.value_id);
|
||||
if (!value)
|
||||
frontier.set(`${op.id}:${input.input_index}`, {
|
||||
operation: op.id,
|
||||
input_index: input.input_index,
|
||||
reason: "native-null-input",
|
||||
});
|
||||
else if (!value.definition && !value.is_constant && !value.is_annotation)
|
||||
frontier.set(value.id, {
|
||||
value_id: value.id,
|
||||
reason: value.is_input ? "function-input" : "definition-unavailable",
|
||||
});
|
||||
}
|
||||
return {
|
||||
...result,
|
||||
value_frontier: [...frontier.values()],
|
||||
coverage: {
|
||||
...result.coverage,
|
||||
status:
|
||||
result.coverage.status === "complete-for-printed-definitions"
|
||||
? "complete-for-retained-value-definitions"
|
||||
: result.coverage.status,
|
||||
source_scope: model.coverage.scope,
|
||||
missing_reasons: result.coverage.missing.map((definition) => ({
|
||||
definition,
|
||||
reason: frontierReasons.get(definition) ?? "not-retained",
|
||||
})),
|
||||
relation_authority: "native SSA value identities and operand indexes",
|
||||
},
|
||||
};
|
||||
}
|
||||
@@ -1,68 +0,0 @@
|
||||
// Sparse facts retain their own resource, root and frontier contract.
|
||||
import { z } from "zod";
|
||||
import {
|
||||
operationId,
|
||||
value,
|
||||
rowParser,
|
||||
requireFact,
|
||||
readNativeFacet,
|
||||
} from "./native-contract.mjs";
|
||||
import { operation, parseOperation } from "./native-operation-contract.mjs";
|
||||
|
||||
const frontier = z.strictObject({
|
||||
definition: operationId,
|
||||
reason: z.enum(["operation-budget", "outside-recovered-membership"]),
|
||||
});
|
||||
const coverage = z.strictObject({
|
||||
scope: z.string(),
|
||||
operation_budget: z.literal(4096),
|
||||
frontier: z.array(frontier).max(12288),
|
||||
issues: z.array(z.string()).max(4096),
|
||||
machine_semantic_equivalence: z.literal("unknown"),
|
||||
});
|
||||
const schema = z.strictObject({
|
||||
schema_version: z.literal(1),
|
||||
roots: z.array(operationId).max(4096),
|
||||
operations: z.array(operation).max(4096),
|
||||
values: z.array(value).max(12288),
|
||||
coverage,
|
||||
});
|
||||
const parseCoverage = rowParser(coverage, {
|
||||
frontier: (row) => frontier.parse(row),
|
||||
issues: (row) => z.string().parse(row),
|
||||
});
|
||||
const parseProjection = rowParser(
|
||||
schema,
|
||||
{
|
||||
roots: (row) => operationId.parse(row),
|
||||
operations: parseOperation,
|
||||
values: (row) => value.parse(row),
|
||||
},
|
||||
{ coverage: parseCoverage },
|
||||
);
|
||||
export const readProjectionFacet = (fn) =>
|
||||
readNativeFacet(fn, "atomic_projection", "operation projection", (raw) => {
|
||||
requireFact(
|
||||
raw?.schema_version === 1,
|
||||
"Unsupported native projection schema",
|
||||
);
|
||||
for (const [rows, limit] of [
|
||||
[raw.roots, 4096],
|
||||
[raw.operations, 4096],
|
||||
[raw.values, 12288],
|
||||
[raw.coverage?.frontier, 12288],
|
||||
[raw.coverage?.issues, 4096],
|
||||
])
|
||||
requireFact(
|
||||
Array.isArray(rows) && rows.length <= limit,
|
||||
"Native projection exceeds capacity or has invalid shape",
|
||||
);
|
||||
requireFact(
|
||||
raw.operations.reduce(
|
||||
(n, op) => n + (Array.isArray(op?.inputs) ? op.inputs.length : 0),
|
||||
0,
|
||||
) <= 16384,
|
||||
"Native projection input capacity exceeded",
|
||||
);
|
||||
return parseProjection(raw);
|
||||
});
|
||||
@@ -1,85 +0,0 @@
|
||||
import { requireFact } from "./native-contract.mjs";
|
||||
import { freezeFacts as freeze } from "./fact-reads.mjs";
|
||||
import {
|
||||
indexOperationFacts,
|
||||
createOperationViews,
|
||||
coverageIssues,
|
||||
} from "./native-operation-views.mjs";
|
||||
|
||||
export function indexProjection(facts) {
|
||||
const reasons = new Map();
|
||||
for (const row of facts.coverage.frontier) {
|
||||
requireFact(!reasons.has(row.definition), "Duplicate projection frontier");
|
||||
reasons.set(row.definition, row.reason);
|
||||
requireFact(
|
||||
row.reason !== "operation-budget" ||
|
||||
facts.operations.length === facts.coverage.operation_budget,
|
||||
"Projection claims an unexhausted operation budget",
|
||||
);
|
||||
}
|
||||
const index = indexOperationFacts(facts.operations, facts.values, [
|
||||
...reasons.keys(),
|
||||
]);
|
||||
const roots = new Set(facts.roots);
|
||||
requireFact(roots.size === facts.roots.length, "Duplicate projection root");
|
||||
const callIDs = [...index.operations.values()]
|
||||
.filter((op) => ["CPUI_CALL", "CPUI_CALLIND"].includes(op.opcode))
|
||||
.map((op) => op.id);
|
||||
requireFact(
|
||||
roots.size === callIDs.length && callIDs.every((id) => roots.has(id)),
|
||||
"Projection roots differ from retained call operations",
|
||||
);
|
||||
const reached = new Set(roots),
|
||||
queue = [...roots];
|
||||
for (const id of queue)
|
||||
for (const edge of index.normalized.get(id).definitions)
|
||||
if (
|
||||
index.operations.has(edge.definition) &&
|
||||
!reached.has(edge.definition)
|
||||
) {
|
||||
reached.add(edge.definition);
|
||||
queue.push(edge.definition);
|
||||
}
|
||||
requireFact(
|
||||
reached.size === index.operations.size,
|
||||
"Projection contains operations unrelated to its roots",
|
||||
);
|
||||
return { ...index, reasons };
|
||||
}
|
||||
|
||||
export function createProjectionViews(facts, code, indexes, unavailableGraph) {
|
||||
freeze(facts);
|
||||
const coverage = freeze({
|
||||
scope: facts.coverage.scope,
|
||||
operations: indexes.operations.size,
|
||||
values: indexes.values.size,
|
||||
blocks: null,
|
||||
root_count: facts.roots.length,
|
||||
operation_budget: facts.coverage.operation_budget,
|
||||
definition_frontier_count: facts.coverage.frontier.length,
|
||||
...coverageIssues(facts.coverage.issues),
|
||||
control_flow: "not-collected",
|
||||
definition_scope:
|
||||
"retained operand definitions; unexpanded IDs stay unknown",
|
||||
uses_scope: "selected-operations-only; no complete-use claim",
|
||||
value_scope:
|
||||
"operands and outputs of the retained call-rooted operation set",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
memory_alias_dependencies: "unknown",
|
||||
runtime_successors: "unknown",
|
||||
representation:
|
||||
"optimized recovered IR; origins and block IDs do not prove machine extents or branch targets",
|
||||
});
|
||||
const core = createOperationViews(indexes, coverage, indexes.reasons);
|
||||
return Object.freeze({
|
||||
...core,
|
||||
code,
|
||||
block: unavailableGraph,
|
||||
incoming: unavailableGraph,
|
||||
blocks: unavailableGraph,
|
||||
edges: (input) =>
|
||||
["definitions", "uses", "operands"].includes(input.kind)
|
||||
? core.operationRelations(input)
|
||||
: unavailableGraph(),
|
||||
});
|
||||
}
|
||||
@@ -1,318 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
|
||||
const value = (id, definition, extra = {}) => ({
|
||||
id,
|
||||
definition,
|
||||
space: "register",
|
||||
offset: "0x30",
|
||||
size_bytes: 8,
|
||||
is_constant: false,
|
||||
is_input: false,
|
||||
is_annotation: false,
|
||||
type_name: "",
|
||||
...extra,
|
||||
});
|
||||
const input = (input_index, value_id, predecessor = null, edge = null) => ({
|
||||
input_index,
|
||||
value_id,
|
||||
phi_predecessor: predecessor,
|
||||
phi_predecessor_edge: edge,
|
||||
});
|
||||
const op = (id, opcode, inputs, output = null) => ({
|
||||
id,
|
||||
address: id.split(":")[0],
|
||||
block: 9,
|
||||
opcode,
|
||||
inputs,
|
||||
output,
|
||||
statement: "unparseable display",
|
||||
is_dead: false,
|
||||
});
|
||||
function fixture() {
|
||||
const operations = [
|
||||
op("0x1100:1", "CPUI_CALL", [
|
||||
input(0, "v0"),
|
||||
input(1, "v3"),
|
||||
input(2, "v7"),
|
||||
input(3, "v9"),
|
||||
]),
|
||||
op(
|
||||
"0x1080:2",
|
||||
"CPUI_MULTIEQUAL",
|
||||
[input(0, "v1", 7, "7:1"), input(1, "v2", 8, "8:0")],
|
||||
"v3",
|
||||
),
|
||||
op("0x1050:3", "CPUI_PTRSUB", [input(0, "v4"), input(1, "v5")], "v1"),
|
||||
op("0x1060:4", "CPUI_PTRSUB", [input(0, "v4"), input(1, "v6")], "v2"),
|
||||
];
|
||||
const values = [
|
||||
value("v0", null, {
|
||||
space: "fspec",
|
||||
offset: "0x10000",
|
||||
is_annotation: true,
|
||||
}),
|
||||
value("v1", "0x1050:3"),
|
||||
value("v2", "0x1060:4"),
|
||||
value("v3", "0x1080:2"),
|
||||
value("v4", null, { space: "const", offset: "0x0", is_constant: true }),
|
||||
value("v5", null, { space: "const", offset: "0x2100", is_constant: true }),
|
||||
value("v6", null, { space: "const", offset: "0x2200", is_constant: true }),
|
||||
value("v7", null, { is_input: true }),
|
||||
value("v9", "0x9000:0"),
|
||||
];
|
||||
return {
|
||||
binary: "/input",
|
||||
count: 1,
|
||||
language: "c",
|
||||
error: null,
|
||||
functions: [
|
||||
{
|
||||
address_hex: "0x1000",
|
||||
code: "retained C",
|
||||
error: null,
|
||||
atomic_facts_error: "atomic operation capacity exceeded",
|
||||
atomic_projection: {
|
||||
schema_version: 1,
|
||||
roots: ["0x1100:1"],
|
||||
operations,
|
||||
values,
|
||||
coverage: {
|
||||
scope: "call-rooted bounded backward SSA definitions",
|
||||
operation_budget: 4096,
|
||||
frontier: [
|
||||
{
|
||||
definition: "0x9000:0",
|
||||
reason: "outside-recovered-membership",
|
||||
},
|
||||
],
|
||||
issues: [],
|
||||
machine_semantic_equivalence: "unknown",
|
||||
},
|
||||
},
|
||||
atomic_calls: {
|
||||
schema_version: 1,
|
||||
items: [
|
||||
{
|
||||
id: "0x1100:1",
|
||||
origin_address: "0x1100",
|
||||
block: 9,
|
||||
opcode: "CPUI_CALL",
|
||||
is_dead: false,
|
||||
target_value_id: "v0",
|
||||
spec_available: false,
|
||||
target: null,
|
||||
target_name: null,
|
||||
arguments: ["v3", "v7", "v9"].map((value_id, i) => ({
|
||||
input_index: i + 1,
|
||||
value_id,
|
||||
storage: null,
|
||||
storage_authority: null,
|
||||
})),
|
||||
output: null,
|
||||
prototype: null,
|
||||
stack_offset: null,
|
||||
},
|
||||
],
|
||||
values: structuredClone(
|
||||
values.filter((v) => ["v0", "v3", "v7", "v9"].includes(v.id)),
|
||||
),
|
||||
coverage: {
|
||||
scope: "all recovered block-member CALL/CALLIND operations",
|
||||
blocks_examined: 10,
|
||||
operations_examined: 100,
|
||||
value_scope: "call inputs including target and outputs only",
|
||||
machine_call_equivalence: "unknown",
|
||||
},
|
||||
},
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
test("operand relations preserve every slot, page across seeds and select exact caller indices", () => {
|
||||
const source = fixture();
|
||||
source.functions[0].atomic_projection.operations[1].inputs.push(
|
||||
input(2, "v5", 2, "2:0"),
|
||||
input(3, "v7", 3, "3:0"),
|
||||
input(4, null, 4, "4:0"),
|
||||
);
|
||||
const model = parseNativeIR(JSON.stringify(source), "0x1000");
|
||||
const read = (extra = {}) =>
|
||||
model.edges({
|
||||
kind: "operands",
|
||||
seeds: ["0x1080:2"],
|
||||
offset: 0,
|
||||
limit: 2,
|
||||
...extra,
|
||||
});
|
||||
const first = read();
|
||||
assert.equal(first.total, 5);
|
||||
assert.equal(first.next_offset, 2);
|
||||
assert.deepEqual(
|
||||
first.items.map((r) => r.input_index),
|
||||
[0, 1],
|
||||
);
|
||||
assert.deepEqual(
|
||||
read({ offset: 2, limit: 10 }).items.map((r) => r.definition),
|
||||
[null, null, null],
|
||||
);
|
||||
const selected = read({ input_indices: [4, 2, 4, 500], limit: 10 });
|
||||
assert.equal(selected.total, 2);
|
||||
assert.equal(selected.items[0].value_id, null);
|
||||
assert.deepEqual(
|
||||
selected.items.map((r) => [r.input_index, r.phi_predecessor_edge]),
|
||||
[
|
||||
[4, "4:0"],
|
||||
[2, "2:0"],
|
||||
],
|
||||
);
|
||||
assert.equal(read({ input_indices: [] }).total, 0);
|
||||
const batch = read({
|
||||
seeds: ["0x1080:2", "0x1050:3", "0x1050:3", "0x9000:0"],
|
||||
offset: 4,
|
||||
limit: 2,
|
||||
});
|
||||
assert.equal(batch.total, 7);
|
||||
assert.equal(batch.next_offset, 6);
|
||||
assert.deepEqual(
|
||||
batch.items.map((r) => [r.operation, r.input_index]),
|
||||
[
|
||||
["0x1080:2", 4],
|
||||
["0x1050:3", 0],
|
||||
],
|
||||
);
|
||||
assert.deepEqual(batch.unretained_seeds, ["0x9000:0"]);
|
||||
assert.equal(
|
||||
read({ seeds: ["0x1100:1"], input_indices: [3] }).items[0].definition,
|
||||
"0x9000:0",
|
||||
);
|
||||
const summary = model.select({ ids: ["0x1080:2"] }).items[0];
|
||||
assert.equal(summary.input_count, 5);
|
||||
assert.equal(summary.definition_count, 2);
|
||||
assert.equal(summary.constant_count, 1);
|
||||
assert.equal(Object.hasOwn(summary, "inputs"), false);
|
||||
const repeatedValues = model.values({ ids: ["v3", "v3", "vbad", "vbad"] });
|
||||
assert.deepEqual(
|
||||
repeatedValues.items.map((v) => v.id),
|
||||
["v3"],
|
||||
);
|
||||
assert.deepEqual(repeatedValues.missing, ["vbad"]);
|
||||
});
|
||||
|
||||
test("partial definition facts compose through phi operands without implying a CFG or full uses", () => {
|
||||
const model = parseNativeIR(JSON.stringify(fixture()), "0x1000");
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(
|
||||
model.facets.operations.scope,
|
||||
"bounded-call-rooted-definitions",
|
||||
);
|
||||
assert.equal(model.facets.values.scope, "call-rooted-definitions");
|
||||
assert.equal(model.coverage.definition_frontier_count, 1);
|
||||
assert.equal(JSON.stringify(model.coverage).includes("0x9000:0"), false);
|
||||
const phi = model.select({
|
||||
ids: ["0x1080:2"],
|
||||
fields: ["inputs", "output"],
|
||||
limit: 1,
|
||||
}).items[0];
|
||||
assert.equal(phi.output, "v3");
|
||||
assert.equal(phi.inputs[0].phi_predecessor_edge, "7:1");
|
||||
assert.throws(() => {
|
||||
phi.inputs[0].value_id = "vdead";
|
||||
}, TypeError);
|
||||
const sliced = model.slice({
|
||||
seeds: ["0x1100:1"],
|
||||
max_operations: 10,
|
||||
stop_kinds: [],
|
||||
});
|
||||
assert.equal(sliced.operations.length, 4);
|
||||
assert.equal(sliced.coverage.status, "partial");
|
||||
assert.deepEqual(sliced.coverage.missing_reasons, [
|
||||
{ definition: "0x9000:0", reason: "outside-recovered-membership" },
|
||||
]);
|
||||
assert.deepEqual(sliced.value_frontier, [
|
||||
{ value_id: "v7", reason: "function-input" },
|
||||
]);
|
||||
assert.deepEqual(model.findConstant("0x2100"), ["0x1050:3"]);
|
||||
const uses = model.edges({
|
||||
kind: "uses",
|
||||
seeds: ["0x9000:0"],
|
||||
offset: 0,
|
||||
limit: 10,
|
||||
});
|
||||
assert.equal(uses.items.length, 1);
|
||||
assert.deepEqual(uses.unretained_seeds, ["0x9000:0"]);
|
||||
assert.match(uses.relation_coverage.scope, /selected-operations-only/);
|
||||
for (const read of [
|
||||
() => model.blocks({}),
|
||||
() => model.edges({ kind: "controllers", seeds: [9] }),
|
||||
])
|
||||
assert.throws(read, (e) => e.code === "capability_unavailable");
|
||||
});
|
||||
|
||||
test("bad or contradictory projections preserve independently valid call facts and code", () => {
|
||||
const changes = [
|
||||
(p) => {
|
||||
p.operations[1].inputs[0].value_id = "vdead";
|
||||
},
|
||||
(p) => {
|
||||
p.coverage.frontier = [];
|
||||
},
|
||||
(p) => {
|
||||
p.coverage.frontier[0].reason = "operation-budget";
|
||||
},
|
||||
(p) => {
|
||||
p.roots = ["0x1050:3"];
|
||||
},
|
||||
(p) => {
|
||||
p.roots.push("0x1100:1");
|
||||
},
|
||||
(p) => {
|
||||
p.operations.push(op("0x1070:5", "CPUI_COPY", [input(0, "v7")]));
|
||||
},
|
||||
(p) => {
|
||||
p.values[3].offset = "0xbad";
|
||||
},
|
||||
(p) => {
|
||||
p.coverage.frontier = Array(12289).fill(null);
|
||||
},
|
||||
];
|
||||
for (const change of changes) {
|
||||
const doc = fixture();
|
||||
change(doc.functions[0].atomic_projection);
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.code, "retained C");
|
||||
assert.equal(model.facets.operations.availability, "unavailable");
|
||||
assert.equal(model.facets.calls.availability, "available");
|
||||
assert.equal(model.facets.values.scope, "call-boundary-only");
|
||||
assert.ok(JSON.stringify(model.facets.operations.error).length < 4096);
|
||||
}
|
||||
});
|
||||
|
||||
test("bad call facets do not erase valid projection definitions", () => {
|
||||
const doc = fixture();
|
||||
doc.functions[0].atomic_calls.items[0].arguments[0].value_id = "vdead";
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.facets.calls.availability, "unavailable");
|
||||
assert.equal(model.facets.operations.availability, "available");
|
||||
assert.equal(model.values({ ids: ["v5"] }).items[0].offset, "0x2100");
|
||||
});
|
||||
|
||||
test("compact coverage bounds issue text while slices retain the actual issue count", () => {
|
||||
const doc = fixture();
|
||||
doc.functions[0].atomic_projection.coverage.issues = Array(10).fill(
|
||||
"x".repeat(10000),
|
||||
);
|
||||
const model = parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
assert.equal(model.coverage.issue_count, 10);
|
||||
assert.equal(model.coverage.issues.length, 8);
|
||||
assert.equal(model.coverage.issues_omitted, 2);
|
||||
assert.equal(model.coverage.issue_messages_truncated, true);
|
||||
assert.ok(JSON.stringify(model.coverage).length < 4096);
|
||||
assert.equal(
|
||||
model.slice({ seeds: ["0x1100:1"], max_operations: 10 }).coverage
|
||||
.source_issues,
|
||||
10,
|
||||
);
|
||||
});
|
||||
@@ -1,28 +0,0 @@
|
||||
import { pageFacts } from "./fact-reads.mjs";
|
||||
import { requireFact } from "./native-contract.mjs";
|
||||
|
||||
export const nativePage = (
|
||||
rows,
|
||||
input = {},
|
||||
identity = ["id", "origin_address", "block"],
|
||||
) => pageFacts(rows, { offset: 0, limit: 20, ...input }, identity);
|
||||
export const uniqueNative = (rows) => {
|
||||
const index = new Map();
|
||||
for (const row of rows) {
|
||||
requireFact(!index.has(row.id), `Duplicate native identity ${row.id}`);
|
||||
index.set(row.id, row);
|
||||
}
|
||||
return index;
|
||||
};
|
||||
export const readNativeIdentities = (index, input, filter = () => true) => ({
|
||||
...nativePage(
|
||||
(input.ids
|
||||
? [...new Set(input.ids)].flatMap((id) => index.get(id) ?? [])
|
||||
: [...index.values()]
|
||||
).filter(filter),
|
||||
input,
|
||||
),
|
||||
missing: input.ids
|
||||
? [...new Set(input.ids)].filter((id) => !index.has(id))
|
||||
: [],
|
||||
});
|
||||
@@ -1,50 +0,0 @@
|
||||
const read = async (name,input) => {
|
||||
const result=await rea[name](input);
|
||||
if(!result.ok)throw new Error(JSON.stringify(result.error));
|
||||
return result.data;
|
||||
};
|
||||
const complete=(page)=>{
|
||||
if(page.next_offset!==null)throw new Error("Continue the selected fact page before drawing a complete conclusion");
|
||||
return page;
|
||||
};
|
||||
const found=await read("functions",{query:"sqlite3VdbeExec",exact:true});
|
||||
complete(found);
|
||||
const formats=[];
|
||||
for(const query of ["cannot open savepoint - SQL statements in progress","cannot release savepoint - SQL statements in progress"]) {
|
||||
const strings=await read("strings",{query});
|
||||
complete(strings);
|
||||
const addresses=[...new Set(strings.items.filter((s)=>s.value===query).map((s)=>s.address))];
|
||||
if(addresses.length!==1)throw new Error("Ambiguous format: "+JSON.stringify(strings));
|
||||
const references=await read("references",{addresses,direction:"incoming",fields:["fromFunction","instruction"]});
|
||||
for(const result of references.results) {
|
||||
if(!result.ok)throw new Error(JSON.stringify(result.error));
|
||||
complete(result);
|
||||
}
|
||||
formats.push({query,address:addresses[0],string_evidence_id:strings.source.evidence_id,references});
|
||||
}
|
||||
const owners=formats.map((f)=>new Set(f.references.results.flatMap((r)=>r.ok?r.items.map((ref)=>ref.fromFunction?.address).filter(Boolean):[])));
|
||||
const candidates=found.items.filter((f)=>owners.every((set)=>set.has(f.address)));
|
||||
if(candidates.length!==1)throw new Error("Unresolved format-owning function: "+JSON.stringify({found,formats}));
|
||||
const capture=await read("snapshot",{entry:candidates[0].address});
|
||||
const snapshot=capture.snapshot;
|
||||
for(const format of formats) {
|
||||
const producers=await read("operations",{snapshot,constant:format.address,fields:["opcode","output"]});
|
||||
complete(producers);
|
||||
const setup=producers.items.filter((op)=>op.opcode==="CPUI_PTRSUB");
|
||||
if(setup.length!==1)throw new Error("Ambiguous format producer: "+JSON.stringify(producers));
|
||||
const uses=await read("relations",{snapshot,kind:"uses",seeds:[setup[0].id]});
|
||||
complete(uses);
|
||||
Object.assign(format,{producer:setup[0],uses});
|
||||
}
|
||||
const controls=await read("relations",{snapshot,kind:"controllers",seeds:formats.map((f)=>f.producer.block)});
|
||||
complete(controls);
|
||||
const branchIDs=[...new Set(controls.items.flatMap((relation)=>relation.edge.condition_value?[relation.edge.condition_value]:[]))];
|
||||
const conditions=await read("values",{snapshot,ids:branchIDs});
|
||||
complete(conditions);
|
||||
const definitions=await read("operations",{snapshot,ids:conditions.items.map((v)=>v.definition).filter(Boolean),fields:["opcode","inputs","output","constants"]});
|
||||
complete(definitions);
|
||||
const values=await read("values",{snapshot,ids:[...new Set(definitions.items.flatMap((op)=>op.inputs.map((i)=>i.value_id).filter(Boolean)))]});
|
||||
complete(values);
|
||||
const evidence=await read("evidence",{evidence_id:capture.source.evidence_id,component:"recovered_ir",limit:128});
|
||||
if((await read("snapshot",{entry:candidates[0].address})).snapshot!==snapshot)throw new Error("Capture was not reused");
|
||||
return {capture,formats,controls,conditions,definitions,values,evidence};
|
||||
@@ -1,104 +0,0 @@
|
||||
// Native switch framing/rows are admitted independently of graph and calls.
|
||||
import { z } from "zod";
|
||||
import {
|
||||
hex,
|
||||
ordinal,
|
||||
edgeId,
|
||||
operationId,
|
||||
value,
|
||||
rowParser,
|
||||
readNativeFacet,
|
||||
requireFact,
|
||||
} from "./native-contract.mjs";
|
||||
|
||||
const edge = z.strictObject({ id: edgeId, index: ordinal, target: ordinal });
|
||||
const location = z.strictObject({ space: z.string(), offset: hex });
|
||||
const row = z.strictObject({
|
||||
index: ordinal,
|
||||
label: z
|
||||
.strictObject({ status: z.enum(["value", "no-label"]), raw_bits: hex })
|
||||
.nullable(),
|
||||
recovered_target: location.nullable(),
|
||||
edges: z.array(edge),
|
||||
});
|
||||
const table = z.strictObject({
|
||||
partial: z.boolean(),
|
||||
labelled: z.boolean(),
|
||||
consumed_mask: hex,
|
||||
signed_render_hint: z.boolean(),
|
||||
default_raw_index: z.number().int().min(-1).max(0x7fffffff),
|
||||
default_edge: edge.nullable(),
|
||||
default_is_folded: z.boolean(),
|
||||
rows: z.array(row),
|
||||
});
|
||||
const item = z.strictObject({
|
||||
id: operationId,
|
||||
origin_address: hex,
|
||||
block: ordinal,
|
||||
selector: value.nullable(),
|
||||
table: table.nullable(),
|
||||
});
|
||||
const facet = z.strictObject({
|
||||
schema_version: z.literal(1),
|
||||
items: z.array(item).max(12000),
|
||||
coverage: z.strictObject({
|
||||
scope: z.string(),
|
||||
blocks_examined: ordinal.min(1),
|
||||
operations_examined: z
|
||||
.number()
|
||||
.int()
|
||||
.nonnegative()
|
||||
.max(Number.MAX_SAFE_INTEGER),
|
||||
machine_semantic_equivalence: z.literal("unknown"),
|
||||
}),
|
||||
});
|
||||
const parseTable = rowParser(table, {
|
||||
rows: rowParser(row, { edges: (v) => edge.parse(v) }),
|
||||
});
|
||||
const parseItem = rowParser(
|
||||
item,
|
||||
{},
|
||||
{ table: (v) => (v === null ? null : parseTable(v)) },
|
||||
);
|
||||
const parseFacet = rowParser(facet, { items: parseItem });
|
||||
|
||||
export const readSwitchFacet = (fn) =>
|
||||
readNativeFacet(fn, "atomic_switches", "recovered switches", (raw) => {
|
||||
requireFact(
|
||||
raw?.schema_version === 1,
|
||||
"Unsupported native switch facet schema; expected 1",
|
||||
);
|
||||
requireFact(
|
||||
Array.isArray(raw.items) && raw.items.length <= 12000,
|
||||
"Native switch count exceeds capacity or has invalid shape",
|
||||
);
|
||||
let rows = 0,
|
||||
mappings = 0,
|
||||
text = 0;
|
||||
for (const item of raw.items) {
|
||||
for (const s of [item?.selector?.space, item?.selector?.type_name])
|
||||
if (typeof s === "string") text += Buffer.byteLength(s);
|
||||
if (item?.table === null) continue;
|
||||
requireFact(
|
||||
Array.isArray(item?.table?.rows),
|
||||
"Native switch rows must be an array",
|
||||
);
|
||||
rows += item.table.rows.length;
|
||||
requireFact(rows <= 100000, "Native switch rows exceed capacity");
|
||||
for (const row of item.table.rows) {
|
||||
requireFact(
|
||||
Array.isArray(row?.edges),
|
||||
"Native case edges must be an array",
|
||||
);
|
||||
mappings += row.edges.length;
|
||||
requireFact(
|
||||
mappings <= 100000,
|
||||
"Native switch mappings exceed capacity",
|
||||
);
|
||||
if (typeof row?.recovered_target?.space === "string")
|
||||
text += Buffer.byteLength(row.recovered_target.space);
|
||||
}
|
||||
}
|
||||
requireFact(text <= 2 * 1024 * 1024, "Native switch text exceeds capacity");
|
||||
return parseFacet(raw);
|
||||
});
|
||||
@@ -1,320 +0,0 @@
|
||||
// Native table ownership is separate from optimized CFG and machine transfers.
|
||||
import { requireFact, requireSameNativeValue } from "./native-contract.mjs";
|
||||
import { uniqueNative, readNativeIdentities } from "./native-reads.mjs";
|
||||
import { freezeFacts } from "./fact-reads.mjs";
|
||||
|
||||
function validateEdge(edge, block) {
|
||||
requireFact(
|
||||
edge.id === `${block}:${edge.index}`,
|
||||
"Native switch edge identity mismatch",
|
||||
);
|
||||
}
|
||||
function indexTable(sw) {
|
||||
const { rows, ...metadata } = sw.table;
|
||||
requireFact(
|
||||
(metadata.default_raw_index === -1) === (metadata.default_edge === null),
|
||||
"Native switch default availability mismatch",
|
||||
);
|
||||
if (metadata.default_edge) {
|
||||
validateEdge(metadata.default_edge, sw.block);
|
||||
requireFact(
|
||||
metadata.default_edge.index === metadata.default_raw_index,
|
||||
"Native switch default index mismatch",
|
||||
);
|
||||
}
|
||||
const counts = {
|
||||
case_count: rows.length,
|
||||
mapping_count: 0,
|
||||
known_label_count: 0,
|
||||
missing_label_count: 0,
|
||||
no_label_count: 0,
|
||||
unmapped_row_count: 0,
|
||||
unknown_target_count: 0,
|
||||
};
|
||||
const byLabel = new Map();
|
||||
for (const [i, row] of rows.entries()) {
|
||||
requireFact(row.index === i, "Native case row ordinal mismatch");
|
||||
const edges = new Set();
|
||||
for (const edge of row.edges) {
|
||||
validateEdge(edge, sw.block);
|
||||
requireFact(
|
||||
!edges.has(edge.id),
|
||||
"Duplicate native case/out-edge binding",
|
||||
);
|
||||
edges.add(edge.id);
|
||||
}
|
||||
if (row.label === null) counts.missing_label_count++;
|
||||
else if (row.label.status === "no-label") counts.no_label_count++;
|
||||
else {
|
||||
counts.known_label_count++;
|
||||
if (!byLabel.has(row.label.raw_bits)) byLabel.set(row.label.raw_bits, []);
|
||||
byLabel.get(row.label.raw_bits).push(row);
|
||||
}
|
||||
counts.mapping_count += row.edges.length;
|
||||
if (row.edges.length === 0) counts.unmapped_row_count++;
|
||||
if (row.recovered_target === null) counts.unknown_target_count++;
|
||||
}
|
||||
const hasRawLabels = counts.missing_label_count < rows.length;
|
||||
requireFact(
|
||||
metadata.labelled === hasRawLabels,
|
||||
"Native switch label availability mismatch",
|
||||
);
|
||||
const prefix = caseOffsets(rows);
|
||||
const table = { ...metadata, ...counts, relation_count: prefix.at(-1) };
|
||||
// Retain source rows and label references, not a second full row projection.
|
||||
// A numeric row lookup is O(1); selected label reads use this index.
|
||||
return { table, lookup: { rows, prefix, byLabel, labelOffsets: new Map() } };
|
||||
}
|
||||
|
||||
export function indexSwitchFacet(facts) {
|
||||
freezeFacts(facts);
|
||||
const switches = uniqueNative(facts.items),
|
||||
summaries = new Map(),
|
||||
cases = new Map();
|
||||
requireFact(
|
||||
switches.size <= facts.coverage.operations_examined,
|
||||
"Native switch count exceeds examined operations",
|
||||
);
|
||||
for (const sw of switches.values()) {
|
||||
requireFact(
|
||||
sw.id.split(":")[0] === sw.origin_address,
|
||||
"Native switch origin/identity mismatch",
|
||||
);
|
||||
let table = null;
|
||||
if (sw.table) {
|
||||
const indexed = indexTable(sw);
|
||||
table = indexed.table;
|
||||
cases.set(sw.id, indexed.lookup);
|
||||
}
|
||||
summaries.set(
|
||||
sw.id,
|
||||
freezeFacts({
|
||||
id: sw.id,
|
||||
origin_address: sw.origin_address,
|
||||
block: sw.block,
|
||||
selector: sw.selector,
|
||||
table,
|
||||
}),
|
||||
);
|
||||
}
|
||||
return { switches, summaries, cases };
|
||||
}
|
||||
export function validateSwitchGraph(index, graph) {
|
||||
for (const sw of index.switches.values()) {
|
||||
const op = graph.operations.get(sw.id),
|
||||
block = graph.blocks.get(sw.block);
|
||||
requireFact(
|
||||
op?.opcode === "CPUI_BRANCHIND" &&
|
||||
op.block === sw.block &&
|
||||
op.address === sw.origin_address &&
|
||||
block,
|
||||
"Native switch/graph ownership mismatch",
|
||||
);
|
||||
requireFact(
|
||||
(op.inputs[0]?.value_id ?? null) === (sw.selector?.id ?? null),
|
||||
"Native switch/graph selector mismatch",
|
||||
);
|
||||
if (sw.selector)
|
||||
requireSameNativeValue(
|
||||
sw.selector,
|
||||
graph.values.get(sw.selector.id),
|
||||
"Native switch/graph selector value mismatch",
|
||||
);
|
||||
const checkEdge = (edge) =>
|
||||
requireFact(
|
||||
block.successors[edge.index]?.id === edge.id &&
|
||||
block.successors[edge.index]?.target === edge.target,
|
||||
"Native switch/graph edge mapping mismatch",
|
||||
);
|
||||
if (sw.table?.default_edge) checkEdge(sw.table.default_edge);
|
||||
for (const row of sw.table?.rows ?? [])
|
||||
for (const edge of row.edges) checkEdge(edge);
|
||||
}
|
||||
for (const op of graph.operations.values())
|
||||
requireFact(
|
||||
op.opcode !== "CPUI_BRANCHIND" || index.switches.has(op.id),
|
||||
"Native indirect branch omitted from switch coverage",
|
||||
);
|
||||
}
|
||||
export function validateSwitchValues(index, values) {
|
||||
if (!values) return;
|
||||
for (const sw of index.switches.values())
|
||||
if (sw.selector && values.has(sw.selector.id))
|
||||
requireSameNativeValue(
|
||||
sw.selector,
|
||||
values.get(sw.selector.id),
|
||||
"Native switch/shared value mismatch",
|
||||
);
|
||||
}
|
||||
function caseOffsets(rows) {
|
||||
const prefix = new Uint32Array(rows.length + 1);
|
||||
for (let i = 0; i < rows.length; i++)
|
||||
prefix[i + 1] = prefix[i] + Math.max(1, rows[i].edges.length);
|
||||
return prefix;
|
||||
}
|
||||
function selectCaseRows(lookup, labels, indices) {
|
||||
if (!labels && !indices) return lookup;
|
||||
if (!indices && labels?.size === 1) {
|
||||
const label = labels.values().next().value;
|
||||
const rows = lookup.byLabel.get(label) ?? [];
|
||||
if (!lookup.labelOffsets.has(label) && lookup.byLabel.has(label))
|
||||
lookup.labelOffsets.set(label, caseOffsets(rows));
|
||||
return {
|
||||
rows,
|
||||
prefix: lookup.labelOffsets.get(label) ?? new Uint32Array(1),
|
||||
};
|
||||
}
|
||||
const rows = indices
|
||||
? indices
|
||||
.flatMap((i) => lookup.rows[i] ?? [])
|
||||
.filter(
|
||||
(r) =>
|
||||
!labels ||
|
||||
(r.label?.status === "value" && labels.has(r.label.raw_bits)),
|
||||
)
|
||||
: [...labels]
|
||||
.flatMap((label) => lookup.byLabel.get(label) ?? [])
|
||||
.sort((a, b) => a.index - b.index);
|
||||
return { rows, prefix: caseOffsets(rows) };
|
||||
}
|
||||
function firstCaseAt(prefix, offset) {
|
||||
let low = 0,
|
||||
high = prefix.length - 1;
|
||||
while (low < high) {
|
||||
const mid = (low + high) >>> 1;
|
||||
if (prefix[mid + 1] <= offset) low = mid + 1;
|
||||
else high = mid;
|
||||
}
|
||||
return low;
|
||||
}
|
||||
function casePage(sw, selected, start, end) {
|
||||
const { id, block } = sw;
|
||||
const items = [];
|
||||
for (
|
||||
let i = firstCaseAt(selected.prefix, start);
|
||||
i < selected.rows.length && selected.prefix[i] < end;
|
||||
i++
|
||||
) {
|
||||
const r = selected.rows[i],
|
||||
case_id = `${id}:case:${r.index}`;
|
||||
const begin = Math.max(0, start - selected.prefix[i]);
|
||||
const finish = Math.min(
|
||||
Math.max(1, r.edges.length),
|
||||
end - selected.prefix[i],
|
||||
);
|
||||
for (let j = begin; j < finish; j++) {
|
||||
const edge = r.edges[j] ?? null;
|
||||
items.push({
|
||||
id: `${case_id}:edge:${edge?.index ?? "unknown"}`,
|
||||
case_id,
|
||||
switch: id,
|
||||
block,
|
||||
case_index: r.index,
|
||||
label: r.label,
|
||||
recovered_target: r.recovered_target,
|
||||
edge,
|
||||
});
|
||||
}
|
||||
}
|
||||
return items;
|
||||
}
|
||||
function selectedCaseCoverage(seeds, summaries) {
|
||||
return seeds.flatMap((id) => {
|
||||
const sw = summaries.get(id),
|
||||
t = sw?.table;
|
||||
return sw
|
||||
? [
|
||||
{
|
||||
id,
|
||||
table_available: t !== null,
|
||||
partial: t?.partial ?? null,
|
||||
case_count: t?.case_count ?? null,
|
||||
relation_count: t?.relation_count ?? null,
|
||||
mapping_count: t?.mapping_count ?? null,
|
||||
known_label_count: t?.known_label_count ?? null,
|
||||
missing_label_count: t?.missing_label_count ?? null,
|
||||
no_label_count: t?.no_label_count ?? null,
|
||||
unmapped_row_count: t?.unmapped_row_count ?? null,
|
||||
},
|
||||
]
|
||||
: [];
|
||||
});
|
||||
}
|
||||
function readCases(index, coverage, input) {
|
||||
const seeds = [...new Set(input.seeds)],
|
||||
labels = input.case_values && new Set(input.case_values),
|
||||
indices =
|
||||
input.case_indices &&
|
||||
[...new Set(input.case_indices)].sort((a, b) => a - b),
|
||||
offset = input.offset ?? 0,
|
||||
limit = input.limit ?? 20,
|
||||
items = [];
|
||||
let total = 0;
|
||||
for (const id of seeds) {
|
||||
const lookup = index.cases.get(id);
|
||||
if (!lookup) continue;
|
||||
const selected = selectCaseRows(lookup, labels, indices),
|
||||
count = selected.prefix.at(-1);
|
||||
const start = Math.max(0, offset - total),
|
||||
end = Math.min(count, offset + limit - total);
|
||||
if (start < end)
|
||||
items.push(...casePage(index.summaries.get(id), selected, start, end));
|
||||
total += count;
|
||||
}
|
||||
return {
|
||||
items,
|
||||
total,
|
||||
next_offset: offset + limit < total ? offset + limit : null,
|
||||
unretained_seeds: seeds.filter((id) => !index.switches.has(id)),
|
||||
coverage,
|
||||
relation_coverage: {
|
||||
scope:
|
||||
"one relation per captured case/out-edge; an unmapped case has one explicit unknown-edge row; not all feasible runtime cases",
|
||||
switches: selectedCaseCoverage(seeds, index.summaries),
|
||||
},
|
||||
};
|
||||
}
|
||||
|
||||
export function createSwitchViews(facts, index) {
|
||||
const coverage = freezeFacts({
|
||||
...facts.coverage,
|
||||
case_identity:
|
||||
"snapshot + switch operation + native address-table row index",
|
||||
relation_identity:
|
||||
"case_id + native out-edge index, or one explicit unknown-edge row",
|
||||
label_authority:
|
||||
"native recovered label and explicit reserved-no-label status; row/edge indexes are not label values",
|
||||
selector_authority:
|
||||
"native current BRANCHIND input; may be a normalized selector rather than a machine jump address",
|
||||
target_authority:
|
||||
"native address-table target and separate recovered out-edge mapping; optimized block origins may differ",
|
||||
default_authority:
|
||||
"native model default out-edge, not necessarily a machine out-of-range guard destination",
|
||||
signedness: "signed_render_hint is only a native rendering hint",
|
||||
runtime_feasibility: "unknown",
|
||||
memory_alias_dependencies: "unknown",
|
||||
});
|
||||
const tables = [...index.summaries.values()].map((s) => s.table);
|
||||
const summary = freezeFacts({
|
||||
count: index.switches.size,
|
||||
table_count: tables.filter(Boolean).length,
|
||||
partial_table_count: tables.filter((t) => t?.partial).length,
|
||||
case_count: tables.reduce((n, t) => n + (t?.case_count ?? 0), 0),
|
||||
relation_count: tables.reduce((n, t) => n + (t?.relation_count ?? 0), 0),
|
||||
coverage,
|
||||
});
|
||||
return {
|
||||
summary,
|
||||
switches: (input) => ({
|
||||
...readNativeIdentities(
|
||||
index.summaries,
|
||||
input,
|
||||
(sw) =>
|
||||
!input.origin_addresses ||
|
||||
input.origin_addresses.includes(sw.origin_address),
|
||||
),
|
||||
coverage,
|
||||
}),
|
||||
cases: (input) => readCases(index, coverage, input),
|
||||
};
|
||||
}
|
||||
@@ -1,351 +0,0 @@
|
||||
import assert from "node:assert/strict";
|
||||
import test from "node:test";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
|
||||
const switchID = "0x1000:1";
|
||||
const edge = (index, target) => ({ id: `0:${index}`, index, target });
|
||||
function tableFixture() {
|
||||
return {
|
||||
partial: true,
|
||||
labelled: true,
|
||||
consumed_mask: "0xff",
|
||||
signed_render_hint: true,
|
||||
default_raw_index: 1,
|
||||
default_edge: edge(1, 2),
|
||||
default_is_folded: true,
|
||||
rows: [
|
||||
{
|
||||
index: 0,
|
||||
label: { status: "value", raw_bits: "0x42" },
|
||||
recovered_target: { space: "ram", offset: "0xabc" },
|
||||
edges: [edge(0, 1)],
|
||||
},
|
||||
{
|
||||
index: 1,
|
||||
label: { status: "value", raw_bits: "0x7" },
|
||||
recovered_target: { space: "ram", offset: "0xdef" },
|
||||
edges: [edge(0, 1)],
|
||||
},
|
||||
{
|
||||
index: 2,
|
||||
label: { status: "no-label", raw_bits: "0xbad1abe1bad1abe1" },
|
||||
recovered_target: { space: "ram", offset: "0xf00" },
|
||||
edges: [edge(1, 2)],
|
||||
},
|
||||
{ index: 3, label: null, recovered_target: null, edges: [] },
|
||||
],
|
||||
};
|
||||
}
|
||||
function graphFixture(selector) {
|
||||
const operation = (address, block, opcode, inputs = []) => ({
|
||||
id: `${address}:1`,
|
||||
address,
|
||||
block,
|
||||
opcode,
|
||||
inputs,
|
||||
output: null,
|
||||
statement: "uninterpreted display",
|
||||
is_dead: false,
|
||||
});
|
||||
return {
|
||||
schema_version: 3,
|
||||
blocks: [
|
||||
{
|
||||
id: 0,
|
||||
start: "0x1000",
|
||||
operations: [switchID],
|
||||
predecessors: [],
|
||||
successors: [edge(0, 1), edge(1, 2)].map((e) => ({
|
||||
...e,
|
||||
polarity: null,
|
||||
condition_value: null,
|
||||
condition_test: null,
|
||||
})),
|
||||
},
|
||||
{
|
||||
id: 1,
|
||||
start: "0x1010",
|
||||
operations: ["0x1010:1"],
|
||||
predecessors: [0],
|
||||
successors: [],
|
||||
},
|
||||
{
|
||||
id: 2,
|
||||
start: "0x1020",
|
||||
operations: ["0x1020:1"],
|
||||
predecessors: [0],
|
||||
successors: [],
|
||||
},
|
||||
],
|
||||
operations: [
|
||||
operation("0x1000", 0, "CPUI_BRANCHIND", [
|
||||
{
|
||||
input_index: 0,
|
||||
value_id: "v1",
|
||||
phi_predecessor: null,
|
||||
phi_predecessor_edge: null,
|
||||
},
|
||||
]),
|
||||
operation("0x1010", 1, "CPUI_RETURN"),
|
||||
operation("0x1020", 2, "CPUI_RETURN"),
|
||||
],
|
||||
values: [{ ...selector }],
|
||||
coverage: {
|
||||
scope: "fixture recovered graph",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
indirect_branch_operations: [switchID],
|
||||
omitted_definition_ids: [],
|
||||
issues: [],
|
||||
},
|
||||
};
|
||||
}
|
||||
function fixture() {
|
||||
const selector = {
|
||||
id: "v1",
|
||||
space: "register",
|
||||
offset: "0x10",
|
||||
size_bytes: 8,
|
||||
is_constant: false,
|
||||
is_input: true,
|
||||
is_annotation: false,
|
||||
definition: null,
|
||||
type_name: "uint64",
|
||||
};
|
||||
return {
|
||||
binary: "/input",
|
||||
count: 1,
|
||||
language: "c",
|
||||
error: null,
|
||||
functions: [
|
||||
{
|
||||
address_hex: "0x1000",
|
||||
code: "retained C",
|
||||
error: null,
|
||||
atomic_facts: graphFixture(selector),
|
||||
atomic_switches: {
|
||||
schema_version: 1,
|
||||
items: [
|
||||
{
|
||||
id: switchID,
|
||||
origin_address: "0x1000",
|
||||
block: 0,
|
||||
selector,
|
||||
table: tableFixture(),
|
||||
},
|
||||
],
|
||||
coverage: {
|
||||
scope: "captured native tables",
|
||||
blocks_examined: 3,
|
||||
operations_examined: 3,
|
||||
machine_semantic_equivalence: "unknown",
|
||||
},
|
||||
},
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
const load = (doc) => parseNativeIR(JSON.stringify(doc), "0x1000");
|
||||
const cases = (model, extra = {}) =>
|
||||
model.edges({ kind: "cases", seeds: [switchID], ...extra });
|
||||
|
||||
test("case-to-edge paging bounds a single case with several mappings and keeps a stable case identity", () => {
|
||||
const doc = fixture();
|
||||
doc.functions[0].atomic_switches.items[0].table.rows[0].edges.push(
|
||||
edge(1, 2),
|
||||
);
|
||||
const model = load(doc),
|
||||
first = cases(model, { case_values: ["0x42"], limit: 1 }),
|
||||
second = cases(model, { case_values: ["0x42"], offset: 1, limit: 1 });
|
||||
assert.equal(model.switches({}).items[0].table.case_count, 4);
|
||||
assert.equal(model.switches({}).items[0].table.relation_count, 5);
|
||||
assert.equal(first.total, 2);
|
||||
assert.equal(first.items.length, 1);
|
||||
assert.equal(first.next_offset, 1);
|
||||
assert.equal(first.items[0].case_id, second.items[0].case_id);
|
||||
assert.notEqual(first.items[0].id, second.items[0].id);
|
||||
assert.equal(first.items[0].edge.index, 0);
|
||||
assert.equal(second.items[0].edge.index, 1);
|
||||
assert.equal(second.next_offset, null);
|
||||
});
|
||||
|
||||
test("a late two-edge page does not materialize thousands of unrequested mappings", () => {
|
||||
const doc = fixture(),
|
||||
fn = doc.functions[0];
|
||||
delete fn.atomic_facts;
|
||||
const table = fn.atomic_switches.items[0].table;
|
||||
table.rows = table.rows.slice(0, 1);
|
||||
table.rows[0].edges = Array.from({ length: 4096 }, (_, i) => edge(i, i + 1));
|
||||
fn.atomic_switches.coverage.blocks_examined = 4097;
|
||||
fn.atomic_switches.coverage.operations_examined = 4097;
|
||||
const model = load(doc);
|
||||
const page = cases(model, { case_values: ["0x42"], offset: 4090, limit: 2 });
|
||||
assert.equal(page.total, 4096);
|
||||
assert.equal(page.next_offset, 4092);
|
||||
assert.deepEqual(
|
||||
page.items.map((r) => r.edge.index),
|
||||
[4090, 4091],
|
||||
);
|
||||
assert.equal(page.items[0].case_id, page.items[1].case_id);
|
||||
assert.ok(Buffer.byteLength(JSON.stringify(page.items)) < 1024);
|
||||
});
|
||||
|
||||
test("case ordinals, labels, original targets and optimized edges remain distinct and paged", () => {
|
||||
const model = load(fixture());
|
||||
assert.equal(model.facets.switches.availability, "available");
|
||||
const sw = model.switches({ fields: ["table"] }).items[0];
|
||||
assert.equal(sw.selector, undefined);
|
||||
assert.equal(sw.table.case_count, 4);
|
||||
assert.equal(sw.table.known_label_count, 2);
|
||||
assert.equal(sw.table.missing_label_count, 1);
|
||||
assert.equal(sw.table.no_label_count, 1);
|
||||
assert.equal(sw.table.unmapped_row_count, 1);
|
||||
assert.equal(sw.table.default_edge.id, "0:1");
|
||||
const page = cases(model, {
|
||||
seeds: [switchID, switchID, "unknown"],
|
||||
case_values: ["0x7", "0x42", "0xbad1abe1bad1abe1"],
|
||||
limit: 1,
|
||||
});
|
||||
assert.equal(page.total, 2);
|
||||
assert.equal(page.next_offset, 1);
|
||||
assert.deepEqual(page.unretained_seeds, ["unknown"]);
|
||||
assert.equal(page.items[0].case_index, 0);
|
||||
assert.equal(page.items[0].label.raw_bits, "0x42");
|
||||
assert.equal(page.items[0].recovered_target.offset, "0xabc");
|
||||
assert.equal(page.items[0].edge.target, 1);
|
||||
assert.throws(() => {
|
||||
page.items[0].edge.target = 2;
|
||||
}, TypeError);
|
||||
assert.equal(cases(model, { case_indices: [0] }).items[0].edge.target, 1);
|
||||
assert.equal(model.block(1).origin_start, "0x1010");
|
||||
const second = cases(model, {
|
||||
case_values: ["0x42", "0x7"],
|
||||
offset: 1,
|
||||
limit: 1,
|
||||
});
|
||||
assert.equal(second.items[0].case_index, 1);
|
||||
assert.equal(second.next_offset, null);
|
||||
assert.deepEqual(
|
||||
cases(model, { case_indices: [2, 0, 1], case_values: ["0x7"] }).items.map(
|
||||
(r) => r.case_index,
|
||||
),
|
||||
[1],
|
||||
);
|
||||
assert.equal(cases(model, { case_values: [] }).total, 0);
|
||||
});
|
||||
test("unlabelled/unmapped and partial or unavailable tables do not become absent runtime cases", () => {
|
||||
const doc = fixture(),
|
||||
model = load(doc);
|
||||
assert.equal(cases(model, { case_values: ["0xbad1abe1bad1abe1"] }).total, 0);
|
||||
const source = cases(model, { case_indices: [2, 3] });
|
||||
assert.equal(source.items[0].label.status, "no-label");
|
||||
assert.equal(source.items[1].label, null);
|
||||
assert.deepEqual(source.items[1].edge, null);
|
||||
assert.equal(source.relation_coverage.switches[0].partial, true);
|
||||
assert.equal(source.coverage.runtime_feasibility, "unknown");
|
||||
doc.functions[0].atomic_switches.items[0].table = null;
|
||||
const unknown = cases(load(doc));
|
||||
assert.equal(unknown.relation_coverage.switches[0].table_available, false);
|
||||
assert.equal(unknown.relation_coverage.switches[0].case_count, null);
|
||||
});
|
||||
test("switch rows survive graph failure and missing old-engine capture remains unavailable", () => {
|
||||
const doc = fixture();
|
||||
delete doc.functions[0].atomic_facts;
|
||||
doc.functions[0].atomic_facts_error = "graph capacity";
|
||||
const model = load(doc);
|
||||
assert.equal(model.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(model.switches({}).items[0].selector.size_bytes, 8);
|
||||
assert.equal(cases(model).total, 4);
|
||||
delete doc.functions[0].atomic_switches;
|
||||
const legacy = load(doc);
|
||||
assert.equal(legacy.facets.switches.availability, "unavailable");
|
||||
assert.throws(() => cases(legacy));
|
||||
assert.equal(legacy.code, "retained C");
|
||||
doc.functions[0].atomic_switches = {
|
||||
schema_version: 1,
|
||||
items: [],
|
||||
coverage: {
|
||||
scope: "all recovered switches",
|
||||
blocks_examined: 3,
|
||||
operations_examined: 3,
|
||||
machine_semantic_equivalence: "unknown",
|
||||
},
|
||||
};
|
||||
const empty = load(doc);
|
||||
assert.equal(empty.facets.switches.availability, "available");
|
||||
assert.equal(empty.facets.switches.count, 0);
|
||||
});
|
||||
test("contradictory switch ownership or values fail only that independent facet", () => {
|
||||
for (const change of [
|
||||
(sw) => (sw.table.rows[0].edges[0].target = 2),
|
||||
(sw) => (sw.table.rows[0].index = 1),
|
||||
(sw) => (sw.selector.size_bytes = 4),
|
||||
(sw) => (sw.table.default_raw_index = 0),
|
||||
(sw) => sw.table.rows[0].edges.push(edge(0, 1)),
|
||||
]) {
|
||||
const doc = fixture();
|
||||
change(doc.functions[0].atomic_switches.items[0]);
|
||||
const model = load(doc);
|
||||
assert.equal(model.facets.switches.availability, "unavailable");
|
||||
assert.equal(model.facets.recovered_ir.availability, "available");
|
||||
assert.equal(model.code, "retained C");
|
||||
}
|
||||
});
|
||||
test("partial facets cannot disagree about a shared value identity when the graph is unavailable", () => {
|
||||
const doc = fixture(),
|
||||
fn = doc.functions[0];
|
||||
delete fn.atomic_facts;
|
||||
fn.atomic_calls = {
|
||||
schema_version: 1,
|
||||
items: [
|
||||
{
|
||||
id: "0x1100:0",
|
||||
origin_address: "0x1100",
|
||||
block: 0,
|
||||
opcode: "CPUI_CALL",
|
||||
is_dead: false,
|
||||
target_value_id: null,
|
||||
spec_available: false,
|
||||
target: null,
|
||||
target_name: null,
|
||||
arguments: [
|
||||
{
|
||||
input_index: 1,
|
||||
value_id: "v1",
|
||||
storage: null,
|
||||
storage_authority: null,
|
||||
},
|
||||
],
|
||||
output: null,
|
||||
prototype: null,
|
||||
stack_offset: null,
|
||||
},
|
||||
],
|
||||
values: [{ ...fn.atomic_switches.items[0].selector }],
|
||||
coverage: {
|
||||
scope: "all recovered calls",
|
||||
blocks_examined: 3,
|
||||
operations_examined: 3,
|
||||
value_scope: "call inputs and outputs",
|
||||
machine_call_equivalence: "unknown",
|
||||
},
|
||||
};
|
||||
assert.equal(load(doc).facets.switches.availability, "available");
|
||||
fn.atomic_switches.items[0].selector.size_bytes = 4;
|
||||
const invalid = load(doc);
|
||||
assert.equal(invalid.facets.switches.availability, "unavailable");
|
||||
assert.equal(invalid.facets.calls.availability, "available");
|
||||
assert.equal(invalid.calls({}).total, 1);
|
||||
});
|
||||
test("switch capacity rejection precedes row validation and preserves the valid graph", () => {
|
||||
for (const oversized of [
|
||||
(table) => (table.rows = Array(100001).fill({})),
|
||||
(table) => (table.rows[0].edges = Array(100001).fill({})),
|
||||
]) {
|
||||
const doc = fixture();
|
||||
oversized(doc.functions[0].atomic_switches.items[0].table);
|
||||
const model = load(doc);
|
||||
assert.equal(model.facets.switches.availability, "unavailable");
|
||||
assert.match(model.facets.switches.error.message, /capacity/);
|
||||
assert.equal(model.facets.recovered_ir.availability, "available");
|
||||
}
|
||||
});
|
||||
@@ -1,23 +0,0 @@
|
||||
const read = async (promise) => {
|
||||
const result = await promise;
|
||||
if (!result.ok) throw new Error(JSON.stringify(result));
|
||||
return result.data;
|
||||
};
|
||||
const found = await read(rea.strings({ query: "incorrect header check", limit: 1 }));
|
||||
const literal = found.items[0].address;
|
||||
const refs = await read(rea.references({ addresses: [literal] }));
|
||||
const entry = refs.results[0].items[0].fromFunction.address;
|
||||
const snapshot = await read(rea.snapshot({ entry }));
|
||||
const s = snapshot.snapshot;
|
||||
const seed = await read(rea.operations({ snapshot: s, constant: literal, fields: ["kind", "statement", "inputs", "output"] }));
|
||||
const uses = await read(rea.relations({ snapshot: s, kind: "uses", seeds: seed.items.map((row) => row.id) }));
|
||||
const consumers = await read(rea.operations({ snapshot: s, ids: uses.items.map((row) => row.use) }));
|
||||
const blockIds = [...new Set(consumers.items.map((row) => row.block))];
|
||||
const blocks = await read(rea.blocks({ snapshot: s, ids: blockIds }));
|
||||
const controls = await read(rea.relations({ snapshot: s, kind: "controllers", seeds: blockIds, limit: 128 }));
|
||||
const guards = await read(rea.operations({ snapshot: s, block_ids: [...new Set(controls.items.map((row) => row.controller))], kind: "branch", fields: ["kind", "statement", "inputs"] }));
|
||||
const phi = await read(rea.operations({ snapshot: s, kind: "phi", fields: ["inputs", "output"], limit: 2 }));
|
||||
const valueIds = [...new Set(phi.items.flatMap((op) => op.inputs.map((input) => input.value_id)).filter(Boolean))];
|
||||
const values = await read(rea.values({ snapshot: s, ids: valueIds }));
|
||||
const slice = await read(rea.slice({ snapshot: s, seeds: guards.items.map((row) => row.id), max_operations: 64 }));
|
||||
return { literal, entry, source: snapshot.source, coverage: snapshot.coverage, blocks, controls, guards, phi, values, slice: { coverage: slice.coverage, value_frontier: slice.value_frontier, operations: slice.operations.map(({ id, statement }) => ({ id, statement })) } };
|
||||
@@ -1,44 +0,0 @@
|
||||
# Wide native operand retrieval oracle
|
||||
|
||||
Pinned before any model launch on this question; not supplied to measured models.
|
||||
This is a seeded relationship-retrieval task in a complex function, not an
|
||||
unseeded behavior reconstruction, source-level explanation or runtime proof.
|
||||
|
||||
Target: Debian SQLite 3.40.1-2+deb12u2 amd64,
|
||||
`/usr/lib/x86_64-linux-gnu/libsqlite3.so.0` -> `.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Engine: preserved `a0685486518e09d3c65f756325a7e9fed8df123a2a988aca2972a72335ce8edc`.
|
||||
|
||||
From the retained public-budget native projection: operation `0xea998:14649` is
|
||||
`CPUI_MULTIEQUAL`, block 1203, output `v30000f159`, 156 native input slots.
|
||||
It is not dead. Input relationships must retain their own identities even when
|
||||
four selected slots refer to one value:
|
||||
|
||||
| Input index | Value | Predecessor | Predecessor edge |
|
||||
| ----------- | ---------- | ----------: | ---------------- |
|
||||
| 0 | v300034de6 | 7 | 7:1 |
|
||||
| 1 | v300034de6 | 8 | 8:72 |
|
||||
| 77 | v300034de6 | 633 | 633:0 |
|
||||
| 105 | v10000cb23 | 738 | 738:0 |
|
||||
| 155 | v300034de6 | 668 | 668:0 |
|
||||
|
||||
Distinct values and retained immediate definitions:
|
||||
|
||||
| Value | Space | Offset | Width | Definition | Opcode/output |
|
||||
| ---------- | ------ | ------------------ | ----: | ------------- | ---------------------------- |
|
||||
| v300034de6 | stack | 0xfffffffffffffec0 | 8 | 0xea7d8:146d5 | CPUI_MULTIEQUAL / v300034de6 |
|
||||
| v10000cb23 | unique | 0x23e00 | 8 | 0xebd6c:1a344 | CPUI_CAST / v10000cb23 |
|
||||
|
||||
Both values are nonconstant, nonannotation and not native function inputs.
|
||||
Their offsets are storage offsets, not computed values or machine destinations.
|
||||
Slot 156 is absent for this retained 156-slot operation. `0xffff:0` is not
|
||||
retained; its existence outside the projection remains unknown, not an empty
|
||||
operation. Selected operands plus definition identities suffice; no complete
|
||||
input array, C report, exhaustive graph or transitive slice is required.
|
||||
|
||||
Public projection: 4,096 operations, 7,469 values and 326 definition-frontier
|
||||
entries. Complete recovered CFG is unavailable; uses concern the retained set,
|
||||
aliases and runtime feasibility remain unknown. Selected immediate definitions
|
||||
are retained, without closing the global frontier. Evidence references and
|
||||
target digest must be real. These producer-gold identities were also compared
|
||||
to the earlier full diagnostic capture; neither is independent machine/IR proof.
|
||||
@@ -1,52 +0,0 @@
|
||||
# Sequential and overlapping public requests
|
||||
|
||||
This follows the unexplained 107.804-second source capture in the
|
||||
[guided model pair](savepoint-control-study.md). The same preserved `677b9795...`
|
||||
engine, target digest, VM entry and three instruction windows are used through
|
||||
real stdio MCP. Each case starts a fresh lease/private native cache; cases run
|
||||
serially in code/sequential, direct/overlap, direct/sequential, code/overlap order.
|
||||
No model is involved and no token savings are inferred.
|
||||
|
||||
`verify-overlap.mjs` first requests one full VM capture and the windows, either
|
||||
sequentially or with the second request starting 100 ms after the first. The
|
||||
client records starts/finishes to prove request overlap; it does not assume
|
||||
provider concurrency. `AtomicSession.call` serializes provider admission for
|
||||
both adapters. Thus an instruction request can wait behind the snapshot. Its
|
||||
API wall time is not its native command execution time.
|
||||
|
||||
Every full recovered graph and ordinary function field matches the complete
|
||||
golden. Selected instruction items/coverage agree across all four cases. Both
|
||||
snapshot and instruction reads repeat under the same retained Evidence and
|
||||
handle. Target/frame/decoding selection and all limits are preserved.
|
||||
|
||||
| Case | Fresh task seconds | Fresh family CPU s | VM capture s | Warm task ms | Warm family CPU s |
|
||||
| ------------------ | -----------------: | -----------------: | -----------: | -----------: | ----------------: |
|
||||
| Code, sequential | 39.488 | 39.432 | 37.112 | 43.737 | 0.044 |
|
||||
| Direct, overlap | 38.931 | 38.865 | 36.635 | 6.561 | 0.007 |
|
||||
| Direct, sequential | 39.171 | 39.098 | 36.987 | 7.759 | 0.008 |
|
||||
| Code, overlap | 40.616 | 40.564 | 38.262 | 32.288 | 0.033 |
|
||||
|
||||
CPU is a delta over each complete phase in the owned cgroup; no overlapping
|
||||
per-request CPU attribution is attempted. Kernel peaks are cumulative and
|
||||
include charged file cache, not RSS; the warm peak is only an upper bound.
|
||||
Complete family totals/peaks and cleanup counters are in the resource receipts
|
||||
under `.cache/atomic-codemode/overlap-v1-*`. One CPU, 4 GiB hard / 3 GiB early-stop,
|
||||
192-task bounds apply. All four cases have verified cleanup, no abort and OOM 0.
|
||||
|
||||
| Case | Whole family wall s | Whole family CPU s | Kernel peak MiB |
|
||||
| ------------------ | ------------------: | -----------------: | --------------: |
|
||||
| Code, sequential | 41.299 | 41.026 | 623.1 |
|
||||
| Direct, overlap | 40.695 | 40.427 | 616.2 |
|
||||
| Direct, sequential | 40.893 | 40.660 | 617.6 |
|
||||
| Code, overlap | 42.506 | 42.239 | 630.0 |
|
||||
|
||||
The 108-second capture does not reproduce in this matched sample. Request
|
||||
queueing is confirmed by the owner and explains waiting outside native source
|
||||
times; it does not explain the original long capture itself. There is no basis
|
||||
to change provider concurrency or claim a native speed fix. Host contention,
|
||||
OS caches and unrecorded native phases remain uncontrolled/unmeasured. This is
|
||||
a four-case diagnostic, not a distribution or proof that the anomaly is fixed.
|
||||
|
||||
Both repeated tasks are small; Code Mode's additional guest setup is visible
|
||||
here. This is not evidence against composition over a larger dependency graph,
|
||||
nor a comparison with a client that lacks its own JavaScript executor.
|
||||
-80
@@ -1,80 +0,0 @@
|
||||
{
|
||||
"name": "rea-atomic-codemode-experiment",
|
||||
"lockfileVersion": 3,
|
||||
"requires": true,
|
||||
"packages": {
|
||||
"": {
|
||||
"name": "rea-atomic-codemode-experiment",
|
||||
"dependencies": {
|
||||
"quickjs-emscripten": "0.32.0"
|
||||
}
|
||||
},
|
||||
"node_modules/@jitl/quickjs-ffi-types": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/@jitl/quickjs-ffi-types/-/quickjs-ffi-types-0.32.0.tgz",
|
||||
"integrity": "sha512-v9T+GQpmk43VDJ7d72sf0Nexhk+ArvtUihW27dy7lqAl0zBObFKtSBBIm5RBjwIhE8VwsPPm9PNuvPvNqLWUEg==",
|
||||
"license": "MIT"
|
||||
},
|
||||
"node_modules/@jitl/quickjs-wasmfile-debug-asyncify": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/@jitl/quickjs-wasmfile-debug-asyncify/-/quickjs-wasmfile-debug-asyncify-0.32.0.tgz",
|
||||
"integrity": "sha512-EX8zbXwGqCgAE764M+qvkHtyXDi/FUoMBea0JnES7vCM3P7a2+EOZOjGv85wtZ2sJhI1oJ+nekmqpOODFDY+hw==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-ffi-types": "0.32.0"
|
||||
}
|
||||
},
|
||||
"node_modules/@jitl/quickjs-wasmfile-debug-sync": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/@jitl/quickjs-wasmfile-debug-sync/-/quickjs-wasmfile-debug-sync-0.32.0.tgz",
|
||||
"integrity": "sha512-LeYWrPGC1uNCTBWvibo3ZLJj0CSVNYUXvJpXMCmuQ5Sap2cCACc3uvGvYV4homHHBAzfw5akoTqMMS4YFRtw+Q==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-ffi-types": "0.32.0"
|
||||
}
|
||||
},
|
||||
"node_modules/@jitl/quickjs-wasmfile-release-asyncify": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/@jitl/quickjs-wasmfile-release-asyncify/-/quickjs-wasmfile-release-asyncify-0.32.0.tgz",
|
||||
"integrity": "sha512-3oSwPfja12ICz4aIblB58cuY8JlEq5Txt8Cut4VLo+LH47QN+mzCnSgnbB03hWzg1LBcc+VyyI9UOag7a1NF+Q==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-ffi-types": "0.32.0"
|
||||
}
|
||||
},
|
||||
"node_modules/@jitl/quickjs-wasmfile-release-sync": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/@jitl/quickjs-wasmfile-release-sync/-/quickjs-wasmfile-release-sync-0.32.0.tgz",
|
||||
"integrity": "sha512-BKNDI/TPBfGlLNGYpLrhcDGXmIk4xHm4MRAisOBnOzpXVn9HZWsfmMAc9WMBrAHjvvds6HOikKeaOBKdPdpVrg==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-ffi-types": "0.32.0"
|
||||
}
|
||||
},
|
||||
"node_modules/quickjs-emscripten": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/quickjs-emscripten/-/quickjs-emscripten-0.32.0.tgz",
|
||||
"integrity": "sha512-So0Sqw869y/S2oE3Nuc0uT3Dhqgvsj8FSrwBdsuTosVsG8ME5/OcudU1GxsrIFdFABgy17GHnTVO9TYV/bLQcA==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-wasmfile-debug-asyncify": "0.32.0",
|
||||
"@jitl/quickjs-wasmfile-debug-sync": "0.32.0",
|
||||
"@jitl/quickjs-wasmfile-release-asyncify": "0.32.0",
|
||||
"@jitl/quickjs-wasmfile-release-sync": "0.32.0",
|
||||
"quickjs-emscripten-core": "0.32.0"
|
||||
},
|
||||
"engines": {
|
||||
"node": ">=16.0.0"
|
||||
}
|
||||
},
|
||||
"node_modules/quickjs-emscripten-core": {
|
||||
"version": "0.32.0",
|
||||
"resolved": "https://registry.npmjs.org/quickjs-emscripten-core/-/quickjs-emscripten-core-0.32.0.tgz",
|
||||
"integrity": "sha512-QFnPfjFey8EqknSrSxe1hZrf1/8z7/6s1QzGOmKo6++02r7QRRX7ZoyNaZh7JuVjWsVW87KnQrbZqnHkOAzUyg==",
|
||||
"license": "MIT",
|
||||
"dependencies": {
|
||||
"@jitl/quickjs-ffi-types": "0.32.0"
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,12 +0,0 @@
|
||||
{
|
||||
"dependencies": {
|
||||
"quickjs-emscripten": "0.32.0"
|
||||
},
|
||||
"name": "rea-atomic-codemode-experiment",
|
||||
"private": true,
|
||||
"type": "module",
|
||||
"packageManager": "npm@11.16.0",
|
||||
"scripts": {
|
||||
"test": "node --test *.test.mjs"
|
||||
}
|
||||
}
|
||||
@@ -1,221 +0,0 @@
|
||||
// Diagnostic native timings, using REA's target, environment and process owners.
|
||||
import assert from "node:assert/strict";
|
||||
import { createHash } from "node:crypto";
|
||||
import { readFile, writeFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
import { parseArgs } from "node:util";
|
||||
import { parseBinaryTarget } from "../../dist/application/BinaryTargetResolver.js";
|
||||
import { inspectKunaInstallation } from "../../dist/kuna/KunaInstallation.js";
|
||||
import { KunaTargetCommands } from "../../dist/kuna/KunaTargetCommands.js";
|
||||
import { PrivateRuntimeRoot } from "../../dist/process/PrivateRuntimeRoot.js";
|
||||
import { spawnOwnedProviderProcess } from "../../dist/process/ProviderProcess.js";
|
||||
import { requireResult } from "./fact-error.mjs";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
import { NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const { values: options, positionals } = parseArgs({
|
||||
allowPositionals: true,
|
||||
options: {
|
||||
"without-timing": { type: "boolean", default: false },
|
||||
"output-bytes": { type: "string", default: String(NATIVE_OUTPUT_BYTES) },
|
||||
},
|
||||
});
|
||||
const [engine, binary, entry, output, goldenPath] = positionals;
|
||||
const timingEnabled = !options["without-timing"];
|
||||
assert.ok(engine && binary && entry && output);
|
||||
const outputBytes = Number(options["output-bytes"]);
|
||||
assert.ok(
|
||||
Number.isSafeInteger(outputBytes) &&
|
||||
outputBytes > 0 &&
|
||||
outputBytes <= NATIVE_OUTPUT_BYTES,
|
||||
"Choose a positive native output budget within the supervisor's 8 MiB capture capacity",
|
||||
);
|
||||
const sha = (s) => createHash("sha256").update(s).digest("hex");
|
||||
const expectedPhases = [
|
||||
"decompile_drive",
|
||||
"pre_render_metadata",
|
||||
"render_c",
|
||||
"metadata_extract",
|
||||
"membership",
|
||||
"calls",
|
||||
"full_graph",
|
||||
"projection",
|
||||
"metadata_tail",
|
||||
];
|
||||
const verifier = createVerifierRun(),
|
||||
runs = [],
|
||||
failures = [];
|
||||
const root = await PrivateRuntimeRoot.create({ prefix: "rea-atomic-profile-" });
|
||||
let commands, installation, target;
|
||||
try {
|
||||
target = requireResult(await parseBinaryTarget(binary));
|
||||
installation = requireResult(
|
||||
await inspectKunaInstallation({
|
||||
...process.env,
|
||||
REA_KUNA_COMMAND: engine,
|
||||
REA_KUNA_CACHE_DIR: join(root.path, "engine-cache"),
|
||||
}),
|
||||
);
|
||||
commands = new KunaTargetCommands(target, {
|
||||
signal: new AbortController().signal,
|
||||
launcher: (options) =>
|
||||
spawnOwnedProviderProcess({
|
||||
...options,
|
||||
env: {
|
||||
...options.env,
|
||||
...(timingEnabled ? { KUNA_EXPERIMENT_ATOMIC_TIMING: "1" } : {}),
|
||||
},
|
||||
}),
|
||||
});
|
||||
let firstFunction;
|
||||
for (const phase of [
|
||||
"fresh-private-native-cache",
|
||||
"same-private-native-cache-new-process",
|
||||
]) {
|
||||
const before = performance.now();
|
||||
const captured = requireResult(
|
||||
await commands.run({
|
||||
installation,
|
||||
operation: "experiment:atomic-phase-profile",
|
||||
arguments: (input) => [
|
||||
"decompile",
|
||||
input,
|
||||
entry,
|
||||
"--addr",
|
||||
"--json",
|
||||
"--atomic-facts",
|
||||
"--atomic-output-bytes",
|
||||
String(outputBytes),
|
||||
"--max-fn-memory",
|
||||
"1024",
|
||||
],
|
||||
outputBytes: NATIVE_OUTPUT_BYTES,
|
||||
timeoutMs: 120000,
|
||||
signal: new AbortController().signal,
|
||||
}),
|
||||
);
|
||||
const wallMs = performance.now() - before;
|
||||
const stdout = captured.stdout.text,
|
||||
stderr = captured.stderr.text;
|
||||
const stdoutPath = `${output}.run-${runs.length}.stdout.json`,
|
||||
stderrPath = `${output}.run-${runs.length}.stderr.jsonl`;
|
||||
await writeFile(stdoutPath, stdout, { flag: "wx" });
|
||||
await writeFile(stderrPath, stderr, { flag: "wx" });
|
||||
const record = {
|
||||
phase,
|
||||
wall_ms: wallMs,
|
||||
status: "unverified",
|
||||
source: {
|
||||
stdout_path: stdoutPath,
|
||||
stdout_sha256: sha(stdout),
|
||||
stdout_bytes: Buffer.byteLength(stdout),
|
||||
stderr_path: stderrPath,
|
||||
stderr_sha256: sha(stderr),
|
||||
},
|
||||
};
|
||||
runs.push(record);
|
||||
assert.notEqual(captured.diagnosticTruncated, true);
|
||||
const timings = stderr
|
||||
.split("\n")
|
||||
.filter((line) => line.startsWith('{"kind":"rea_atomic_phase",'))
|
||||
.map((line) => JSON.parse(line));
|
||||
assert.deepEqual(
|
||||
timings.map((row) => row.phase),
|
||||
timingEnabled ? expectedPhases : [],
|
||||
);
|
||||
assert.ok(
|
||||
timings.every(
|
||||
(row) =>
|
||||
row.address === entry &&
|
||||
Number.isSafeInteger(row.wall_ns) &&
|
||||
row.wall_ns >= 0,
|
||||
),
|
||||
);
|
||||
const model = parseNativeIR(stdout, entry, captured.input);
|
||||
const functionRecord = JSON.parse(stdout).functions[0];
|
||||
if (firstFunction) assert.deepEqual(functionRecord, firstFunction);
|
||||
firstFunction ??= functionRecord;
|
||||
if (goldenPath)
|
||||
assert.deepEqual(
|
||||
functionRecord,
|
||||
JSON.parse(await readFile(goldenPath, "utf8")).functions[0],
|
||||
);
|
||||
const accountedMs = timingEnabled
|
||||
? timings.reduce((n, row) => n + row.wall_ns / 1e6, 0)
|
||||
: null;
|
||||
if (accountedMs !== null) assert.ok(accountedMs <= wallMs);
|
||||
Object.assign(record, {
|
||||
status: "verified",
|
||||
accounted_wall_ms: accountedMs,
|
||||
unprofiled_wall_ms: accountedMs === null ? null : wallMs - accountedMs,
|
||||
timings,
|
||||
code_bytes: Buffer.byteLength(functionRecord.code),
|
||||
facets: model.facets,
|
||||
function_record_matches_prior_run: runs.length > 1,
|
||||
function_record_matches_golden: goldenPath ? true : null,
|
||||
});
|
||||
}
|
||||
} catch (error) {
|
||||
failures.push({
|
||||
code: error.code ?? "verification_failed",
|
||||
message: error.message,
|
||||
details: error.details ?? {},
|
||||
stack: error.stack,
|
||||
});
|
||||
} finally {
|
||||
let closed = true;
|
||||
if (commands) {
|
||||
const result = await commands.close();
|
||||
closed = result.ok;
|
||||
if (!closed)
|
||||
failures.push({
|
||||
code: result.error.code,
|
||||
message: result.error.message,
|
||||
details: result.error.details,
|
||||
});
|
||||
}
|
||||
if (closed) {
|
||||
try {
|
||||
await root.close();
|
||||
} catch (error) {
|
||||
failures.push({
|
||||
code: "cleanup_incomplete",
|
||||
message: error.message,
|
||||
resources: [root.path],
|
||||
});
|
||||
}
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify(
|
||||
{
|
||||
engine: installation && {
|
||||
path: installation.executable,
|
||||
sha256: installation.sha256,
|
||||
},
|
||||
target: target && { path: target.path, sha256: target.sha256 },
|
||||
entry,
|
||||
timing_enabled: timingEnabled,
|
||||
output_budget_bytes: outputBytes,
|
||||
supervisor_capture_budget_bytes: NATIVE_OUTPUT_BYTES,
|
||||
runs,
|
||||
failures,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Diagnostic native processes through production owners with an explicit timing-only launcher environment. Phase walls only, not phase CPU. Whole-run CPU/peak memory are separate resource-run receipts; OS cache uncontrolled. Not a model or public MCP measurement.",
|
||||
},
|
||||
null,
|
||||
2,
|
||||
),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(JSON.stringify({ output, runs: runs.length, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,136 +0,0 @@
|
||||
# Selected definitions after whole-graph capacity failure
|
||||
|
||||
Historical probe at REA experiment revision `4d7fccbce`. Use that revision's
|
||||
patches to reproduce its preserved diagnostic engine and results below. The
|
||||
subsequent callable projection is documented in [native-engine.md](native-engine.md);
|
||||
its real-model question/oracle are [pinned separately](projection-oracle.md).
|
||||
|
||||
This is a diagnostic feasibility probe following the completed boundary model
|
||||
pairs, not a new public provider capability. The analyst question is whether
|
||||
selected call-argument definitions can remain readable without materializing an
|
||||
entire recovered graph in the public transport. The alternatives are a smaller
|
||||
whole-graph encoding and a bounded selection of operations/values. Neither
|
||||
fewer bytes nor an offline graph traversal establishes model-token or time gains.
|
||||
|
||||
## Actual graph and capacity boundaries
|
||||
|
||||
The artifact remains Debian SQLite `3.40.1-2+deb12u2`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
The selected implementation is `sqlite3VdbeExec`, entry `0xea6d0`.
|
||||
The disposable diagnostic patch temporarily admits 32,000 operations, 72,000
|
||||
values, 128,000 inputs and 16 MiB text. It is applied after
|
||||
`kuna-native-ir.patch`, against the same pinned Kuna upstream revision. Only
|
||||
this diagnostic capture uses a 32 MiB encoded/stdout budget. The committed
|
||||
public patch, preserved public engine and 8 MiB public budget are unchanged.
|
||||
|
||||
| Observed quantity | Value |
|
||||
| ----------------- | ----: |
|
||||
| Recovered operations | 21,349 |
|
||||
| Recovered values | 35,755 |
|
||||
| Recovered blocks | 1,229 |
|
||||
| Operand inputs | 51,951 |
|
||||
| CFG successor edges | 2,061 |
|
||||
| Operation statement UTF-8 bytes | 2,255,616 |
|
||||
| Value type-name UTF-8 bytes | 90,013 |
|
||||
| Compact ordinary graph JSON bytes | 17,008,223 |
|
||||
| Entire native document bytes | 17,801,972 |
|
||||
|
||||
Raising only the current 12,000-operation bound would not suffice: inputs also
|
||||
exceed 50,000, display text exceeds 2 MiB, and the entire document exceeds 8 MiB.
|
||||
Values happen to remain below 36,000, with little headroom. Native coverage
|
||||
reports no producer issues or omitted defining-operation IDs. This is the
|
||||
retained recovered graph, not proof of complete/correct machine or runtime
|
||||
semantics. The ordinary function record, including C and mappings, and the
|
||||
independent call facet deeply equal the preserved public-engine capture.
|
||||
|
||||
`measure-native-projection.py` constructs a lossless tuple encoding of every
|
||||
graph field and verifies decoding equals the original graph. Its graph occupies
|
||||
8,225,047 bytes; estimated whole-document size is 9,018,796 bytes after retaining
|
||||
C/calls/metadata. Thus this simple encoding still does not fit the unchanged
|
||||
8,388,608-byte public budget. This is an exact synthetic serialization size and
|
||||
roundtrip check, not an implemented native codec or measured codec speed.
|
||||
|
||||
## Focused definition facts
|
||||
|
||||
Using the two existing call identities at `0xf115c` and `0xf238b` as roots, the
|
||||
script follows native operand values to their defining operations. It retains
|
||||
exact IDs and complete selected operation/value records. A 512-operation budget
|
||||
is not reached in these depth samples; unresolved definitions remain explicit.
|
||||
|
||||
| Requested backward depth | Operations | Values | Encoded payload bytes | Unexpanded definition frontier |
|
||||
| ------------------------ | ---------: | -----: | --------------------: | -----------------------------: |
|
||||
| 1 | 5 | 12 | 4,335 | 6 |
|
||||
| 2 | 11 | 24 | 8,589 | 3 |
|
||||
| 3 | 14 | 29 | 10,447 | 1 |
|
||||
| 4 | 15 | 32 | 11,457 | 1 |
|
||||
| 5 | 16 | 33 | 11,934 | 1 |
|
||||
|
||||
The same experiment over all 416 recovered call roots gives:
|
||||
|
||||
| Requested backward depth | Operations | Values | Encoded payload bytes |
|
||||
| ------------------------ | ---------: | -----: | --------------------: |
|
||||
| 1 | 965 | 2,269 | 830,233 |
|
||||
| 2 | 1,494 | 3,076 | 1,188,208 |
|
||||
| 3 | 2,007 | 3,920 | 1,541,735 |
|
||||
| 4 | 2,401 | 4,670 | 1,849,015 |
|
||||
|
||||
The 4,096-operation probe budget is not reached in these samples. Depth three
|
||||
still has 394 unexpanded definitions. These payload sizes cover selected
|
||||
operations/values alone, before source/provenance/transport framing. Depths are
|
||||
probe parameters, not proposed product limits. This is backward SSA dependence,
|
||||
without CFG, control dependence, complete use lists, memory aliases or feasible
|
||||
runtime paths. No model receives these offline projections in the completed
|
||||
trials. They do not become accessible through an existing snapshot by this probe.
|
||||
|
||||
The signal favors trying a bounded call-rooted operation/value facet when the
|
||||
full graph is unavailable. The native producer already has the live function at
|
||||
the capture boundary, so this could reuse that analysis rather than reconstruct
|
||||
definitions from C or force repeated decompilation. A focused initial capture
|
||||
is another option; later requests would need a new analysis or a separately
|
||||
owned live native model. That lifecycle/time tradeoff remains unresolved.
|
||||
|
||||
Any prototype must share native identity/value/operand interpretation, retain
|
||||
frontier reasons and source Evidence, expose partial definition/use scope, and
|
||||
keep CFG-dependent reads unavailable. A native block/origin ID does not establish
|
||||
a machine instruction extent or branch target. Both direct and Code Mode must
|
||||
receive the same facts through the shared API before another model comparison.
|
||||
|
||||
## Reproduction and measured scope
|
||||
|
||||
Apply `kuna-native-ir.patch`, then `kuna-graph-scale-diagnostic.patch`, to pinned
|
||||
upstream `fe01d6294eb0d1053ce343c8943646df88e143fc`. Reuse its stock SLA assets
|
||||
and the one-job embedded-spec build profile in [native-engine.md](native-engine.md).
|
||||
The preserved diagnostic executable has SHA-256
|
||||
`6d0d1cef9de13180e19af77a6f62075c417178831740bccc7174bbac0af764cc`.
|
||||
It is copied outside Cargo outputs, separately from the public `e7d3f0cc...` engine.
|
||||
The initial build launch failed before Cargo because its PATH omitted the existing
|
||||
Cargo installation; that setup failure remains recorded, followed by a corrected
|
||||
build. No toolchain was installed or upgraded.
|
||||
|
||||
The capture reuses `KunaTargetCommands.run`, a fresh private installation
|
||||
`configuration.cacheRoot`, owned target staging and normal supervision. Its argv
|
||||
is `decompile INPUT 0xea6d0 --addr --json --atomic-facts
|
||||
--atomic-output-bytes 33554432 --max-fn-memory 1024`, with matching 32 MiB
|
||||
supervised stdout capacity. It does not execute the target. The temporary native
|
||||
source changes are restored after capture; diagnostic patch reverse/forward and
|
||||
public patch reverse checks pass.
|
||||
|
||||
The diagnostic build takes 316.764 family wall / 316.136 CPU seconds, 2,113.5 MiB
|
||||
kernel family peak. Native capture takes 37.078 / 36.869 seconds, 551.4 MiB.
|
||||
Offline sizing/roundtrip takes 1.428 / 1.242 seconds, 170.0 MiB. All use one CPU,
|
||||
4 GiB hard memory, 3 GiB early stop and verify cleanup/OOM 0; peaks include charged
|
||||
file cache. OS caches are uncontrolled. There is no equivalent completed-task
|
||||
speed comparison, public-provider success claim or model-token measurement here.
|
||||
|
||||
Run the bounded sizing script against the retained diagnostic native JSON:
|
||||
|
||||
```sh
|
||||
python3 experiments/analysis-efficiency/resource-run.py --observe-payload-exit -- \
|
||||
python3 experiments/atomic-codemode/measure-native-projection.py \
|
||||
/absolute/native.json /absolute/new-sizing-report.json
|
||||
```
|
||||
|
||||
Raw receipts are `graph-scale-diagnostic-engine.json`,
|
||||
`graph-scale-diagnostic-receipt.json`, `graph-scale-repro-sizing.json` and their
|
||||
resource logs under `.cache/atomic-codemode/`. Binaries and the 17.8 MB native
|
||||
capture remain local; they are not committed.
|
||||
@@ -1,66 +0,0 @@
|
||||
# Phi alternatives through bounded native facts
|
||||
|
||||
Fix `sqlite-projection-question.txt` and this oracle before either model launch.
|
||||
Both lanes receive the same fourteen methods, native engine, model/effort request,
|
||||
common guidance and resource bounds. This is a seeded follow-up investigation,
|
||||
not discovery from an unknown entry. The task differs from counting boundary
|
||||
calls, so its costs are not a before/after speedup comparison with that task.
|
||||
|
||||
Artifact: Debian SQLite `3.40.1-2+deb12u2`, `libsqlite3.so.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Implementation entry is `0xea6d0`, with a distinct same-name PLT entry.
|
||||
|
||||
Real public projection verification and comparison with the earlier full native
|
||||
diagnostic graph establish these retained identities. The complete graph remains
|
||||
unavailable; bounded operations/values are available. It is not necessary for the
|
||||
answer to enumerate every global budget/counter or all unselected frontier IDs.
|
||||
|
||||
| Call origin / identity | Recovered RSI value | Immediate PHI definition |
|
||||
| ---------------------- | ------------------- | ------------------------ |
|
||||
| `0xf115c` / `0xf115c:6a84` | `v1000340ac` | `0xf1152:cf0f` |
|
||||
| `0xf238b` / `0xf238b:7130` | `v100033f25` | `0xf238b:ceb7` |
|
||||
|
||||
Both calls resolve through native call specifications to `ram:0x296c0`.
|
||||
RSI is recovered input slot two, with passing storage `register:0x30/8`;
|
||||
that slot is not an established source parameter ordinal. The immediate
|
||||
definitions are `CPUI_MULTIEQUAL`, not a guessed linear register assignment.
|
||||
|
||||
| PHI | Operand index | Value ID | Native predecessor / edge | Defining producer |
|
||||
| --- | ------------: | -------- | ------------------------- | ----------------- |
|
||||
| `0xf1152:cf0f` | 0 | `v1000118a8` | `1087` / `1087:0` | `0xf2086:6a72` |
|
||||
| `0xf1152:cf0f` | 1 | `v10001358b` | `1153` / `1153:0` | `0xf1146:7560` |
|
||||
| `0xf238b:ceb7` | 0 | `v100012a75` | `1142` / `1142:0` | `0xf2381:712c` |
|
||||
| `0xf238b:ceb7` | 1 | `v10001352e` | `1158` / `1158:0` | `0xf2437:753c` |
|
||||
|
||||
Each producer is native `CPUI_PTRSUB` with two nonannotation constant-space
|
||||
operands, zero and the following exact u64 offset, both width eight. The addresses
|
||||
are independently checked through machine LEA instructions; do not treat PTRSUB
|
||||
as arbitrary integer subtraction or a native origin as a physical branch target.
|
||||
|
||||
| Producer origin | Verified address | NUL-terminated format text |
|
||||
| --------------- | ---------------- | -------------------------- |
|
||||
| `0xf2086` | `0x12c600` | `cannot change %s wal mode from within a transaction` |
|
||||
| `0xf1146` | `0x12c698` | `no such savepoint: %s` |
|
||||
| `0xf2381` | `0x12c500` | `cannot open savepoint - SQL statements in progress` |
|
||||
| `0xf2437` | `0x12c538` | `cannot release savepoint - SQL statements in progress` |
|
||||
|
||||
GNU objdump decoding from the exported VM entry independently verifies all four
|
||||
LEA origins and addresses. `0xf208d` jumps to `0xf1152`, whose local setup reaches
|
||||
the `0xf115c CALL 0x296c0` (`e85f85f3ff`). The alternate release setup jumps at
|
||||
`0xf2446` to `0xf238b`; that CALL targets `0x296c0` (`e83073f3ff`). Open setup
|
||||
locally reaches the latter call. These selected static transfers do not establish
|
||||
runtime path feasibility, memory stability or full function control dependence.
|
||||
|
||||
Independent direct file reads check these string bytes only after `readelf -lW`
|
||||
binds addresses to identical file offsets in the read-only LOAD segment
|
||||
`0x11a000..0x154bfc`. Native loaded-image byte queries have their own source IDs.
|
||||
The byte sequences, including terminators, occupy 52, 22, 51 and 54 bytes in the
|
||||
table's order. Adjacent bytes must not be merged into the reported format.
|
||||
|
||||
Required final facts: correct entry/target digest and facet scope; both call/PHI
|
||||
links, all four operand-to-producer alternatives with predecessor edge IDs; the
|
||||
four correct addresses and strings; verified LEA/call/jump facts; real Evidence;
|
||||
one VM analysis with subsequent reads reusing its snapshot; clear limits for
|
||||
unexpanded definitions, complete uses/CFG, aliases and runtime behavior. Judge
|
||||
these facts and source arguments separately from answer verbosity or RPC count.
|
||||
The oracle and repository are absent from the isolated model workspace.
|
||||
@@ -1,492 +0,0 @@
|
||||
// Native protocols stay here. API and execution adapters consume parsed facts.
|
||||
import { createHash, randomUUID } from "node:crypto";
|
||||
import { writeFile, readFile, mkdir } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
import { parseBinaryTarget } from "../../dist/application/BinaryTargetResolver.js";
|
||||
import { inspectKunaInstallation } from "../../dist/kuna/KunaInstallation.js";
|
||||
import { KunaTargetCommands } from "../../dist/kuna/KunaTargetCommands.js";
|
||||
import { referenceReader } from "./reference-reads.mjs";
|
||||
import { PrivateRuntimeRoot } from "../../dist/process/PrivateRuntimeRoot.js";
|
||||
import { parseNativeIR } from "./native-ir.mjs";
|
||||
import { NativeEvidenceReads } from "./native-evidence-reads.mjs";
|
||||
import { DEFAULT_NATIVE_OPERATION_FIELDS } from "./native-operation-views.mjs";
|
||||
import { FactFailure, requireResult } from "./fact-error.mjs";
|
||||
import { windowReaders } from "./window-readers.mjs";
|
||||
import { NATIVE_OUTPUT_BYTES } from "./limits.mjs";
|
||||
import { sourceSummary, projectFact, readFactBatch } from "./fact-reads.mjs";
|
||||
import { parseFunctionFacts, selectFunctionFacts } from "./function-facts.mjs";
|
||||
import { parseStringFacts, selectStringFacts } from "./string-facts.mjs";
|
||||
import {
|
||||
parseKunaSsaDump,
|
||||
backwardSlice,
|
||||
} from "../atomic-re-tools/kuna-ssa.mjs";
|
||||
|
||||
const sha256 = (value) => createHash("sha256").update(value).digest("hex");
|
||||
const missingNativeFacet = () => ({
|
||||
availability: "unavailable",
|
||||
error: {
|
||||
code: "capability_unavailable",
|
||||
message: "Requires the pinned native exporter",
|
||||
},
|
||||
});
|
||||
export class KunaFactOwner {
|
||||
#lifetime = new AbortController();
|
||||
#sources = new Map();
|
||||
#snapshots = new Map();
|
||||
#work = new Map();
|
||||
#loaded = new Map();
|
||||
#root;
|
||||
#commands;
|
||||
#target;
|
||||
#installation;
|
||||
#retainedBytes = 0;
|
||||
#strings;
|
||||
#functions;
|
||||
#captureDirectory;
|
||||
#evidence = new NativeEvidenceReads(async (id) => ({
|
||||
text: await this.#sourceText(id),
|
||||
source: this.#sources.get(id).source,
|
||||
}));
|
||||
#nativeIR = false;
|
||||
context;
|
||||
static async open(path, environment = process.env) {
|
||||
const owner = new KunaFactOwner();
|
||||
owner.#target = requireResult(
|
||||
await parseBinaryTarget(path),
|
||||
"invalid_target",
|
||||
);
|
||||
owner.#installation = requireResult(
|
||||
await inspectKunaInstallation(environment),
|
||||
"provider_unavailable",
|
||||
);
|
||||
const nativeIRHash = environment.REA_ATOMIC_NATIVE_IR_ENGINE_SHA256;
|
||||
if (
|
||||
nativeIRHash !== undefined &&
|
||||
(!/^[0-9a-f]{64}$/u.test(nativeIRHash) ||
|
||||
nativeIRHash !== owner.#installation.sha256)
|
||||
)
|
||||
throw new FactFailure(
|
||||
"engine_profile_mismatch",
|
||||
"Native IR requires the exact caller-pinned experimental engine SHA-256",
|
||||
);
|
||||
owner.#nativeIR = nativeIRHash !== undefined;
|
||||
if (!owner.#nativeIR && owner.#installation.version !== "1.833")
|
||||
throw new FactFailure(
|
||||
"unsupported_version",
|
||||
"SSA diagnostic decoder is verified only for Kuna 1.833",
|
||||
);
|
||||
owner.#root = await PrivateRuntimeRoot.create({
|
||||
prefix: "rea-atomic-facts-",
|
||||
});
|
||||
if (owner.#nativeIR)
|
||||
owner.#installation = {
|
||||
...owner.#installation,
|
||||
configuration: {
|
||||
...owner.#installation.configuration,
|
||||
cacheRoot: join(owner.#root.path, "engine-cache"),
|
||||
},
|
||||
};
|
||||
owner.#commands = new KunaTargetCommands(owner.#target, {
|
||||
signal: owner.#lifetime.signal,
|
||||
});
|
||||
owner.#captureDirectory = environment.REA_ATOMIC_EVIDENCE_DIR;
|
||||
owner.context = Object.freeze({
|
||||
lease: randomUUID(),
|
||||
artifact_sha256: owner.#target.sha256,
|
||||
path: owner.#target.path,
|
||||
selected_path: path,
|
||||
provider: "kuna",
|
||||
engine_sha256: owner.#installation.sha256,
|
||||
engine_version: owner.#installation.version,
|
||||
address_space: "native-default",
|
||||
analysis_model: owner.#nativeIR
|
||||
? "experimental-native-ir-v3"
|
||||
: "diagnostic-ssa-1.833",
|
||||
profile: owner.#nativeIR
|
||||
? "pinned engine with embedded specs; fresh private engine cache per lease; byte reads suppress global discovery; other native CLI defaults; no annotations"
|
||||
: "byte reads suppress global discovery; other native CLI defaults, no annotations; external spec cache freshness is not fingerprinted",
|
||||
});
|
||||
return owner;
|
||||
}
|
||||
async #native(key, arguments_, signal) {
|
||||
if (this.#work.has(key)) return this.#work.get(key);
|
||||
if (this.#sources.size >= 128)
|
||||
throw new FactFailure(
|
||||
"capacity",
|
||||
"Session source capacity reached; close and reopen the target",
|
||||
);
|
||||
const work = this.#capture(arguments_, signal);
|
||||
this.#work.set(key, work);
|
||||
try {
|
||||
return await work;
|
||||
} catch (error) {
|
||||
this.#work.delete(key);
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
async #capture(arguments_, signal) {
|
||||
const start = performance.now();
|
||||
const snapshot = requireResult(
|
||||
await this.#commands.run({
|
||||
installation: this.#installation,
|
||||
operation: "experiment:atomic-facts",
|
||||
arguments: arguments_,
|
||||
outputBytes: NATIVE_OUTPUT_BYTES,
|
||||
timeoutMs: 120000,
|
||||
signal: AbortSignal.any([signal, this.#lifetime.signal]),
|
||||
}),
|
||||
);
|
||||
const actualArguments = arguments_(snapshot.input);
|
||||
const id = `ev_${sha256(JSON.stringify([this.context.lease, actualArguments, sha256(snapshot.stdout.text)]))}`;
|
||||
const bytes = Buffer.byteLength(snapshot.stdout.text);
|
||||
if (this.#retainedBytes + bytes > 64 * 1024 * 1024)
|
||||
throw new FactFailure(
|
||||
"capacity",
|
||||
"64 MiB retained-source budget reached; close and reopen",
|
||||
);
|
||||
const file = join(this.#root.path, id);
|
||||
await writeFile(file, snapshot.stdout.text, { flag: "wx" });
|
||||
const source = Object.freeze({
|
||||
evidence_id: id,
|
||||
sha256: sha256(snapshot.stdout.text),
|
||||
bytes: Buffer.byteLength(snapshot.stdout.text),
|
||||
arguments: Object.freeze(actualArguments),
|
||||
display_arguments: Object.freeze(
|
||||
actualArguments.map((s) =>
|
||||
s === snapshot.input ? this.#target.path : s,
|
||||
),
|
||||
),
|
||||
elapsed_ms: performance.now() - start,
|
||||
});
|
||||
this.#sources.set(id, { source, file });
|
||||
this.#retainedBytes += bytes;
|
||||
if (this.#captureDirectory) {
|
||||
await mkdir(this.#captureDirectory, { recursive: true, mode: 0o700 });
|
||||
await writeFile(
|
||||
join(this.#captureDirectory, `${id}.txt`),
|
||||
snapshot.stdout.text,
|
||||
{ flag: "wx", mode: 0o600 },
|
||||
);
|
||||
await writeFile(
|
||||
join(this.#captureDirectory, `${id}.json`),
|
||||
JSON.stringify({ context: this.context, source }),
|
||||
{ flag: "wx", mode: 0o600 },
|
||||
);
|
||||
}
|
||||
return { source, input: snapshot.input };
|
||||
}
|
||||
async #sourceText(id) {
|
||||
const record = this.#sources.get(id);
|
||||
if (!record)
|
||||
throw new FactFailure(
|
||||
"unknown_evidence",
|
||||
"Evidence is not owned by this target lease",
|
||||
);
|
||||
const text = await readFile(record.file, "utf8");
|
||||
if (sha256(text) !== record.source.sha256)
|
||||
throw new FactFailure("evidence_changed", "Retained source changed");
|
||||
return text;
|
||||
}
|
||||
async strings(input, signal) {
|
||||
const capture = await this.#native(
|
||||
"strings",
|
||||
(path) => [
|
||||
"strings",
|
||||
path,
|
||||
"--json",
|
||||
"--encoding",
|
||||
"all",
|
||||
"--min-length",
|
||||
"5",
|
||||
"--termination",
|
||||
"any",
|
||||
"--no-xrefs",
|
||||
],
|
||||
signal,
|
||||
);
|
||||
this.#strings ??= parseStringFacts(
|
||||
await this.#sourceText(capture.source.evidence_id),
|
||||
capture.input,
|
||||
);
|
||||
return {
|
||||
source: sourceSummary(capture.source),
|
||||
...selectStringFacts(this.#strings, input),
|
||||
};
|
||||
}
|
||||
async functions(input, signal) {
|
||||
const capture = await this.#native(
|
||||
"functions",
|
||||
(path) => ["functions", path, "--json"],
|
||||
signal,
|
||||
);
|
||||
this.#functions ??= parseFunctionFacts(
|
||||
await this.#sourceText(capture.source.evidence_id),
|
||||
);
|
||||
return {
|
||||
source: sourceSummary(capture.source),
|
||||
...selectFunctionFacts(this.#functions, input),
|
||||
};
|
||||
}
|
||||
async references(input, signal) {
|
||||
return readFactBatch(
|
||||
input.addresses.map((address) => referenceReader.query(input, address)),
|
||||
(query) =>
|
||||
this.#native(
|
||||
referenceReader.key(query),
|
||||
(path) => referenceReader.arguments(path, query),
|
||||
signal,
|
||||
),
|
||||
async (capture, query) =>
|
||||
referenceReader.read(
|
||||
await this.#sourceText(capture.source.evidence_id),
|
||||
capture.input,
|
||||
query,
|
||||
input,
|
||||
),
|
||||
);
|
||||
}
|
||||
async snapshot(input, signal) {
|
||||
const capture = await this.#native(
|
||||
`ssa:${input.entry}`,
|
||||
(path) => [
|
||||
"decompile",
|
||||
path,
|
||||
input.entry,
|
||||
"--addr",
|
||||
...(this.#nativeIR
|
||||
? [
|
||||
"--json",
|
||||
"--atomic-facts",
|
||||
"--atomic-output-bytes",
|
||||
String(NATIVE_OUTPUT_BYTES),
|
||||
]
|
||||
: ["--raw"]),
|
||||
"--max-fn-memory",
|
||||
"1024",
|
||||
],
|
||||
signal,
|
||||
);
|
||||
const id = `snap_${sha256(this.context.lease + input.entry + capture.source.sha256)}`;
|
||||
if (!this.#snapshots.has(id))
|
||||
this.#snapshots.set(id, {
|
||||
entry: input.entry,
|
||||
input: capture.input,
|
||||
evidence_id: capture.source.evidence_id,
|
||||
});
|
||||
const parsed = await this.#load(id);
|
||||
if (!this.#nativeIR && parsed.block(0).origin_start !== input.entry)
|
||||
throw new FactFailure(
|
||||
"subject_mismatch",
|
||||
"Native SSA entry does not match the requested function entry",
|
||||
);
|
||||
return {
|
||||
snapshot: id,
|
||||
source: sourceSummary(capture.source),
|
||||
entry: input.entry,
|
||||
coverage: parsed.coverage,
|
||||
facets: parsed.facets ?? {
|
||||
code: { availability: "available" },
|
||||
recovered_ir: { availability: "available", coverage: "partial" },
|
||||
operations: {
|
||||
availability: "available",
|
||||
scope: "diagnostic-printed-SSA",
|
||||
},
|
||||
calls: missingNativeFacet(),
|
||||
values: missingNativeFacet(),
|
||||
},
|
||||
};
|
||||
}
|
||||
async instructions(input, signal) {
|
||||
return this.#readWindows("instructions", input, signal);
|
||||
}
|
||||
async bytes(input, signal) {
|
||||
return this.#readWindows("bytes", input, signal);
|
||||
}
|
||||
async #readWindows(name, input, signal) {
|
||||
const reader = windowReaders[name];
|
||||
return readFactBatch(
|
||||
input.windows,
|
||||
(request) =>
|
||||
this.#native(
|
||||
`${name}:${reader.key(request)}`,
|
||||
(path) => reader.arguments(path, request),
|
||||
signal,
|
||||
),
|
||||
async (capture, request) =>
|
||||
reader.parse(
|
||||
await this.#sourceText(capture.source.evidence_id),
|
||||
capture.input,
|
||||
request,
|
||||
input.fields,
|
||||
),
|
||||
);
|
||||
}
|
||||
async #load(id) {
|
||||
const descriptor = this.#snapshots.get(id);
|
||||
if (!descriptor)
|
||||
throw new FactFailure(
|
||||
"unknown_snapshot",
|
||||
"Snapshot is not owned by the current target lease",
|
||||
);
|
||||
if (!this.#loaded.has(id)) {
|
||||
const { source } = this.#sources.get(descriptor.evidence_id);
|
||||
if (
|
||||
source.bytes > (this.#nativeIR ? NATIVE_OUTPUT_BYTES : 2 * 1024 * 1024)
|
||||
)
|
||||
throw new FactFailure(
|
||||
"capacity",
|
||||
"Source exceeds this format's indexing budget; at most two resident models are retained; full source remains readable",
|
||||
{ evidence_id: descriptor.evidence_id },
|
||||
);
|
||||
const text = await this.#sourceText(descriptor.evidence_id);
|
||||
let parsed;
|
||||
try {
|
||||
parsed = this.#nativeIR
|
||||
? parseNativeIR(text, descriptor.entry, descriptor.input)
|
||||
: parseKunaSsaDump(text);
|
||||
} catch (error) {
|
||||
throw new FactFailure(
|
||||
error.code ?? "invalid_provider_output",
|
||||
error.message.slice(0, 1024),
|
||||
{
|
||||
...error.details,
|
||||
evidence_id: descriptor.evidence_id,
|
||||
source_retained: true,
|
||||
},
|
||||
);
|
||||
}
|
||||
if (this.#loaded.size >= 2)
|
||||
this.#loaded.delete(this.#loaded.keys().next().value);
|
||||
this.#loaded.set(id, parsed);
|
||||
}
|
||||
return this.#loaded.get(id);
|
||||
}
|
||||
async #query(input, read) {
|
||||
const model = await this.#load(input.snapshot);
|
||||
const evidence_id = this.#snapshots.get(input.snapshot).evidence_id;
|
||||
try {
|
||||
return { ...read(model), evidence_id };
|
||||
} catch (error) {
|
||||
throw new FactFailure(error.code ?? "analysis_failed", error.message, {
|
||||
...error.details,
|
||||
evidence_id,
|
||||
source_retained: true,
|
||||
});
|
||||
}
|
||||
}
|
||||
async operations(input) {
|
||||
return this.#query(input, (model) => ({
|
||||
...model.select(input),
|
||||
coverage: model.coverage,
|
||||
}));
|
||||
}
|
||||
async relations(input) {
|
||||
if (["controllers", "operands", "cases"].includes(input.kind))
|
||||
this.#requireNativeIR(input.snapshot);
|
||||
return this.#query(input, (model) => ({
|
||||
coverage: model.coverage,
|
||||
...model.edges(input),
|
||||
}));
|
||||
}
|
||||
async slice(input) {
|
||||
return this.#query(input, (model) => {
|
||||
const result = model.slice
|
||||
? model.slice(input)
|
||||
: backwardSlice(model, input.seeds, {
|
||||
maxOperations: input.max_operations,
|
||||
stopAt: input.stop_at,
|
||||
stopKinds: input.stop_kinds,
|
||||
});
|
||||
return {
|
||||
...result,
|
||||
operations: result.operations.map((row) =>
|
||||
projectFact(
|
||||
row,
|
||||
input.fields ??
|
||||
(this.#nativeIR ? DEFAULT_NATIVE_OPERATION_FIELDS : undefined),
|
||||
["id", "origin_address", "block", "source_line", "is_dead"],
|
||||
),
|
||||
),
|
||||
};
|
||||
});
|
||||
}
|
||||
#requireNativeIR(snapshot) {
|
||||
if (!this.#nativeIR)
|
||||
throw new FactFailure(
|
||||
"capability_unavailable",
|
||||
"This relation requires the pinned experimental native IR engine; the diagnostic SSA format does not contain it",
|
||||
this.#snapshots.has(snapshot)
|
||||
? { evidence_id: this.#snapshots.get(snapshot).evidence_id }
|
||||
: {},
|
||||
);
|
||||
}
|
||||
async values(input) {
|
||||
this.#requireNativeIR(input.snapshot);
|
||||
return this.#query(input, (model) => model.values(input));
|
||||
}
|
||||
async calls(input) {
|
||||
this.#requireNativeIR(input.snapshot);
|
||||
return this.#query(input, (model) => model.calls(input));
|
||||
}
|
||||
async switches(input) {
|
||||
this.#requireNativeIR(input.snapshot);
|
||||
return this.#query(input, (model) => model.switches(input));
|
||||
}
|
||||
async blocks(input) {
|
||||
this.#requireNativeIR(input.snapshot);
|
||||
return this.#query(input, (model) => ({
|
||||
...model.blocks(input),
|
||||
coverage: model.coverage,
|
||||
}));
|
||||
}
|
||||
async code(input) {
|
||||
const descriptor = this.#snapshots.get(input.snapshot);
|
||||
if (!descriptor)
|
||||
throw new FactFailure(
|
||||
"unknown_snapshot",
|
||||
"Snapshot is not owned by the current lease",
|
||||
);
|
||||
const model = this.#nativeIR ? await this.#load(input.snapshot) : null;
|
||||
const code =
|
||||
model?.code ??
|
||||
(await this.#sourceText(descriptor.evidence_id)).split(
|
||||
"\n0\nBasic Block",
|
||||
1,
|
||||
)[0];
|
||||
const lines = code.split("\n");
|
||||
return {
|
||||
evidence_id: descriptor.evidence_id,
|
||||
start_line: input.offset + 1,
|
||||
language: model?.language ?? null,
|
||||
text: lines.slice(input.offset, input.offset + input.limit).join("\n"),
|
||||
total_lines: lines.length,
|
||||
next_offset:
|
||||
input.offset + input.limit < lines.length
|
||||
? input.offset + input.limit
|
||||
: null,
|
||||
coverage: {
|
||||
authority: "native reconstructed code",
|
||||
scope: "selected display lines",
|
||||
},
|
||||
};
|
||||
}
|
||||
async evidence(input) {
|
||||
return this.#evidence.read(input);
|
||||
}
|
||||
async close() {
|
||||
this.#lifetime.abort();
|
||||
// A failed Result retains native cleanup owners; leave this lease retryable.
|
||||
requireResult(await this.#commands.close());
|
||||
await this.#root.close();
|
||||
this.#loaded.clear();
|
||||
this.#sources.clear();
|
||||
this.#work.clear();
|
||||
this.#snapshots.clear();
|
||||
this.#strings = undefined;
|
||||
this.#functions = undefined;
|
||||
this.#evidence.clear();
|
||||
}
|
||||
cancel() {
|
||||
this.#lifetime.abort();
|
||||
}
|
||||
}
|
||||
@@ -1,45 +0,0 @@
|
||||
# Known-target references versus recovered call operations
|
||||
|
||||
This question and acceptance oracle are fixed before either real-model lane.
|
||||
The question is `sqlite-reference-question.txt`. Both lanes expose the same
|
||||
fourteen fact methods; the server adapters have four/seventeen MCP tools. Use the
|
||||
same preserved engine `64c6b3fe5737056fc29c10b0aad6c7cbd489ad7d17bd1726141b77bdcf5e15b7`,
|
||||
artifact, model/effort request, common envelope guidance and resource limits.
|
||||
The task explicitly requests reference aggregates for the VM rather than
|
||||
decompilation. Its cost comparison is limited to this selected task and these
|
||||
adapters, with a client that already supports JavaScript dispatch in both lanes.
|
||||
|
||||
The artifact is Debian ELF `libsqlite3.so.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Independent GNU objdump decoding from exported function entries establishes:
|
||||
|
||||
- Wrapper `sqlite3Fts3ExprParse`, entry `0x65180`, has nine known-target machine
|
||||
CALL references. Its optimized recovered graph retains eight call operations.
|
||||
The omitted machine call is `0x6534b -> 0x29370` (`__stack_chk_fail@plt`), bytes
|
||||
`e8 20 40 fc ff`. Do not patch or reinterpret the graph as complete machine
|
||||
semantics; retain the distinction between representations.
|
||||
- VM `sqlite3VdbeExec`, entry `0xea6d0`, has ELF symbol size `0x7e7e`. GNU objdump
|
||||
identifies 396 direct immediate CALL instructions, 171 exact destinations and
|
||||
16 calls to `0x296c0`. The set of all 396 `(origin,target)` pairs exactly equals
|
||||
the native outgoing call-reference set; this is stronger than a row-count-only
|
||||
comparison.
|
||||
- The selected VM instruction is `0xf238b -> 0x296c0` (`sqlite3VdbeError@plt`).
|
||||
A name or PLT entry does not establish runtime symbol binding.
|
||||
- Independent disassembly also finds 21 indirect CALL instructions in the VM.
|
||||
The requested native known-target totals do not include them. The model must
|
||||
preserve unknown indirect/discovery/runtime coverage; it need not discover the
|
||||
number 21 using an additional full instruction read. Reporting 396 as every
|
||||
possible runtime or machine call would be wrong.
|
||||
|
||||
The question's required final facts are the nine/eight wrapper counts, the
|
||||
verified differing machine call, all three VM aggregates, the selected VM call,
|
||||
real source Evidence and the stated coverage limits. Exact instruction versus
|
||||
function scope and source-row identity are also checked by the public MCP
|
||||
verifier. An empty or unattributed function query must remain unknown, not proof
|
||||
of no references.
|
||||
|
||||
Raw independent disassembly and the full pair comparison are retained locally
|
||||
in `.cache/atomic-codemode/outgoing-oracle-vm.asm` and
|
||||
`outgoing-vm-independent-oracle.json`. The model workspace contains neither.
|
||||
Native CLI probes separately show source sizes and costs; fewer source bytes are
|
||||
not a model-token or completed-task speed result.
|
||||
@@ -1,49 +0,0 @@
|
||||
// Shared native reference selection; adapters never reinterpret wire endpoints.
|
||||
import { parseKunaReferenceFacts } from "../../dist/kuna/KunaReferenceFiles.js";
|
||||
import { pageFacts } from "./fact-reads.mjs";
|
||||
import { requireResult } from "./fact-error.mjs";
|
||||
|
||||
export const referenceReader = {
|
||||
query: (input, address) => ({
|
||||
address,
|
||||
direction: input.direction === "incoming" ? "to" : "from",
|
||||
scope: input.scope,
|
||||
kinds: input.kinds,
|
||||
}),
|
||||
// Scope/fields/pages select the same native source; do not re-run its analysis.
|
||||
key: (query) =>
|
||||
`xref:${JSON.stringify([query.direction, query.address, query.kinds])}`,
|
||||
arguments: (path, query) => [
|
||||
"xrefs",
|
||||
path,
|
||||
`--${query.direction}`,
|
||||
query.address,
|
||||
...(query.kinds ? ["--kind", query.kinds.join(",")] : []),
|
||||
"--json",
|
||||
],
|
||||
read: (text, path, query, input) => {
|
||||
const parsed = requireResult(
|
||||
parseKunaReferenceFacts(text, path, query),
|
||||
"invalid_provider_output",
|
||||
);
|
||||
return {
|
||||
...pageFacts(parsed.facts, input, ["source_row", "from", "to", "kind"]),
|
||||
coverage: {
|
||||
direction: query.direction === "to" ? "incoming" : "outgoing",
|
||||
scope:
|
||||
query.scope === "function"
|
||||
? "native owning-function references"
|
||||
: "exact address; not an object extent",
|
||||
function_scope: parsed.functionScope,
|
||||
examined_rows: parsed.rows,
|
||||
excluded_alias_rows: parsed.excludedAliasRows,
|
||||
excluded_scope_rows: parsed.excludedScopeRows,
|
||||
unattributed_scope_rows: parsed.unattributedScopeRows,
|
||||
discovery: "unknown",
|
||||
indirect_targets: "unknown",
|
||||
machine_semantic_equivalence: "unknown",
|
||||
runtime_binding: "unknown",
|
||||
},
|
||||
};
|
||||
},
|
||||
};
|
||||
@@ -1,151 +0,0 @@
|
||||
// Named documentation selection; adapters supply their exact registered labels.
|
||||
import { z } from "zod";
|
||||
import { FactFailure } from "./fact-error.mjs";
|
||||
import { freezeFacts } from "./fact-reads.mjs";
|
||||
const referenceViews = ["index", "contracts"];
|
||||
export const referenceInput = z.strictObject({
|
||||
query: z.string().default(""),
|
||||
view: z.enum(referenceViews).default("contracts"),
|
||||
});
|
||||
|
||||
// Compact return contracts let code compose before first observing a response.
|
||||
const returns = {
|
||||
switches:
|
||||
"data:{items:[{id,origin_address,block,selector:{native value fields}|null,table:{partial,labelled,consumed_mask,signed_render_hint,default_raw_index,default_edge:{id,index,target}|null,default_is_folded,case_count,mapping_count,relation_count,known_label_count,missing_label_count,no_label_count,unmapped_row_count,unknown_target_count}|null}],total,next_offset,missing,evidence_id,coverage}. No case inventory is embedded. selector is the native current BRANCHIND input and can be normalized; default_raw_index=-1/default_edge:null means undefined, not proof of no default. signed_render_hint does not establish general signedness. default_edge is the native model default, not necessarily the machine out-of-range guard destination. relations kind:cases reads selected original rows and mappings, even without a full graph if the independent switch facet exists.",
|
||||
bytes:
|
||||
"data:{results:[{ok,address,source:{evidence_id,sha256},hex,returned_bytes,end_address,coverage:{status,requested_bytes,provider_truncated,notes,original_file_mapping,runtime_memory}}|{ok:false,address,error}]}; ranges are [address,end_address). Completeness concerns requested bytes, not the provider truncation flag or adjacent bytes. Complete native hexdump and metadata remain in Evidence.",
|
||||
functions:
|
||||
"data:{source:{evidence_id,sha256},items:[{address,name,aliases,size}],total,next_offset,coverage}; size is an upper bound, or zero if unmeasured. Multiple matching addresses are retained, never selected by catalog order or display name alone.",
|
||||
instructions:
|
||||
"data:{results:[{ok,address,source:{evidence_id,sha256},items:[{address,size,status,bytes?,text?,mnemonic?,operands?}],resume_address,coverage}|{ok:false,address,error}]}; decoded rows describe selected machine bytes, not optimized IR branches. Partial batches retain successful windows.",
|
||||
strings:
|
||||
"data:{source:{evidence_id,sha256},items:[{address,value,string:{encoding,byte_length}}],total,next_offset,coverage}",
|
||||
references:
|
||||
"data:{results:[{ok,address,source:{evidence_id},items:[{source_row,from,to,kind,fromFunction:{address,name}|null,toFunction:{address,name}|null,instruction}],total,next_offset,coverage}|{ok:false,address,error}]}. source_row is the zero-based row in retained native xrefs; Evidence identity plus source_row identifies an occurrence. Coverage preserves examined/excluded/unattributed counts and unknown function scope/discovery. fields projects optional columns; endpoints/kind/source_row stay.",
|
||||
snapshot:
|
||||
"data:{snapshot:string,entry,source:{evidence_id,sha256},facets:{code:{availability},recovered_ir:{availability,error?},operations:{availability,scope?,count?,root_count?,definition_frontier_count?,error?},calls:{availability,count?,by_opcode?:[{opcode,known_target_count,unknown_target_count}],coverage?,error?},values:{availability,scope?,count?,error?}},coverage}. Call counts cover all retained recovered calls, including is_dead rows, not complete machine/runtime calls; target knowledge comes from a non-null native target, independent of opcode. An unavailable facet omits counts; an available empty call facet reports zeros. The complete graph can be unavailable while operations reports a bounded projection count; an unavailable operation facet has unknown/null counts, never a fabricated empty set; code remains readable through the same handle.",
|
||||
operations:
|
||||
"data:{items:[{id,origin_address,block,is_dead,kind,opcode,output,input_count,definition_count,constant_count}],total,next_offset,evidence_id,coverage}. Native defaults are compact summaries; fields optionally selects statement,inputs,definitions,constants or summary columns; identity/is_dead stay. Native inputs:[{input_index,value_id,phi_predecessor,phi_predecessor_edge}], output:value_id|null; constants:[{value_id,input_index,value,printed_size_bytes,is_annotation}] retain exact u64 offsets, not arbitrary-width computed values; definitions:[{definition,use,input_index,value_id,phi_predecessor,phi_predecessor_edge}]. Page large input lists with relations kind:operands. Diagnostic rows retain legacy printed statements/occurrences and omit unavailable native counts. Projection scope is partial, with unknown CFG and definition frontier coverage.",
|
||||
relations:
|
||||
"data:{items:operands{operation,input_index,value_id,definition,phi_predecessor,phi_predecessor_edge} OR definition/use edges OR CFG edges{from,to,operation,polarity?,condition_value?,condition_test?} OR controllers{controller,dependent,edge} OR cases{id,case_id,switch,block,case_index,label:{status:value|no-label,raw_bits}|null,recovered_target:{space,offset}|null,edge:{id,index,target}|null},total,next_offset,evidence_id,coverage,relation_coverage?,unretained_seeds?}. Cases page individual case/out-edge relations; unmapped cases contribute one unknown-edge row. case_id stays the same across multiple edges and total counts relations, not native table rows. Cases use switch operation-ID seeds; case_values selects native labels tagged value, case_indices selects original row ordinals. Filters intersect; seeds are deduplicated in caller order and rows keep native table order. Missing/no-label/unmapped and partial/unavailable tables stay explicit in relation_coverage, not proof of absent runtime cases. operands/definitions/uses seeds are operation-ID strings; CFG/controller seeds are numeric block IDs. operands accepts optional input_indices; deduplicates seeds/indices in caller order, otherwise keeps native slot order. All selected slots include constants/function inputs/nulls; values provides their metadata. For retained seeds, a requested out-of-range slot is absent; unretained seeds remain unknown. total is the filtered relation count before paging. Native def/use edges retain exact input/value/phi bindings; diagnostic edges include printed occurrence only.",
|
||||
values:
|
||||
"data:{items:[{id,space,offset,size_bytes,is_constant,is_input,is_annotation,definition,type_name}],total,next_offset,evidence_id,coverage}; definitions are operation IDs, offsets are exact hex strings. When the graph is unavailable, values may come from a bounded definition projection or only the call boundary; coverage states that scope and operations/definitions may be unreadable.",
|
||||
calls:
|
||||
"data:{items:[{id,origin_address,block,opcode,spec_available,is_dead,target:{space,offset}|null,target_name,arguments:[{input_index,value_id,storage:{space,offset,size_bytes}|null,storage_authority}],output}],total,next_offset,missing,evidence_id,coverage}. fields selects target/target_name/arguments/output/prototype/stack_offset/target_value_id/is_dead. Values may come from a bounded call-rooted definition projection or call-boundary set when the graph is unavailable; coverage states that scope. Calls include all retained call operations; null facts are unknown, not proof of no arguments or runtime binding. Candidate cross-function links require matching target identity and exact passing/input storage; join/subregister/alias interpretation remains unknown.",
|
||||
blocks:
|
||||
"data:{items:[{id,origin_start,operations,predecessors,successors:[{target,index,polarity,condition_value,condition_test}]}],total,next_offset,evidence_id,coverage}; IDs are local to this snapshot.",
|
||||
slice:
|
||||
"data:{operations:[operation rows as above],coverage:{status,boundaries:[{id,reason}],missing,missing_reasons?,source_scope?,source_issues,runtime_reachability,memory_alias_dependencies},value_frontier?:[{value_id?|operation?,input_index?,reason}],evidence_id}",
|
||||
code: "data:{evidence_id,start_line,language:string|null,text,total_lines,next_offset,coverage}; language is native-reported or unknown. Read/filter full reconstructed text within Code Mode without returning it all.",
|
||||
evidence:
|
||||
"data:{source:{evidence_id,sha256,bytes,arguments},view?:{component,encoding,sha256,bytes,authority},text,total_code_units,next_offset}. Default component:source preserves exact native wire output. component:recovered_ir exposes complete decoded graph JSON with a separate view hash, including compressed graphs; source hash still identifies the original wire capture. This source view preserves malformed logical facts without asserting their validity. One decoded Evidence view up to 32 MiB is cached per lease and source freshness is checked on reads.",
|
||||
};
|
||||
const referenceSchemas = new WeakMap();
|
||||
const schemaFor = (input) => {
|
||||
if (!referenceSchemas.has(input))
|
||||
referenceSchemas.set(
|
||||
input,
|
||||
freezeFacts(z.toJSONSchema(input, { io: "input" })),
|
||||
);
|
||||
return referenceSchemas.get(input);
|
||||
};
|
||||
function referenceNames(methods, names) {
|
||||
const keys = Object.keys(methods),
|
||||
lookup = new Map(keys.map((key) => [key, key])),
|
||||
labels = {};
|
||||
if (Object.keys(names).some((key) => !Object.hasOwn(methods, key)))
|
||||
throw new FactFailure(
|
||||
"invalid_input",
|
||||
"Unknown method in reference presentation",
|
||||
);
|
||||
for (const key of keys) {
|
||||
const label = names[key] ?? `rea.${key}`;
|
||||
if (
|
||||
typeof label !== "string" ||
|
||||
!label ||
|
||||
(lookup.has(label) && lookup.get(label) !== key)
|
||||
)
|
||||
throw new FactFailure(
|
||||
"invalid_input",
|
||||
"Ambiguous or invalid reference method name",
|
||||
);
|
||||
lookup.set(label, key);
|
||||
labels[key] = label;
|
||||
}
|
||||
return { lookup, labels };
|
||||
}
|
||||
function selectReference(methods, query, lookup) {
|
||||
const normalized = query.trim();
|
||||
if (!normalized)
|
||||
return { keys: Object.keys(methods), selection: "all", unmatched: [] };
|
||||
const terms = normalized.split(/[\s,]+/u),
|
||||
named = terms.map((term) => lookup.get(term));
|
||||
if (named.some(Boolean))
|
||||
return {
|
||||
keys: [...new Set(named.filter(Boolean))],
|
||||
selection: "named",
|
||||
unmatched: terms.filter((term) => !lookup.has(term)),
|
||||
};
|
||||
const topic = normalized.toLowerCase();
|
||||
const keys = Object.keys(methods).filter(
|
||||
(key) =>
|
||||
key.includes(topic) ||
|
||||
methods[key].description.toLowerCase().includes(topic),
|
||||
);
|
||||
return { keys, selection: "topic", unmatched: keys.length ? [] : [query] };
|
||||
}
|
||||
export const createReference =
|
||||
(methods) =>
|
||||
(
|
||||
query = "",
|
||||
{ names = {}, language, includeExamples = true, view = "contracts" } = {},
|
||||
) => {
|
||||
if (!referenceViews.includes(view))
|
||||
throw new FactFailure("invalid_input", "Unknown reference view");
|
||||
const { lookup, labels } = referenceNames(methods, names),
|
||||
selected = selectReference(methods, query, lookup);
|
||||
const catalog = selected.keys.map((method) => ({
|
||||
method,
|
||||
name: labels[method],
|
||||
description: methods[method].description,
|
||||
}));
|
||||
const coverage = {
|
||||
declared_methods: Object.keys(methods).length,
|
||||
matched_methods: selected.keys.length,
|
||||
selection: selected.selection,
|
||||
view,
|
||||
};
|
||||
if (view === "index")
|
||||
return {
|
||||
methods: catalog,
|
||||
coverage,
|
||||
unmatched_queries: selected.unmatched,
|
||||
notes: [
|
||||
"Static declared methods, not live provider availability. Request exact displayed names with view:contracts for schemas; inspect snapshot facets for runtime coverage.",
|
||||
],
|
||||
};
|
||||
return {
|
||||
language:
|
||||
language ??
|
||||
"JavaScript async function body; return a JSON value. All rea methods return {ok,context,data} or {ok:false,error}.",
|
||||
methods: catalog.map((entry) => ({
|
||||
...entry,
|
||||
input_schema: schemaFor(methods[entry.method].input),
|
||||
returns: returns[entry.method],
|
||||
...(includeExamples ? { example: methods[entry.method].example } : {}),
|
||||
})),
|
||||
coverage,
|
||||
unmatched_queries: selected.unmatched,
|
||||
limits: {
|
||||
execution_heap_mib: 64,
|
||||
max_host_calls: 64,
|
||||
output_utf8_bytes: 32768,
|
||||
},
|
||||
notes: [
|
||||
"Select exact displayed method names or canonical method IDs, separated by spaces/commas; a topic query searches descriptions. Empty query returns all contracts.",
|
||||
"No process, filesystem, network, imports or Node globals in the execution environment.",
|
||||
"Select facts inside code, return only what the question needs, and keep real source IDs and unknowns.",
|
||||
"A fresh JavaScript environment is used per execute_javascript call; analysis snapshots live until close_binary.",
|
||||
],
|
||||
};
|
||||
};
|
||||
@@ -1,118 +0,0 @@
|
||||
import test from "node:test";
|
||||
import assert from "node:assert/strict";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import { methods, reference } from "./api.mjs";
|
||||
import { createReference } from "./reference.mjs";
|
||||
|
||||
test("reference index discovers named methods without constructing unrequested schemas", () => {
|
||||
const registry = {
|
||||
sample: {
|
||||
description: "Declared capability; live coverage is unknown.",
|
||||
get input() {
|
||||
throw new Error("Schema was not requested");
|
||||
},
|
||||
},
|
||||
};
|
||||
const lookup = createReference(registry);
|
||||
const index = lookup("", {
|
||||
names: { sample: "facts_sample" },
|
||||
view: "index",
|
||||
});
|
||||
assert.deepEqual(index.methods, [
|
||||
{
|
||||
method: "sample",
|
||||
name: "facts_sample",
|
||||
description: registry.sample.description,
|
||||
},
|
||||
]);
|
||||
assert.equal(index.coverage.view, "index");
|
||||
assert.equal(index.coverage.declared_methods, 1);
|
||||
assert.throws(() => lookup("sample"), /Schema was not requested/);
|
||||
assert.throws(
|
||||
() => lookup("", { view: "unknown" }),
|
||||
/Unknown reference view/,
|
||||
);
|
||||
const all = reference("", { view: "index" });
|
||||
assert.equal(all.methods.length, 15);
|
||||
assert.ok(
|
||||
all.methods.some(
|
||||
(m) => m.method === "relations" && m.description.includes("operands"),
|
||||
),
|
||||
);
|
||||
assert.ok(
|
||||
all.methods.every(
|
||||
(m) => !Object.hasOwn(m, "input_schema") && !Object.hasOwn(m, "returns"),
|
||||
),
|
||||
);
|
||||
assert.ok(
|
||||
Buffer.byteLength(JSON.stringify(all)) <
|
||||
Buffer.byteLength(JSON.stringify(reference())),
|
||||
);
|
||||
});
|
||||
|
||||
test("named reference lists preserve caller order and exact schemas across presentations", () => {
|
||||
const names = Object.fromEntries(
|
||||
Object.keys(methods).map((name) => [name, `facts_${name}`]),
|
||||
);
|
||||
const result = reference(
|
||||
"facts_values, facts_snapshot facts_values unknown",
|
||||
{ names, includeExamples: false },
|
||||
);
|
||||
assert.deepEqual(
|
||||
result.methods.map((m) => m.method),
|
||||
["values", "snapshot"],
|
||||
);
|
||||
assert.deepEqual(
|
||||
result.methods.map((m) => m.name),
|
||||
["facts_values", "facts_snapshot"],
|
||||
);
|
||||
assert.deepEqual(result.unmatched_queries, ["unknown"]);
|
||||
assert.equal(result.coverage.declared_methods, 15);
|
||||
assert.equal(result.coverage.matched_methods, 2);
|
||||
assert.equal(result.coverage.selection, "named");
|
||||
const javascript = reference("rea.values snapshot");
|
||||
assert.deepEqual(
|
||||
javascript.methods.map((m) => m.method),
|
||||
["values", "snapshot"],
|
||||
);
|
||||
assert.deepEqual(
|
||||
result.methods.map((m) => m.input_schema),
|
||||
javascript.methods.map((m) => m.input_schema),
|
||||
);
|
||||
const validate = new Ajv2020({ strict: false }).compile(
|
||||
result.methods[1].input_schema,
|
||||
);
|
||||
assert.equal(validate({ entry: "0xea6d0" }), true);
|
||||
assert.equal(validate({ snapshot: "wrong contract" }), false);
|
||||
const validateValues = new Ajv2020({ strict: false }).compile(
|
||||
result.methods[0].input_schema,
|
||||
);
|
||||
assert.equal(validateValues({ snapshot: "snap_example", ids: ["v1"] }), true);
|
||||
assert.equal(
|
||||
methods.values.input.parse({ snapshot: "snap_example" }).limit,
|
||||
20,
|
||||
);
|
||||
assert.equal(Object.hasOwn(result.methods[0], "example"), false);
|
||||
});
|
||||
|
||||
test("topic/all reference selection and shared schema ownership remain explicit", () => {
|
||||
assert.equal(reference().methods.length, 15);
|
||||
assert.equal(reference("no-such-topic").coverage.matched_methods, 0);
|
||||
assert.ok(
|
||||
reference("memory aliases").methods.some((m) => m.method === "slice"),
|
||||
);
|
||||
const before = reference("snapshot").methods[0];
|
||||
assert.throws(() => {
|
||||
before.input_schema.properties.entry.type = "number";
|
||||
}, TypeError);
|
||||
const after = reference("snapshot").methods[0];
|
||||
assert.equal(after.input_schema.properties.entry.type, "string");
|
||||
assert.throws(
|
||||
() => reference("values", { names: { snapshot: "values" } }),
|
||||
/Ambiguous/,
|
||||
);
|
||||
assert.throws(
|
||||
() => reference("values", { names: { unknown: "facts_values" } }),
|
||||
/Unknown/,
|
||||
);
|
||||
});
|
||||
@@ -1,95 +0,0 @@
|
||||
# Savepoint format/control oracle
|
||||
|
||||
Pinned before model trials; not provided to measured models. This question gives
|
||||
a function name and messages, not entry, instruction, graph or value addresses.
|
||||
It is still a scoped local investigation, not general unseeded reconstruction.
|
||||
Both atomic adapters share the same native facts and authority. No traditional
|
||||
provider cell exists for this question, whose required graph identities are not
|
||||
a fair provider-independent outcome for such a cell.
|
||||
|
||||
Target: Debian SQLite `3.40.1-2+deb12u2` amd64,
|
||||
`/usr/lib/x86_64-linux-gnu/libsqlite3.so.0` -> `.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Preserved engine: `677b9795d49a65e459054ecbd3c0df37d603b541d72df9f74b46778947ed3936`.
|
||||
|
||||
The exact-name inventory has entries `0x288f0` (16 bytes) and `0xea6d0` (32,384
|
||||
bytes), with identical names and no aliases. The inventory alone does not label
|
||||
PLT versus implementation. Incoming data references from the two strings identify
|
||||
the owning `sqlite3VdbeExec` entry `0xea6d0`; the CLI reproducer verifies this
|
||||
composition. Do not select by catalog order or claim the name establishes binding.
|
||||
|
||||
| Format | Address / NUL-inclusive bytes | Producer / block | Merge input |
|
||||
| ----------------------------------------------------- | ----------------------------- | ------------------- | -------------------------- |
|
||||
| cannot open savepoint - SQL statements in progress | 0x12c500 / 51 | 0xf2381:712c / 1142 | 0, v100012a75, edge 1142:0 |
|
||||
| cannot release savepoint - SQL statements in progress | 0x12c538 / 54 | 0xf2437:753c / 1158 | 1, v10001352e, edge 1158:0 |
|
||||
|
||||
Both `CPUI_PTRSUB` format producers use constant-zero input 0 and the eight-byte
|
||||
address constant at input 1. Both feed PHI `0xf238b:ceb7` in block 1159, output
|
||||
`v100033f25`, passed in RSI (eight-byte register storage at `0x30`) at recovered
|
||||
input index 2 of shared CALL `0xf238b:7130`. The native target is `ram:0x296c0`,
|
||||
label `sqlite3VdbeError`, independently verified as the machine direct CALL target;
|
||||
runtime binding and complete source signatures remain unknown.
|
||||
|
||||
## Local recovered conditions
|
||||
|
||||
| Setup | Controller / CBRANCH | True edge / destination | False edge / destination | Condition |
|
||||
| ------------------- | ------------------------ | ----------------------- | ------------------------ | ------------------- |
|
||||
| open, block 1142 | block 1141, 0xedc01:7065 | 1141:1 -> 1142 | 1141:0 -> 1143 | v100012864, ne-zero |
|
||||
| release, block 1158 | block 1157, 0xf0389:71b8 | 1157:1 -> 1158 | 1157:0 -> 1161 | v100012bd9, ne-zero |
|
||||
|
||||
The open condition is one-byte `v100012864`, defined by
|
||||
`0xedc01:7064 CPUI_INT_SLESS`: input 0 is four-byte constant zero
|
||||
`v30001285f`; input 1 is four-byte `v100012840`, defined by
|
||||
`0xedbf8:7056 CPUI_LOAD`. Thus the local recovered guard is a **signed** positive
|
||||
loaded 32-bit value, not an unsigned nonzero test.
|
||||
|
||||
The release condition is one-byte `v100012bd9`, defined by
|
||||
`0xf0384:71b3 CPUI_INT_EQUAL`: input 0 is four-byte `v10001280a`, defined by
|
||||
`0xedbe5:7041 CPUI_LOAD`; input 1 is four-byte constant one `v3000393a3`.
|
||||
Its local guard is equality to one. Both load contents and aliases remain unknown.
|
||||
Models need not invent source member names for either memory object.
|
||||
|
||||
Derived controllers return precisely the two relations above for dependents
|
||||
1142/1158 in the actual retained-source replay. Coverage is complete for finite
|
||||
paths to recovered zero-successor exits; nontermination and runtime reachability
|
||||
remain unknown. These are direct/local control relations, not all transitive
|
||||
earlier guards or proof of feasible execution.
|
||||
|
||||
## Independent selected machine evidence
|
||||
|
||||
GNU objdump over the same ELF independently supplies these bytes/transfers;
|
||||
ELF file/virtual mappings also verify their raw bytes. The complete recovered
|
||||
graph, including these identities, deeply equals the earlier diagnostic native
|
||||
golden. That producer comparison is not independent machine/IR equivalence.
|
||||
|
||||
| Address | Bytes | Machine fact |
|
||||
| ------- | -------------- | ------------------------------- |
|
||||
| 0xedbe5 | 8b7004 | MOV ESI,[RAX+4] |
|
||||
| 0xedbec | 89742440 | MOV [RSP+0x40],ESI |
|
||||
| 0xedbf8 | 418b9dd8000000 | MOV EBX,[R13+0xd8] |
|
||||
| 0xedbff | 85db | TEST EBX,EBX |
|
||||
| 0xedc01 | 0f8f6b470000 | JG 0xf2372; fallthrough 0xedc07 |
|
||||
| 0xf0384 | 837c244001 | CMP dword [RSP+0x40],1 |
|
||||
| 0xf0389 | 0f849e200000 | JE 0xf242d; fallthrough 0xf038f |
|
||||
| 0xf2381 | 488d3578a10300 | LEA RSI,[0x12c500] |
|
||||
| 0xf238b | e83073f3ff | CALL 0x296c0 |
|
||||
| 0xf2437 | 488d35faa00300 | LEA RSI,[0x12c538] |
|
||||
| 0xf2446 | e940ffffff | JMP 0xf238b |
|
||||
|
||||
The earlier load at `0xedbe5` and spill at `0xedbec` correspond to the native
|
||||
value later compared at `0xf0384`; selected bytes do not prove all intervening
|
||||
memory/runtime behavior. Machine setup/branch windows and native optimized CFG
|
||||
edges are separate evidence representations. PHI is an IR operation at a shared
|
||||
origin, not a machine instruction.
|
||||
|
||||
Expected recovered coverage is now a full block-member graph: 21,349 operations,
|
||||
35,755 values, 1,229 blocks and zero omitted definition IDs. It includes 416
|
||||
recovered calls; 395 known CALL targets and 21 unresolved CALLIND targets. This
|
||||
does not establish full machine-flow/call coverage or memory alias resolution.
|
||||
The previous call-rooted projection lacked both CBRANCH/Boolean definitions and
|
||||
the open-path LOAD, as well as all CFG access; those absences remain recorded.
|
||||
|
||||
Grade the complete answer and actual trace, not keywords alone: both formats and
|
||||
shared call/PHI alternatives, both edges for each local condition, precise Boolean
|
||||
and memory widths/signedness, focused machine checks, real source IDs/digest,
|
||||
scope/earlier-guard/alias/nontermination/runtime limits, and retained-capture reuse.
|
||||
@@ -1,169 +0,0 @@
|
||||
# Real-agent savepoint control investigation
|
||||
|
||||
The [question and oracle](savepoint-control-oracle.md) were committed at
|
||||
`e2205313a` before either model launch. Both lanes have the same atomic facts,
|
||||
target, engine, question and common envelope/stopping guidance. They start with
|
||||
the function name and two messages, without supplied code addresses. This tests
|
||||
the newly available full recovered graph, not provider parity or superiority
|
||||
over a traditional RE interface.
|
||||
|
||||
## Initial pair: natural discovery
|
||||
|
||||
Both complete answers pass the frozen source/answer review:
|
||||
|
||||
- Incoming string references select `0xea6d0` from two same-name entries before
|
||||
one retained VM capture. Neither lane fetches the complete C report or graph.
|
||||
- Both format pointers merge through `0xf238b:ceb7` and reach CALL
|
||||
`0xf238b:7130`, with the correct predecessor/edge bindings and RSI storage.
|
||||
- Each answer supplies both successor edges for each nearest local conditional
|
||||
predecessor. The open guard is signed four-byte `0 < loaded32`; release is
|
||||
four-byte equality to one. Both trace one-byte Booleans to the right LOADs,
|
||||
preserve operand indexes/widths and stop before guessing loaded contents.
|
||||
- Focused instruction windows check loads, the spill, signed JG/equality JZ,
|
||||
format setups and shared machine CALL. Independent GNU/ELF verification checks
|
||||
eleven instruction sites against the same target bytes. PHI/edge identities
|
||||
remain separate from machine instructions and transfers.
|
||||
- Real source IDs and the target digest are present. Each answer limits itself
|
||||
to recovered graph/local guards and states machine-flow, alias,
|
||||
nontermination, earlier-path and runtime-feasibility limits. Neither uses
|
||||
derived controllers or claims a SQL trigger. Both close the lease.
|
||||
|
||||
Each answer cites ten observed source IDs. Both have one successful VM capture,
|
||||
whose recovered graph equals the complete producer golden; facts come from the
|
||||
preserved `677b9795...` engine. Full answers, events, tool context, traces and
|
||||
resource receipts are retained locally under
|
||||
`.cache/atomic-codemode/sqlite-savepoint-control-{direct,codemode}-v1*`.
|
||||
|
||||
| Whole-task measure | Direct atomic | Code Mode |
|
||||
| --------------------------------- | ------------: | --------: |
|
||||
| Cumulative input tokens | 515,024 | 610,715 |
|
||||
| Included cached input | 466,304 | 563,328 |
|
||||
| Uncached input | 48,720 | 47,387 |
|
||||
| Output tokens | 5,537 | 6,345 |
|
||||
| Included reasoning output | 1,344 | 1,502 |
|
||||
| Observed model responses | 17 | 18 |
|
||||
| MCP attempts, including rejection | 24 | 15 |
|
||||
| Successful atomic fact reads | 20 | 21 |
|
||||
| Model wall seconds | 206.269 | 223.636 |
|
||||
| Family wall seconds | 207.427 | 224.133 |
|
||||
| Family CPU seconds | 53.794 | 52.777 |
|
||||
| Kernel process-family peak MiB | 874.9 | 883.2 |
|
||||
|
||||
Code Mode's fifteen attempts include a rejected `timeout_ms:200000` request;
|
||||
the actual schema maximum is 120,000. It retries within that bound. The client
|
||||
also initially tries to JSON-parse the validation error's text and gets a
|
||||
SyntaxError. Both costs remain in the ledger. There are no recorded inner-API
|
||||
failures or resource aborts. Direct has no failed MCP attempt. MCP attempts,
|
||||
successful host reads, client execution calls and model responses are distinct
|
||||
counts; the trace has no handler entry for the schema rejection.
|
||||
|
||||
In this one sample, Code Mode reduces uncached input by about 2.7% but increases
|
||||
cumulative input by 18.6%, output by 14.6% and model wall time by 8.4%. Fewer MCP
|
||||
requests do not establish a whole-task advantage. The lanes also choose slightly
|
||||
different fact selections and output detail. This is an observational pair,
|
||||
not a controlled estimate of a single dispatch mechanism's causal effect.
|
||||
|
||||
The first client discovery call in both lanes applies a broad `rea.*...` regular
|
||||
expression to all tool names/descriptions. It matches unrelated tools, including
|
||||
creation tools, and prints large, truncated metadata. This is a real observed
|
||||
agent cost, not an analysis-provider failure; it also adds noise to comparing
|
||||
the two interfaces. A repeat should give both lanes the same exact-name discovery
|
||||
policy while retaining every advertised tool and charging that guidance to the
|
||||
whole task. No analysis steps, addresses or answers should be added.
|
||||
|
||||
Both runs request `gpt-6.1-sol/high` through Codex 0.161.0; the resolved service
|
||||
build is unknown. Per-response usage sums match the runner ledger. Cached input
|
||||
is already included in input, and reasoning is included in output. One CPU,
|
||||
4 GiB hard / 3 GiB early-stop limits apply. Kernel peak includes charged file
|
||||
cache; it is not RSS. Both runs have OOM 0, verified payload/profile cleanup,
|
||||
unchanged runtime fingerprints, engine and target digests. Each lease is fresh;
|
||||
OS/provider-service caches and remote compute costs are uncontrolled/unmeasured.
|
||||
|
||||
## Discovery-controlled repeat
|
||||
|
||||
Executed at `7a7c4a6d7`, Code Mode first, then direct. The repeat keeps the same
|
||||
question, oracle, native engine and all atomic methods, and uses the runner's existing
|
||||
`--enabled-tools` / `--guide-enabled-tools` options with server name `rea`.
|
||||
Each manifest lists that lane's complete catalog: seventeen direct tools or four
|
||||
Code Mode tools. No analysis method is removed. Captured first client calls
|
||||
select these exact function names; neither repeats the broad metadata spill.
|
||||
This changes discovery guidance and namespace together for both lanes; it is a
|
||||
separate stratum from the natural-discovery pair, not a retroactive replacement.
|
||||
|
||||
Both complete answers again pass the same oracle: correct format ownership,
|
||||
PHI/call and native edges, signed-positive versus equality guards, Boolean/load
|
||||
widths and operand indexes, independent selected machine bytes, real provenance
|
||||
and explicit static/alias/runtime limits. Both use one retained VM capture and
|
||||
close the lease. Direct cites ten observed source IDs; Code Mode cites nine,
|
||||
combining some instruction windows. All recovered graph and ordinary function
|
||||
fields equal the earlier full golden in both lanes. Neither fetches a whole C
|
||||
report or graph. Neither uses derived controllers; both keep local guards
|
||||
separate from all earlier conditions and runtime feasibility.
|
||||
|
||||
| Whole-task measure | Direct atomic | Code Mode |
|
||||
| ------------------------------ | ------------: | --------: |
|
||||
| Cumulative input tokens | 538,013 | 608,622 |
|
||||
| Included cached input | 457,088 | 530,688 |
|
||||
| Uncached input | 80,925 | 77,934 |
|
||||
| Output tokens | 5,150 | 5,989 |
|
||||
| Included reasoning output | 697 | 1,462 |
|
||||
| Observed model responses | 19 | 25 |
|
||||
| MCP attempts | 24 | 13 |
|
||||
| Successful atomic fact reads | 20 | 18 |
|
||||
| Model wall seconds | 187.544 | 266.141 |
|
||||
| Family wall seconds | 188.086 | 266.650 |
|
||||
| Family CPU seconds | 52.113 | 102.739 |
|
||||
| Kernel process-family peak MiB | 859.1 | 883.1 |
|
||||
| VM source-capture wall seconds | 37.026 | 107.804 |
|
||||
|
||||
There are no failed MCP attempts, recorded inner-API failures or resource aborts.
|
||||
Per-response usage sums agree; OOM is zero, cleanup is verified, and target,
|
||||
engine and runtime fingerprints remain unchanged. Full records use suffix `-v2`
|
||||
under the same local artifact directory. Request/model settings and resource
|
||||
bounds are unchanged from the first pair.
|
||||
|
||||
Code Mode again has slightly lower uncached input (3.7%) and fewer MCP requests,
|
||||
but more cumulative input, output, model responses, wall time and family CPU in
|
||||
this sample. It has twelve client waits versus three for direct. Discovery
|
||||
guidance alone has not established an overall improvement across strata.
|
||||
|
||||
The VM capture is a substantial confound: its 107.804 seconds differs from
|
||||
37.026 in direct and 38.013/38.871 in the first pair. Source elapsed time includes
|
||||
command supervision and retention; it is not an isolated native drive interval.
|
||||
No per-command CPU or phase diagnostic was captured during these model trials.
|
||||
The additional instruction-window API call overlapped the pending Code Mode
|
||||
snapshot; its own wall time also includes waiting beyond the retained native
|
||||
source times. Whether concurrency, host contention or another lower boundary
|
||||
caused the variation is unresolved. CPU affinity/quota do not give an exclusive
|
||||
core. Do not subtract these intervals from whole-task time to manufacture a
|
||||
Code Mode speedup or attribute the difference to dispatch alone.
|
||||
|
||||
Both lanes actually use the client's `exec`/`wait` JavaScript surface, despite
|
||||
the recorded client Code Mode feature flags being false. Direct atomic already
|
||||
composes MCP calls through that client executor. These results compare a server
|
||||
Code Mode boundary against directly called atomic MCP tools in that environment;
|
||||
they do not compare JavaScript composition against its absence. Tool counts and
|
||||
configured feature flags cannot establish the client's actual available surface.
|
||||
|
||||
## Remaining questions
|
||||
|
||||
Four real runs establish this scoped control/data-flow task on the full recovered
|
||||
graph, without a stable overall efficiency win. There is no traditional-provider
|
||||
cell for this graph-identity question and no general complex-software superiority
|
||||
claim. Whole-task cumulative input across all four runs is 2,272,374 tokens,
|
||||
including 2,017,408 cached and 254,966 uncached; output is 23,021 including 5,005
|
||||
reasoning tokens. Total model wall is 883.590 seconds. These totals retain the
|
||||
initial failed request and discovery costs, not a pooled treatment estimate.
|
||||
|
||||
The retained graph also exposes the next coverage question: `0xea801:fe` is one
|
||||
`CPUI_BRANCHIND` in block 8 with 149 recovered successors. Current edges expose
|
||||
identities/destinations but no case values. The native jump-table owner should
|
||||
be examined before exporting case-to-edge/target relations; a successor ordinal
|
||||
is not a case value. This is a fact-export limitation, not evidence that Kuna
|
||||
itself cannot recover the labels. The source-capture/concurrent-read cost
|
||||
variation also needs a bounded, matched diagnostic before further speed claims.
|
||||
The [subsequent four-case public MCP diagnostic](overlap-study.md) confirms
|
||||
provider admission is serialized and does not reproduce the long capture;
|
||||
the original cause remains unknown rather than being declared fixed.
|
||||
The [native switch extension](switch-study.md) now exports the case labels and
|
||||
checks all 182 original targets independently; its new model question is separate.
|
||||
@@ -1,127 +0,0 @@
|
||||
// Complete experimental catalog; it does not truncate the production REA catalog.
|
||||
import { McpServer } from "@modelcontextprotocol/server";
|
||||
import { serveStdio } from "@modelcontextprotocol/server/stdio";
|
||||
import { z } from "zod";
|
||||
import { AtomicSession, methods, reference } from "./api.mjs";
|
||||
import { executeJavaScript } from "./executor.mjs";
|
||||
import { referenceInput } from "./reference.mjs";
|
||||
import { TARGET_CONTENT_RESULT_META } from "../../dist/server/toolResultEncoding.js";
|
||||
import { appendFile } from "node:fs/promises";
|
||||
|
||||
const DIRECT_REFERENCE_LANGUAGE =
|
||||
"Direct MCP calls: use the displayed facts_* tools with the same JSON arguments; results have {ok,context,data}.";
|
||||
const session = new AtomicSession();
|
||||
const createServer = () => {
|
||||
const server = new McpServer({ name: "rea-codemode", version: "0.1.0" });
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const apiNames = Object.fromEntries(
|
||||
Object.keys(methods).map((name) => [
|
||||
name,
|
||||
direct ? `facts_${name}` : `rea.${name}`,
|
||||
]),
|
||||
);
|
||||
const encode = (value) => ({
|
||||
content: [{ type: "text", text: JSON.stringify(value) }],
|
||||
_meta: TARGET_CONTENT_RESULT_META,
|
||||
});
|
||||
const register = (name, description, inputSchema, handler) =>
|
||||
server.registerTool(
|
||||
name,
|
||||
{ description, inputSchema, annotations: { readOnlyHint: true } },
|
||||
async (input, request) => {
|
||||
try {
|
||||
const result = await handler(input, request.mcpReq.signal);
|
||||
if (process.env.REA_ATOMIC_TRACE)
|
||||
await appendFile(
|
||||
process.env.REA_ATOMIC_TRACE,
|
||||
JSON.stringify({ tool: name, input, result }) + "\n",
|
||||
{ mode: 0o600 },
|
||||
);
|
||||
if (name === "execute_javascript") {
|
||||
const { calls, ...answer } = result;
|
||||
return encode({
|
||||
...answer,
|
||||
execution: { host_calls: calls?.length ?? 0 },
|
||||
authority: "script-derived",
|
||||
context: answer.context ?? null,
|
||||
});
|
||||
}
|
||||
return encode(result);
|
||||
} catch (error) {
|
||||
return encode({
|
||||
ok: false,
|
||||
context: session.context,
|
||||
error: {
|
||||
code: error.code ?? "analysis_failed",
|
||||
message: error.message,
|
||||
details: error.details ?? {},
|
||||
},
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
register(
|
||||
"open_binary",
|
||||
"Open one local binary for static Kuna analysis. Returns artifact/engine identity and a target lease. Does not execute or modify the target.",
|
||||
z.strictObject({ path: z.string().min(1) }),
|
||||
(input) => session.open(input.path),
|
||||
);
|
||||
register(
|
||||
"close_binary",
|
||||
"Close this target lease and owned native resources; retained snapshots and Evidence expire.",
|
||||
z.strictObject({}),
|
||||
() => session.close(),
|
||||
);
|
||||
register(
|
||||
"reference",
|
||||
"Discover fact methods with view:index and an empty query: returns names and descriptions without schemas. Then request only needed full contracts with view:contracts and exact displayed names or canonical IDs separated by spaces/commas. One topic query searches descriptions; empty query with view:contracts returns all contracts. These are static declared methods, not live provider availability.",
|
||||
referenceInput,
|
||||
(input) =>
|
||||
reference(input.query, {
|
||||
names: apiNames,
|
||||
view: input.view,
|
||||
includeExamples: !direct,
|
||||
language: direct ? DIRECT_REFERENCE_LANGUAGE : undefined,
|
||||
}),
|
||||
);
|
||||
if (direct) {
|
||||
for (const [name, contract] of Object.entries(methods))
|
||||
register(
|
||||
apiNames[name],
|
||||
contract.description,
|
||||
contract.input,
|
||||
(input, signal) => session.call(name, input, signal),
|
||||
);
|
||||
} else {
|
||||
register(
|
||||
"execute_javascript",
|
||||
"Execute a JavaScript async function body against the open binary's atomic rea API. Use reference first. Await rea methods, compose/filter locally, and return a small JSON answer with evidence IDs and unknowns. No Node/filesystem/network APIs. Fresh JS environment per call; analysis facts are reused across calls.",
|
||||
z.strictObject({
|
||||
code: z.string().min(1).max(32768),
|
||||
timeout_ms: z.number().int().min(100).max(120000).default(120000),
|
||||
}),
|
||||
(input, signal) =>
|
||||
executeJavaScript(session, input.code, {
|
||||
signal,
|
||||
timeoutMs: input.timeout_ms,
|
||||
}),
|
||||
);
|
||||
}
|
||||
return server;
|
||||
};
|
||||
const handle = serveStdio(createServer, {
|
||||
onerror: (error) => process.stderr.write(`${error.message}\n`),
|
||||
});
|
||||
let closing;
|
||||
const shutdown = () =>
|
||||
(closing ??= (async () => {
|
||||
session.abort();
|
||||
await session.close();
|
||||
await handle.close();
|
||||
})().catch((error) => {
|
||||
process.stderr.write(`${error.message}\n`);
|
||||
process.exitCode = 1;
|
||||
}));
|
||||
process.once("SIGTERM", shutdown);
|
||||
process.once("SIGINT", shutdown);
|
||||
process.stdin.once("end", shutdown);
|
||||
@@ -1,17 +0,0 @@
|
||||
Analyze sqlite3VdbeExec in {{artifact_path}} statically. Use one retained function
|
||||
capture to inspect recovered call sites and their boundary values even if the
|
||||
complete recovered IR graph is unavailable. Report each facet's availability and
|
||||
the scope of any retained value set; do not treat it as a complete function graph.
|
||||
|
||||
Aggregate the retained recovered calls by CALL versus CALLIND and known versus
|
||||
unknown target. At call origins 0xf115c and 0xf238b, report the resolved target,
|
||||
the input-zero value identity and its annotation/address-space facts, recovered
|
||||
argument slots and passing locations. Determine whether the first argument is
|
||||
the same retained SSA value at both sites and whether it is a function input.
|
||||
Verify the selected machine CALL targets and essential argument-register setup
|
||||
with focused instruction windows. Distinguish internal fspec offsets, SSA value
|
||||
locations, passing storage, and code addresses.
|
||||
|
||||
Explain which definition/use or alias claims remain unavailable. Reuse the same
|
||||
capture for selected reads rather than reanalyzing the VM. Keep the answer focused
|
||||
and give real Evidence references, target digest and static/runtime limitations.
|
||||
@@ -1,15 +0,0 @@
|
||||
Analyze {{artifact_path}} statically. In sqlite3Fts3ExprParse, explain how a
|
||||
"malformed MATCH expression" error can be selected even when its parsing helper
|
||||
returns success. Trace the relevant caller-owned context into the helper: identify
|
||||
the call site, actual callee entry, the passed SSA value and passing location, and
|
||||
the corresponding recovered function input. Explain which context field the helper
|
||||
updates for opening/closing parentheses, where the caller checks that same field,
|
||||
and how the successful and malformed paths differ. Verify the essential pointer,
|
||||
field and control-transfer claims against selected machine instructions.
|
||||
In the final answer, explicitly distinguish the wrapper's routing for helper
|
||||
return codes 0, 1 and 18, including the malformed and depth-limit diagnostics.
|
||||
|
||||
Keep the answer focused on this cross-function state flow. Give function and
|
||||
instruction addresses and real Evidence references. Distinguish recovered call
|
||||
slots from original source parameter ordinals, exact storage matches from proven
|
||||
alias analysis, and local static conditions from demonstrated runtime feasibility.
|
||||
@@ -1,16 +0,0 @@
|
||||
Use one retained static capture of sqlite3VdbeExec at 0xea6d0 in {{artifact_path}}.
|
||||
Investigate recovered operation 0xea998:14649 as a wide SSA merge. Report its
|
||||
opcode, output value, block and total input count. Retrieve only input indices
|
||||
0, 1, 77, 105 and 155, retaining each value ID, native predecessor and exact
|
||||
predecessor-edge identity. Read the distinct selected values' address spaces,
|
||||
offsets, widths and defining operation identities; inspect those immediate
|
||||
defining operations' opcode and output. Explain which selected slots share a
|
||||
value, without discarding their distinct edge identities.
|
||||
|
||||
Check requested slot 156 and operation 0xffff:0 separately: distinguish an absent
|
||||
slot of a retained operation from unknown membership of an unretained operation.
|
||||
Use selected/batch/paged facts and reuse the same capture. Do not return or page
|
||||
through the whole wide input list or reconstructed C. State the available
|
||||
operation scope and remaining definition-frontier, CFG, complete-use, alias and
|
||||
runtime-feasibility limits. Keep the answer focused with actual Evidence IDs and
|
||||
target digest; optimized PHI predecessors and origins are not machine transfers.
|
||||
@@ -1,9 +0,0 @@
|
||||
Analyze {{artifact_path}} statically. Find where the diagnostic
|
||||
"Safety level may not be changed inside a transaction" is selected.
|
||||
Explain the local conditions, distinguishing reading the setting from attempting
|
||||
to change it. Identify the nearby state-byte updates when a change is allowed,
|
||||
the diagnostic call, and the cleanup/return behavior. Do not assume source-level
|
||||
field names or a numeric function return code without evidence.
|
||||
Give supporting function/instruction addresses and real Evidence references.
|
||||
Keep the answer focused on this diagnostic. Preserve unavailable-facet reasons
|
||||
and distinguish local static conditions from demonstrated runtime feasibility.
|
||||
@@ -1,20 +0,0 @@
|
||||
Analyze sqlite3VdbeExec in {{artifact_path}} statically. At recovered call origins
|
||||
0xf115c and 0xf238b, trace the second argument (passed in RSI) through its immediate
|
||||
SSA definition and every PHI alternative to the retained format-address
|
||||
producers. Use one retained capture and structured operation/value relationships.
|
||||
Report the call and PHI identities, each PHI operand index and native predecessor
|
||||
edge identity, and which format-address producer belongs to each alternative.
|
||||
Inspect the available operation scope before traversing: a bounded definition
|
||||
projection is not a complete recovered CFG or full use inventory.
|
||||
|
||||
Read the candidate format strings with focused byte or string queries. Verify
|
||||
all alternative setup sites and the two machine CALL targets with selected
|
||||
instruction windows. Include actual jump destinations used by alternate setup
|
||||
paths, where present. Distinguish native PHI predecessors, instruction origins,
|
||||
constant address-space offsets and actual machine branch targets; a single
|
||||
instruction window does not establish the only format reaching a shared call.
|
||||
|
||||
Keep the result focused, with real Evidence references and target digest. Reuse
|
||||
the same capture for selected reads. State any remaining definition frontier,
|
||||
memory-alias, CFG and runtime-reachability limits; do not claim a demonstrated
|
||||
SQL-level trigger or infer complete source signatures from recovered input slots.
|
||||
@@ -1,6 +0,0 @@
|
||||
Analyze {{artifact_path}} statically. Find where the "malformed MATCH expression"
|
||||
diagnostic is produced. Explain the local condition that selects this diagnostic,
|
||||
how the diagnostic value is constructed, and which return/error paths are distinct.
|
||||
Give supporting function and instruction addresses and real Evidence references.
|
||||
Distinguish local static conditions from demonstrated runtime feasibility, and
|
||||
keep the answer focused on this diagnostic rather than the entire parser.
|
||||
@@ -1,15 +0,0 @@
|
||||
Audit static call references in {{artifact_path}}, returning aggregate results
|
||||
with focused supporting addresses and real Evidence references.
|
||||
|
||||
For sqlite3Fts3ExprParse, compare the native outgoing known-target call-reference
|
||||
count with the retained recovered-IR call-operation count. Explain any difference
|
||||
and verify the differing call against selected machine instructions; absence from
|
||||
optimized IR must not be treated as absence from the executable.
|
||||
|
||||
For sqlite3VdbeExec, report the number of selected native known-target call-site
|
||||
references, the number of distinct exact destination addresses, and the number
|
||||
that target 0x296c0. Verify the call site at 0xf238b. This part concerns references
|
||||
only: do not decompile the VM or return its instruction listing/source report.
|
||||
Compose selected pages locally and return the requested aggregates rather than
|
||||
all reference rows. State the limits for unresolved indirect calls, discovery,
|
||||
and runtime symbol binding. Do not execute the target.
|
||||
@@ -1,30 +0,0 @@
|
||||
Investigate sqlite3VdbeExec in {{artifact_path}} statically, starting from these
|
||||
observed error formats:
|
||||
- cannot open savepoint - SQL statements in progress
|
||||
- cannot release savepoint - SQL statements in progress
|
||||
|
||||
Locate both retained format sources and determine whether they reach the same
|
||||
recovered error call, including how their values merge. Names can identify
|
||||
multiple callable entries; establish the format-owning implementation from
|
||||
references or byte evidence before deep analysis. Reuse one retained VM capture
|
||||
and selected structured facts rather than fetching the whole C report or graph.
|
||||
|
||||
For each format-setup block, find the nearest recovered conditional predecessor
|
||||
controlling entry to that setup. Report both successor edges, their identities,
|
||||
polarity and condition value/test. Trace each condition through its Boolean
|
||||
definition to constants and LOAD boundaries, preserving operand indexes and
|
||||
widths. Stop at those memory reads; do not guess source field names or arbitrary
|
||||
runtime contents. Distinct SSA values, storage offsets and computed contents are
|
||||
different facts.
|
||||
|
||||
Verify the selected setup sites, shared machine CALL, conditional branch targets
|
||||
and relevant load/spill instructions with focused instruction windows. Explain
|
||||
the static local guards, including comparison signedness, without claiming these
|
||||
are all earlier path conditions or a demonstrated SQL trigger. Distinguish native
|
||||
edge/PHI identities from machine instruction addresses and transfers.
|
||||
|
||||
Keep the answer focused, with real Evidence IDs and target digest. State the
|
||||
available recovered-graph scope and remaining alias, complete-machine-flow,
|
||||
nontermination and runtime-feasibility limits. If using derived controllers,
|
||||
retain their finite-recovered-exit-path semantics. Complete retained Evidence
|
||||
has a paged source/decoded-graph view; return only the facts needed here.
|
||||
@@ -1,31 +0,0 @@
|
||||
Investigate the format-owning sqlite3VdbeExec implementation in {{artifact_path}},
|
||||
starting from its callable name and these observed formats:
|
||||
- cannot open savepoint - SQL statements in progress
|
||||
- cannot release savepoint - SQL statements in progress
|
||||
|
||||
Establish the implementation from references before deep analysis; a name may
|
||||
identify several entries. Reuse one retained VM capture and selected structured
|
||||
facts. Find its recovered indirect dispatch and distinguish the original byte
|
||||
selector from the width/identity of the retained normalized selector SSA value
|
||||
and from the machine's computed jump address.
|
||||
|
||||
For case values 0x0, 0x2a, 0x34, 0x39 and 0xb5, report the original table-row
|
||||
identity, native case-to-edge identity/destination, and original recovered target.
|
||||
Check whether 0x34 and 0x39 share a target and edge. Do not infer labels from edge
|
||||
ordinals or optimized block origins. Query only the selected cases; do not fetch
|
||||
the complete case inventory, C report or graph.
|
||||
|
||||
Use focused instructions to establish the machine dispatch: input width,
|
||||
unsigned range guard, table base, entry width/signedness and relative target
|
||||
calculation. Read the selected table bytes in a batch and derive the selected
|
||||
machine targets from those bytes, comparing with the retained native facts.
|
||||
Determine what the selected byte 0xb6 would do at that local guard, whether it
|
||||
has a captured explicit table case, and whether the native model's default
|
||||
out-edge is the same as the machine out-of-range guard destination. Keep those
|
||||
defaults, representations and identities separate.
|
||||
|
||||
Return a focused answer with real Evidence IDs, target digest and necessary
|
||||
coverage. Native full-graph/table recovery does not establish complete machine
|
||||
flow, aliases, nontermination or feasible runtime inputs. Do not invent source
|
||||
member/opcode names, arbitrary loaded contents, a SQL trigger or runtime binding.
|
||||
Complete retained Evidence has paged source/decoded-graph access.
|
||||
@@ -1,10 +0,0 @@
|
||||
Analyze {{artifact_path}} statically. In sqlite3GetToken, recover the initial
|
||||
byte-classification lookup and the first indirect dispatch it controls.
|
||||
For first bytes 0x2d, 0x2f, 0x27, 0x22, 0x3f and 0x80, report the actual stored
|
||||
class value and the initial machine dispatch destination. Read the artifact's
|
||||
data tables; do not assume the byte value itself is a switch-case number.
|
||||
Explain table bases, indexing, displacement width/signedness and target-base
|
||||
calculation with supporting instruction addresses and real Evidence references.
|
||||
Keep this about initial dispatch, not complete token recognition. Distinguish
|
||||
static loaded-image observations from demonstrated runtime feasibility, and do
|
||||
not infer final token types or consumed lengths from initial classes alone.
|
||||
@@ -1,10 +0,0 @@
|
||||
Analyze {{artifact_path}} statically. Find where these two diagnostics are
|
||||
selected: "cannot open savepoint - SQL statements in progress" and
|
||||
"cannot release savepoint - SQL statements in progress".
|
||||
Explain the local conditions that distinguish them, how their message selection
|
||||
and error handling join, and how a missing named savepoint differs. Give the
|
||||
supporting function/instruction addresses and relevant structure offsets.
|
||||
Keep the investigation focused on these diagnostics rather than the whole VM.
|
||||
Use real Evidence references and distinguish local static conditions from
|
||||
demonstrated runtime feasibility. If an analysis facet is unavailable, state
|
||||
what remains supported and what remains unknown; do not invent missing facts.
|
||||
@@ -1,24 +0,0 @@
|
||||
import { parseKunaStringInventory } from "../../dist/kuna/KunaStringInventory.js";
|
||||
import { requireResult } from "./fact-error.mjs";
|
||||
import { freezeFacts, pageFacts } from "./fact-reads.mjs";
|
||||
|
||||
export const parseStringFacts = (text, input) =>
|
||||
freezeFacts(
|
||||
requireResult(
|
||||
parseKunaStringInventory(text, input),
|
||||
"invalid_provider_output",
|
||||
),
|
||||
);
|
||||
|
||||
export const selectStringFacts = (inventory, input) => ({
|
||||
...pageFacts(
|
||||
inventory.items.filter((row) => row.value.includes(input.query)),
|
||||
input,
|
||||
),
|
||||
coverage: {
|
||||
scope: "native string inventory, minimum 5",
|
||||
scanned: inventory.scanned,
|
||||
artifact_discovery: "unknown",
|
||||
reference_attribution: "not-computed",
|
||||
},
|
||||
});
|
||||
@@ -1,141 +0,0 @@
|
||||
"""Summarize actual model usage and owned process-family counters, without grading answers."""
|
||||
|
||||
import argparse
|
||||
import collections
|
||||
import json
|
||||
from pathlib import Path
|
||||
|
||||
|
||||
def client_observations(directory, declared_usage):
|
||||
path = directory / 'codex-tool-context.jsonl'
|
||||
receipt_path = directory / 'context-receipt.json'
|
||||
if not path.exists() or not receipt_path.exists():
|
||||
return {'availability': 'unavailable', 'model_responses': None,
|
||||
'tool_calls': None, 'per_response_usage_matches_summary': None}
|
||||
receipt = json.loads(receipt_path.read_text())
|
||||
responses, calls = {}, {}
|
||||
for line in path.read_text().splitlines():
|
||||
event = json.loads(line)
|
||||
payload = event['payload']
|
||||
if event['type'] == 'token_usage_record':
|
||||
key, value = payload['response_id'], payload['usage']
|
||||
if key in responses and responses[key] != value:
|
||||
raise ValueError('Conflicting usage for the same model response')
|
||||
responses[key] = value
|
||||
elif event['type'] == 'response_item' and payload['type'] in (
|
||||
'custom_tool_call', 'function_call'):
|
||||
key = payload['call_id']
|
||||
if key in calls and calls[key] != payload:
|
||||
raise ValueError('Conflicting records for the same client call')
|
||||
calls[key] = payload
|
||||
totals = collections.Counter()
|
||||
for usage in responses.values():
|
||||
totals.update(usage)
|
||||
complete = receipt['status'] == 'captured'
|
||||
matches = (all(totals[key] == value for key, value in declared_usage.items())
|
||||
if complete and responses else None)
|
||||
if matches is False:
|
||||
raise ValueError('Per-response usage differs from the runner summary')
|
||||
return {
|
||||
'availability': receipt['status'],
|
||||
'model_responses': len(responses) if responses else None,
|
||||
'tool_calls': dict(collections.Counter(value['name'] for value in calls.values())),
|
||||
'per_response_usage_matches_summary': matches,
|
||||
'scope': 'Observed unique model responses and client dispatch/wait calls. '
|
||||
'Configured Code Mode feature flags do not establish the actual '
|
||||
'client tool surface. MCP requests are not model response rounds.',
|
||||
}
|
||||
|
||||
|
||||
def trace_observations(path):
|
||||
"""Transport completion is separate from a failed analysis/API response."""
|
||||
if not path.exists():
|
||||
return {'availability': 'unavailable', 'code_mode_calls': None,
|
||||
'code_mode_failures': None, 'direct_fact_calls': None,
|
||||
'api_failures': None, 'code_mode_evidence_count': None}
|
||||
methods, direct = collections.Counter(), collections.Counter()
|
||||
inner_failures, failures, source_ids = [], [], set()
|
||||
for line in path.read_text().splitlines():
|
||||
entry = json.loads(line)
|
||||
result, tool = entry['result'], entry['tool']
|
||||
if tool == 'execute_javascript':
|
||||
for call in result.get('calls', []):
|
||||
methods[call['method']] += 1
|
||||
if call['status'] != 'ok':
|
||||
inner_failures.append(call)
|
||||
failures.append({'tool': tool, 'scope': 'inner-api', **call})
|
||||
source_ids.update(result.get('evidence_ids', []))
|
||||
elif tool.startswith('facts_'):
|
||||
direct[tool] += 1
|
||||
if result.get('ok') is False:
|
||||
failures.append({'tool': tool, 'scope': 'mcp-result',
|
||||
'error': result.get('error')})
|
||||
return {'availability': 'captured', 'code_mode_calls': dict(methods),
|
||||
'code_mode_failures': inner_failures, 'direct_fact_calls': dict(direct),
|
||||
'api_failures': failures, 'code_mode_evidence_count': len(source_ids)}
|
||||
|
||||
|
||||
def summarize(directory):
|
||||
summary = json.loads((directory / 'summary.json').read_text())
|
||||
resource_records = [json.loads(line) for line in directory.with_suffix('.log').read_text().splitlines()
|
||||
if line.startswith('{"resource_scope":')]
|
||||
if len(resource_records) != 1:
|
||||
raise ValueError(f'{directory}: expected one process-family resource receipt')
|
||||
resources = resource_records[0]
|
||||
observation = resources['payload_resource_observation']
|
||||
if observation['phase'] != 'payload-exited-observer-alive':
|
||||
raise ValueError(f'{directory}: payload-exit counters were not observed')
|
||||
usage = collections.Counter()
|
||||
for turn in summary['usage']:
|
||||
usage.update(turn)
|
||||
client = client_observations(directory, usage)
|
||||
calls = summary['completed_mcp_calls'] # This runner key also retains failed attempts.
|
||||
statuses = collections.Counter(call['status'] for call in calls)
|
||||
trace = trace_observations(directory.with_suffix('.trace.jsonl'))
|
||||
usage['uncached_input_tokens'] = usage['input_tokens'] - usage['cached_input_tokens']
|
||||
cleanup = json.loads((directory / 'cleanup.json').read_text())
|
||||
return {
|
||||
'trial': directory.name,
|
||||
'runtime_revision': summary['runtime_revision'],
|
||||
'runtime_inputs_unchanged': summary.get('runtime_inputs_unchanged'),
|
||||
'experimental_engine': summary.get('experimental_engine'),
|
||||
'experimental_engine_unchanged': summary.get('experimental_engine_unchanged'),
|
||||
'requested_model': summary['requested_model'],
|
||||
'reasoning': summary['reasoning'],
|
||||
'usage': dict(usage),
|
||||
'client_observations': client,
|
||||
'outer_attempts': len(calls),
|
||||
'outer_attempt_scope': 'MCP requests, not model responses or client executor calls',
|
||||
'outer_statuses': dict(statuses),
|
||||
'outer_failures': [{k: call[k] for k in ('tool', 'status', 'error')}
|
||||
for call in calls if call['status'] != 'completed'],
|
||||
'trace_availability': trace['availability'],
|
||||
'inner_code_mode_calls': trace['code_mode_calls'],
|
||||
'inner_api_failures': trace['code_mode_failures'],
|
||||
'direct_fact_calls': trace['direct_fact_calls'],
|
||||
'api_failures': trace['api_failures'],
|
||||
'code_mode_evidence_count': trace['code_mode_evidence_count'],
|
||||
'model_wall_seconds': summary['wall_seconds'],
|
||||
'family_wall_seconds': resources['elapsed_seconds'],
|
||||
'family_cpu_seconds': observation['cpu_stat']['usage_usec'] / 1_000_000,
|
||||
'family_memory_peak_bytes': observation['memory_peak_bytes'],
|
||||
'memory_events': observation['resource_events']['memory.events'],
|
||||
'resource_abort': resources['abort'],
|
||||
'payload_cleanup_verified': observation['payload_cleanup_verified'],
|
||||
'profile_cleanup': cleanup,
|
||||
'target_sha256': summary['selected_artifact_sha256_before'],
|
||||
'target_unchanged': summary['selected_artifact_bytes_unchanged'],
|
||||
'correctness': 'requires separate answer/source-oracle review',
|
||||
'scope': 'Actual cumulative usage includes discovery/retries; cached input is included in input, reasoning in output. Memory is kernel cgroup peak including charged file cache. Server-side compute and resolved model build are unmeasured.',
|
||||
}
|
||||
|
||||
|
||||
def main():
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument('directories', type=Path, nargs='+')
|
||||
options = parser.parse_args()
|
||||
print(json.dumps([summarize(path) for path in options.directories], indent=2))
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
main()
|
||||
@@ -1,76 +0,0 @@
|
||||
# SQLite indirect-dispatch oracle
|
||||
|
||||
Freeze this question/oracle before measured model launches. The question supplies
|
||||
a function name, two formats and selected case values, without code/table/graph
|
||||
addresses. This is a scoped investigation. Both atomic adapters share facts;
|
||||
no traditional-provider cell exists for its required native identities. The
|
||||
direct client also has its own JavaScript executor.
|
||||
|
||||
Target: Debian SQLite `3.40.1-2+deb12u2` amd64,
|
||||
`/usr/lib/x86_64-linux-gnu/libsqlite3.so.0` -> `.0.8.6`, SHA-256
|
||||
`2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8`.
|
||||
Final engine SHA-256:
|
||||
`a9ca8a8ac224936afb08a9d4be9c1225feb69473ad9f35764e8d1c4f04875b5b`.
|
||||
|
||||
Callable entries `0x288f0` and `0xea6d0` share the VDBE name; the two format
|
||||
references establish owning entry `0xea6d0` before deep analysis. Names do not
|
||||
establish runtime binding. One snapshot has 21,349 recovered operations, 35,755
|
||||
values, 1,229 blocks, zero omitted definitions and one indirect branch:
|
||||
`0xea801:fe`, block 8. The switch facet has 182 labelled rows and mappings,
|
||||
149 distinct targets/edges, partial=false, and zero missing/no-label/unmapped/
|
||||
unknown-target rows. Consumed mask is `0xff`; signed_render_hint=false is not
|
||||
itself an unsignedness proof.
|
||||
|
||||
The current selector is eight-byte `v100031675` at `register:0x10`, defined by
|
||||
`0xea7f6:c353 CPUI_INT_ZEXT`. Its one-byte input `v10000024f` is defined by LOAD
|
||||
`0xea7e1:de`. Those widths/definitions are verified in structured native facts.
|
||||
The machine reads one byte, indexes through R15B/EDX, computes a target in RAX
|
||||
and jumps through RAX. The normalized BRANCHIND input is not that jump address.
|
||||
Storage offsets do not specify arbitrary computed contents.
|
||||
|
||||
Independent GNU disassembly and ELF program-header/file reads establish:
|
||||
|
||||
| Address | Bytes | Machine fact |
|
||||
| ------- | -------------- | ------------------------------- |
|
||||
| 0xea7e1 | 440fb638 | MOVZX R15D,byte [RAX] |
|
||||
| 0xea7e5 | 4180ffb5 | CMP R15B,0xb5 |
|
||||
| 0xea7e9 | 0f87a9010000 | JA 0xea998; fallthrough 0xea7ef |
|
||||
| 0xea7ef | 488d0d121f0400 | LEA RCX,[0x12c708] |
|
||||
| 0xea7f6 | 410fb6d7 | MOVZX EDX,R15B |
|
||||
| 0xea7fa | 48630491 | MOVSXD RAX,dword [RCX+RDX*4] |
|
||||
| 0xea7fe | 4801c8 | ADD RAX,RCX |
|
||||
| 0xea801 | ffe0 | JMP RAX |
|
||||
|
||||
Table base is `0x12c708`: 182 entries for zero-extended byte values `0..0xb5`,
|
||||
each a signed little-endian 32-bit displacement added to the base. Local
|
||||
`.cache/atomic-codemode/sqlite-switch-machine-oracle.json` records all bytes,
|
||||
file offsets and 149 unique targets, reconstructed before native verification.
|
||||
Both MCP paths compare all 182 native labelled rows to that independent table.
|
||||
|
||||
| Case / row | Table address | Raw bytes / signed displacement | Native edge -> block | Original target |
|
||||
| ---------- | ------------- | ------------------------------- | -------------------- | --------------- |
|
||||
| 0x0 / 0 | 0x12c708 | d814fcff / -256808 | 8:0 -> 1140 | 0xedbe0 |
|
||||
| 0x2a / 42 | 0x12c7b0 | a8e7fbff / -268376 | 8:34 -> 864 | 0xeaeb0 |
|
||||
| 0x34 / 52 | 0x12c7d8 | e8e5fbff / -268824 | 8:42 -> 778 | 0xeacf0 |
|
||||
| 0x39 / 57 | 0x12c7ec | e8e5fbff / -268824 | 8:42 -> 778 | 0xeacf0 |
|
||||
| 0xb5 / 181 | 0x12c9dc | 16eefbff / -266730 | 8:8 -> 1011 | 0xeb51e |
|
||||
|
||||
The middle pair shares an edge/target but keeps distinct case identities.
|
||||
Row and edge ordinals differ: row 181 maps to edge 8. Relations carry case_id
|
||||
and a separate relation id; multiple edges share case_id and page separately.
|
||||
|
||||
Native model default: `8:42`, block 778, origin `0xeacf0`, raw default index 42,
|
||||
default_is_folded=false. Known values `0x34..0x39` share it. It is not the
|
||||
machine out-of-range guard destination `0xea998`. No explicit native row is
|
||||
labelled `0xb6`. If the selected local byte were `0xb6`, unsigned JA bypasses
|
||||
the table and reaches `0xea998`; this does not prove that such input occurs.
|
||||
|
||||
Grade the complete answer and trace: correct ownership before one VM capture,
|
||||
selected case reads, source-byte versus normalized-selector widths, row/edge
|
||||
identities and shared targets, selected raw displacement arithmetic, the two
|
||||
different defaults, real source IDs/digest and static coverage limitations.
|
||||
No whole case inventory/C report/graph fetch, source field/opcode-name guesses,
|
||||
SQL trigger or runtime binding claim. Full recovery does not prove complete
|
||||
machine flow, memory aliases, nontermination or feasible inputs. Producer
|
||||
equality is not independent machine/IR equivalence. Include discovery, schema
|
||||
reads, errors/retries, all responses and final answers in whole-task costs.
|
||||
@@ -1,206 +0,0 @@
|
||||
# Native cases as composable facts
|
||||
|
||||
The SQLite VM has 182 table entries and 149 distinct targets. Its native
|
||||
jump-table owner already knows labels and exact row-to-out-edge mappings;
|
||||
the experiment previously exported only the optimized CFG. This closes that
|
||||
export/query gap without deriving labels from edge order or block addresses.
|
||||
It remains an opt-in experiment, not stock-provider parity or integration.
|
||||
|
||||
## Owners and contracts
|
||||
|
||||
`atomic_membership` gathers BRANCHIND identities during its existing checked
|
||||
membership scan. Read-only `JumpTable` accessors borrow native address/label
|
||||
rows and edge positions. `atomic_switches` checks the exact linked operation,
|
||||
indexes mappings once and exports an independent version-1 facet alongside
|
||||
unchanged graph schema 3 and its optional schema-4 compression envelope.
|
||||
Ordinary non-opt-in analysis and target executable bytes are unchanged.
|
||||
|
||||
Native limits are 12,000 switches, 100,000 rows/mappings and a shared 2 MiB
|
||||
selector/type/address-space text budget, within the existing supervised process
|
||||
and 8 MiB document bounds. Source labels stay exact hex strings. Missing tables,
|
||||
labels, targets/mappings and the native reserved no-label marker remain explicit.
|
||||
No-label status comes from the native owner; it is not guessed from an integer.
|
||||
|
||||
`native-switch-contract` admits rows before expanding validation errors;
|
||||
`native-switch-views` owns identity checks, partial coverage and indexes. A full
|
||||
graph cross-checks operation/value/edge ownership; without it, shared values are
|
||||
checked against available calls/projection facts. Bad switch data makes that
|
||||
facet unavailable while preserving independently valid graph/calls/C.
|
||||
|
||||
`switches` returns metadata and selector facts, never an embedded case inventory.
|
||||
`relations(kind:cases)` selects labels or original row indexes and pages
|
||||
individual case-to-edge relations. A case has a stable case_id; distinct edges
|
||||
have distinct relation IDs. Unmapped cases produce one explicit unknown-edge
|
||||
row. Relation totals are separate from native row and mapping counts. This
|
||||
avoids an unbounded edge array even when limit=1; a regression reads a late
|
||||
two-edge page from a 4,096-mapping case in under 1 KiB of returned items.
|
||||
|
||||
Raw rows are retained once. Row lookup is direct, labels have an index, and
|
||||
prefix offsets permit paging without constructing all relation projections.
|
||||
Single-label offsets are cached only for existing labels, bounded by the
|
||||
retained source; arbitrary missing labels and filter combinations do not grow
|
||||
that cache. The validated model/lease owns its lifetime. Both MCP adapters and
|
||||
the JavaScript CLI consume these same owners.
|
||||
|
||||
The native writer keeps old graph/call coverage ahead of the new facet when
|
||||
space is scarce. A source review found that a later projection omission could
|
||||
free space without restoring saved switch facts. `bounded_reclaiming` retries
|
||||
that saved facet whenever a preferred smaller combination fits. It moves trees,
|
||||
does not clone them, and bounds both encoded string alternatives. A focused
|
||||
native regression and real 2 MiB CLI captures verify the restored facts.
|
||||
|
||||
## Real provider evidence
|
||||
|
||||
Upstream: `fe01d6294eb0d1053ce343c8943646df88e143fc`.
|
||||
Final preserved engine:
|
||||
`a9ca8a8ac224936afb08a9d4be9c1225feb69473ad9f35764e8d1c4f04875b5b`.
|
||||
Patch SHA-256:
|
||||
`9b6642f498f3c9f356b253a2b8b14daf2154cde8eae2d4ab0696f018566e9b73`.
|
||||
The initial public checks used preserved `ac6bc7ab...`, before the formatter
|
||||
reclamation fix. Binaries are copied outside mutable Cargo outputs and not
|
||||
committed. Target/digest and independent machine oracle are in
|
||||
[switch-oracle.md](switch-oracle.md).
|
||||
|
||||
Both final actual MCP paths compare every one of the 182 labelled original
|
||||
targets with independently reconstructed ELF table entries. Selected loaded
|
||||
bytes and dispatch instructions also agree. The native mappings cross-check
|
||||
against the full recovered graph; graph and ordinary C/metadata/calls match the
|
||||
prior complete golden. That producer comparison does not independently prove
|
||||
whole machine/IR equivalence. Each route executes fourteen MCP requests.
|
||||
|
||||
The final wire document is 3,818,818 bytes. Two real native CLI processes with
|
||||
an explicit 2 MiB output budget instead produce identical 823,908-byte records:
|
||||
graph and projection are explicitly unavailable, while C, 416 calls and all
|
||||
182 cases survive. Replaying their retained source through the shared decoder
|
||||
again verifies every original target. This establishes the native CLI/decoder
|
||||
boundary, not a public MCP no-graph run; the MCP owner keeps its 8 MiB budget.
|
||||
|
||||
The recovered selector is eight bytes but comes from a one-byte LOAD/ZEXT. The
|
||||
machine computes its jump address in a different register. Native default edge
|
||||
`8:42` points to block 778, origin `0xeacf0`; the unsigned machine out-of-range
|
||||
guard instead jumps to `0xea998`. Defaults, original addresses and optimized
|
||||
edge identities have separate authority, now stated in the contracts.
|
||||
|
||||
## Executed checks and costs
|
||||
|
||||
The shared JS suites pass 73 cases. Selected native CLI tests pass thirteen;
|
||||
one native jump-table test separately checks nonmatching ordinals/labels/edges,
|
||||
missing labels and inconsistent row lengths. Focused source lint passes without
|
||||
warnings; upstream compiler warnings are outside this patch. Full native corpus,
|
||||
stages/workspace and whole-REA gates remain unexecuted or previously resource
|
||||
limited; no integration readiness is claimed.
|
||||
|
||||
| Executed scope | Family wall s | Family CPU s | Kernel peak MiB |
|
||||
| ----------------------------------------- | ------------: | -----------: | --------------: |
|
||||
| First native export build | 397.637 | 366.618 | 2,169.6 |
|
||||
| Identity/text-budget build | 432.413 | 388.136 | 2,130.6 |
|
||||
| Initial twelve CLI tests | 100.723 | 99.982 | 647.3 |
|
||||
| Native jump-table test, including compile | 277.247 | 276.646 | 2,855.5 |
|
||||
| Final thirteen CLI tests | 52.204 | 40.984 | 476.4 |
|
||||
| Formatter-fix build | 63.414 | 50.208 | 524.5 |
|
||||
| Cached final build | 0.522 | 0.189 | 30.1 |
|
||||
| Initial Code MCP, ac6bc7ab | 41.698 | 41.415 | 656.5 |
|
||||
| Initial direct MCP, ac6bc7ab | 96.920 | 75.949 | 639.2 |
|
||||
| Two 2 MiB native CLI captures | 79.377 | 79.060 | 514.7 |
|
||||
| Final direct MCP | 44.440 | 44.201 | 635.0 |
|
||||
| Final Code MCP | 43.828 | 43.504 | 643.0 |
|
||||
| Both catalog/reference checks | 2.033 | 1.839 | 144.4 |
|
||||
| Real CLI empty switch/ID selection | 3.646 | 3.397 | 188.8 |
|
||||
|
||||
All resource runs use one CPU, 4 GiB hard / 3 GiB early-stop memory bounds,
|
||||
verified cleanup and OOM 0. Peaks include charged file cache, not RSS. Final
|
||||
same-lease snapshot plus selected-case reuse takes 4.8 ms / 0.006 CPU seconds
|
||||
direct and 8.2 ms / 0.009 Code; cumulative memory peaks only upper-bound isolated
|
||||
warm memory. OS caches and host contention are uncontrolled. These scripted
|
||||
measurements do not establish token savings or model outcomes.
|
||||
|
||||
The initial direct capture also ran slowly (about 90 seconds), despite
|
||||
sequential requests and no model. This shows the earlier long capture is not
|
||||
unique to Code Mode or model interaction. The [overlap diagnostic](overlap-study.md)
|
||||
did not reproduce it. Its native/host cause remains unresolved; no speed fix or
|
||||
causal dispatch comparison is claimed.
|
||||
|
||||
## Model question
|
||||
|
||||
The [question](sqlite-switch-question.txt) and [oracle](switch-oracle.md) must be
|
||||
committed before either real model launch. They ask for selected labels,
|
||||
source-byte/normalized-selector widths, signed relative target arithmetic and
|
||||
the native-default versus machine-guard distinction, starting from function
|
||||
name/messages. Both lanes get the full catalog and the same exact discovery,
|
||||
envelope and stopping guidance; reverse engineering facts are not precomputed
|
||||
for the measured models.
|
||||
|
||||
## Real-model selected-case investigation
|
||||
|
||||
The question and oracle were frozen at `f44dca9c3` before both launches. Direct
|
||||
ran first, then Code Mode, with the preserved final engine above, fresh leases,
|
||||
and isolated model workspaces outside the repository. Both receive their full
|
||||
catalog: eighteen direct tools or four Code Mode tools. Exact-name discovery
|
||||
guidance is identical in policy, with no precomputed analysis facts.
|
||||
|
||||
Both complete answers pass the frozen oracle/source review:
|
||||
|
||||
- Two format references select implementation `0xea6d0` over the other callable
|
||||
entry before a single retained VM capture. Neither fetches a full case
|
||||
inventory, C report or recovered graph.
|
||||
- The one-byte loaded selector, eight-byte normalized ZEXT and computed machine
|
||||
jump address remain distinct. Both verify unsigned `JA`, the table base and
|
||||
signed little-endian four-byte displacement arithmetic with selected bytes.
|
||||
- All five requested in-range labels have the correct original row, native
|
||||
edge/block and recovered target. Labels `0x34`/`0x39` retain distinct case IDs
|
||||
despite sharing a destination/edge. Querying six labels returns five rows;
|
||||
`0xb6` has no captured explicit case in this nonpartial table.
|
||||
- Both distinguish native default `8:42` / `0xeacf0` from the machine guard
|
||||
destination `0xea998`, and do not equate default index 42 with label `0x2a`.
|
||||
- Real source IDs and the target digest support the answers. Both preserve
|
||||
machine-flow, memory-alias, nontermination and runtime-feasibility limits and
|
||||
close the lease. Direct cites twelve observed source IDs; Code Mode cites nine.
|
||||
|
||||
Both retained VM sources decode to the full prior graph golden, with ordinary
|
||||
C/metadata/calls unchanged. This establishes producer fidelity separately from
|
||||
the independent selected machine checks; it does not prove global equivalence.
|
||||
The 3,818,818-byte native document stays outside model context. Full answers,
|
||||
events, source Evidence, client tool context and resource receipts are retained
|
||||
locally under `.cache/atomic-codemode/sqlite-switch-{direct,codemode}-v1*`.
|
||||
|
||||
| Whole-task measure | Direct atomic | Code Mode |
|
||||
| ------------------------------ | ------------: | --------: |
|
||||
| Cumulative input tokens | 423,153 | 335,309 |
|
||||
| Included cached input | 367,104 | 308,352 |
|
||||
| Uncached input | 56,049 | 26,957 |
|
||||
| Output tokens | 5,135 | 4,947 |
|
||||
| Included reasoning output | 1,311 | 964 |
|
||||
| Observed model responses | 18 | 15 |
|
||||
| Client execution / wait calls | 11 / 6 | 11 / 3 |
|
||||
| MCP attempts | 17 | 13 |
|
||||
| Successful atomic fact reads | 13 | 14 |
|
||||
| Model wall seconds | 220.581 | 180.656 |
|
||||
| Family wall seconds | 222.275 | 181.173 |
|
||||
| Family CPU seconds | 93.003 | 57.957 |
|
||||
| Kernel process-family peak MiB | 722.4 | 782.3 |
|
||||
| VM source-capture wall seconds | 78.026 | 41.741 |
|
||||
|
||||
There are no failed MCP/API attempts or resource aborts. Per-response usage
|
||||
sums match the ledger. Both request `gpt-6.1-sol/high` through Codex 0.161.0;
|
||||
the resolved service build and remote compute costs are unknown. Cached input
|
||||
is included in input; reasoning is included in output. Runtime, engine and
|
||||
target fingerprints remain unchanged. One CPU, 4 GiB hard / 3 GiB early-stop
|
||||
bounds apply, with OOM 0 and verified payload/profile cleanup. Peaks include
|
||||
charged file cache, not RSS. Leases/provider caches start fresh; OS caches,
|
||||
model prompt caches and host contention are uncontrolled.
|
||||
|
||||
Code Mode uses 20.8% less cumulative input, 51.9% less uncached input and 18.1%
|
||||
less model wall time in this one sample. Native capture takes 78.0 versus 41.7
|
||||
seconds, a substantial confound already observed without models or overlap.
|
||||
The lanes also select different details and discovery contracts. Both clients
|
||||
already compose calls through their JavaScript executor. This is not evidence
|
||||
of a general Code Mode speedup or traditional-provider superiority/parity.
|
||||
|
||||
Client waiting is another observed whole-task cost: each initial snapshot
|
||||
execution yields after about 31 seconds. Both then request one-second waits;
|
||||
direct tries three before switching to longer waits, while Code Mode completes
|
||||
on its third. A slower capture can therefore add model responses and token
|
||||
usage as well as wall time. Those costs remain in the table; subtracting them
|
||||
would not produce a controlled comparison. The next diagnostic should test
|
||||
the same longer-wait guidance in both lanes without changing the question,
|
||||
analysis facts or completeness requirements.
|
||||
@@ -1,81 +0,0 @@
|
||||
"""The cost ledger must not confuse MCP batching with model response rounds."""
|
||||
|
||||
import importlib.util
|
||||
import json
|
||||
from pathlib import Path
|
||||
import tempfile
|
||||
import unittest
|
||||
|
||||
spec = importlib.util.spec_from_file_location(
|
||||
'summarize_trials', Path(__file__).with_name('summarize-trials.py'))
|
||||
ledger = importlib.util.module_from_spec(spec)
|
||||
spec.loader.exec_module(ledger)
|
||||
|
||||
|
||||
class ClientObservations(unittest.TestCase):
|
||||
def test_transport_success_preserves_direct_timeout_and_code_mode_failure(self):
|
||||
with tempfile.TemporaryDirectory() as temporary:
|
||||
path = Path(temporary) / 'trace.jsonl'
|
||||
self.assertIsNone(ledger.trace_observations(path)['api_failures'])
|
||||
entries = [
|
||||
{'tool': 'facts_snapshot', 'result': {
|
||||
'ok': False, 'error': {'code': 'provider_timeout'}}},
|
||||
{'tool': 'facts_snapshot', 'result': {'ok': True, 'data': {}}},
|
||||
{'tool': 'execute_javascript', 'result': {
|
||||
'ok': True, 'evidence_ids': ['ev_a'], 'calls': [
|
||||
{'method': 'snapshot', 'status': 'error',
|
||||
'error': {'code': 'provider_timeout'}}]}},
|
||||
{'tool': 'execute_javascript', 'result': {
|
||||
'ok': False, 'error': {'code': 'output_budget'},
|
||||
'calls': [], 'evidence_ids': ['ev_a']}},
|
||||
]
|
||||
path.write_text('\n'.join(json.dumps(e) for e in entries))
|
||||
result = ledger.trace_observations(path)
|
||||
self.assertEqual(result['direct_fact_calls'], {'facts_snapshot': 2})
|
||||
self.assertEqual(result['code_mode_calls'], {'snapshot': 1})
|
||||
self.assertEqual(len(result['api_failures']), 3)
|
||||
self.assertEqual(result['api_failures'][0]['error']['code'], 'provider_timeout')
|
||||
self.assertEqual(len(result['code_mode_failures']), 1)
|
||||
self.assertEqual(result['code_mode_evidence_count'], 1)
|
||||
|
||||
def test_distinct_responses_and_dispatch_calls_deduplicate_by_identity(self):
|
||||
usage = {'input_tokens': 12, 'cached_input_tokens': 8, 'output_tokens': 2}
|
||||
events = [
|
||||
{'type': 'token_usage_record', 'payload': {'response_id': 'response', 'usage': usage}},
|
||||
{'type': 'response_item', 'payload': {
|
||||
'type': 'custom_tool_call', 'name': 'exec', 'call_id': 'dispatch',
|
||||
'input': 'Two MCP calls may occur inside this one client dispatch.'}},
|
||||
{'type': 'response_item', 'payload': {
|
||||
'type': 'function_call', 'name': 'wait', 'call_id': 'wait', 'arguments': '{}'}},
|
||||
]
|
||||
with tempfile.TemporaryDirectory() as temporary:
|
||||
root = Path(temporary)
|
||||
(root / 'context-receipt.json').write_text(json.dumps({'status': 'captured'}))
|
||||
path = root / 'codex-tool-context.jsonl'
|
||||
path.write_text('\n'.join(json.dumps(event) for event in events * 2))
|
||||
result = ledger.client_observations(root, usage)
|
||||
self.assertEqual(result['model_responses'], 1)
|
||||
self.assertEqual(result['tool_calls'], {'exec': 1, 'wait': 1})
|
||||
self.assertTrue(result['per_response_usage_matches_summary'])
|
||||
with self.assertRaisesRegex(ValueError, 'differs'):
|
||||
ledger.client_observations(root, {'input_tokens': 13})
|
||||
conflicting = {'type': 'token_usage_record', 'payload': {
|
||||
'response_id': 'response', 'usage': {'input_tokens': 13}}}
|
||||
with path.open('a') as output:
|
||||
output.write('\n' + json.dumps(conflicting))
|
||||
with self.assertRaisesRegex(ValueError, 'Conflicting usage'):
|
||||
ledger.client_observations(root, usage)
|
||||
|
||||
def test_missing_or_partial_capture_does_not_claim_zero_calls_or_complete_usage(self):
|
||||
with tempfile.TemporaryDirectory() as temporary:
|
||||
root = Path(temporary)
|
||||
self.assertIsNone(ledger.client_observations(root, {})['model_responses'])
|
||||
(root / 'context-receipt.json').write_text(json.dumps({'status': 'partial'}))
|
||||
(root / 'codex-tool-context.jsonl').write_text('')
|
||||
result = ledger.client_observations(root, {'input_tokens': 100})
|
||||
self.assertIsNone(result['model_responses'])
|
||||
self.assertIsNone(result['per_response_usage_matches_summary'])
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
unittest.main()
|
||||
File diff suppressed because it is too large.
Load diff
@@ -1,128 +0,0 @@
|
||||
// Real public MCP checks for both calling adapters and the stock/patched CLI.
|
||||
import assert from "node:assert/strict";
|
||||
import { readFile, writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
const verifier = createVerifierRun();
|
||||
const client = new Client({ name: "rea-byte-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const calls = [],
|
||||
failures = [];
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const call = async (name, input) => {
|
||||
const result = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(result._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
calls.push({ name, input, value });
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
const read = async (windows) => {
|
||||
const result = direct
|
||||
? await call("facts_bytes", { windows })
|
||||
: await call("execute_javascript", {
|
||||
code: `const r=await rea.bytes({windows:${JSON.stringify(windows)}});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;`,
|
||||
});
|
||||
return direct ? result.data : result.value;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
const byName = new Map(catalog.tools.map((tool) => [tool.name, tool]));
|
||||
assert.ok(byName.has(direct ? "facts_bytes" : "execute_javascript"));
|
||||
assert.equal(byName.has("execute_javascript"), !direct);
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
const opened = await call("open_binary", { path: binary });
|
||||
assert.equal(
|
||||
opened.context.artifact_sha256,
|
||||
"2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8",
|
||||
);
|
||||
const mixed = await read([
|
||||
{ address: "0x129238", length: 53 },
|
||||
{ address: "0x154bf8", length: 16 },
|
||||
{ address: "0xdeadbeef", length: 1 },
|
||||
]);
|
||||
assert.equal(
|
||||
mixed.results[0].hex,
|
||||
Buffer.from(
|
||||
"Safety level may not be changed inside a transaction\0",
|
||||
).toString("hex"),
|
||||
);
|
||||
assert.equal(
|
||||
mixed.results[0].coverage.status,
|
||||
"complete-for-requested-range",
|
||||
);
|
||||
assert.equal(mixed.results[0].coverage.original_file_mapping, "unknown");
|
||||
assert.equal(mixed.results[1].returned_bytes, 4);
|
||||
assert.equal(mixed.results[1].end_address, "0x154bfc");
|
||||
assert.equal(mixed.results[1].coverage.status, "partial");
|
||||
assert.ok(mixed.results[1].coverage.notes.length > 0);
|
||||
assert.equal(mixed.results[2].ok, false);
|
||||
const again = await read([{ address: "0x129238", length: 53 }]);
|
||||
assert.deepEqual(again.results[0], mixed.results[0]);
|
||||
let maximum;
|
||||
if (direct) {
|
||||
const result = await read([{ address: "0x120000", length: 65536 }]);
|
||||
maximum = result.results[0];
|
||||
assert.equal(maximum.hex.length, 131072);
|
||||
// The pinned ELF LOAD maps these VMAs to identical original file offsets.
|
||||
const artifact = await readFile(binary);
|
||||
assert.equal(
|
||||
maximum.hex,
|
||||
artifact.subarray(0x120000, 0x130000).toString("hex"),
|
||||
);
|
||||
} else {
|
||||
const result = await call("execute_javascript", {
|
||||
code: 'const r=await rea.bytes({windows:[{address:"0x120000",length:65536}]});if(!r.ok)throw new Error(JSON.stringify(r));const {hex,...facts}=r.data.results[0];return {...facts,hex_length:hex.length};',
|
||||
});
|
||||
maximum = result.value;
|
||||
assert.equal(maximum.hex_length, 131072);
|
||||
}
|
||||
assert.equal(maximum.returned_bytes, 65536);
|
||||
assert.equal(maximum.coverage.status, "complete-for-requested-range");
|
||||
await call("close_binary", {});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({ direct, calls, failures, lifecycle }),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({
|
||||
output,
|
||||
direct,
|
||||
calls: calls.length,
|
||||
failures,
|
||||
cleanup_verified: true,
|
||||
}),
|
||||
);
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,94 +0,0 @@
|
||||
// A real MCP cold/warm read benchmark. No model runs in this verifier.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import { familyCounters as counters } from "./family-counters.mjs";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output, entry = "0xc1e0"] = process.argv.slice(2);
|
||||
const verifier = createVerifierRun();
|
||||
const client = new Client({ name: "rea-cache-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const phases = [],
|
||||
failures = [];
|
||||
const call = async (name, args = {}) => {
|
||||
const reply = await client.callTool(
|
||||
{ name, arguments: args },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
const value = JSON.parse(reply.content[0].text);
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
await call("open_binary", { path: binary });
|
||||
for (const phase of ["cold_snapshot", "warm_snapshot"]) {
|
||||
const before = await counters(),
|
||||
start = performance.now();
|
||||
const result = await call("execute_javascript", {
|
||||
code: `const r=await rea.snapshot({entry:${JSON.stringify(entry)}});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;`,
|
||||
});
|
||||
const elapsed = performance.now() - start,
|
||||
after = await counters();
|
||||
phases.push({
|
||||
phase,
|
||||
elapsed_ms: elapsed,
|
||||
family_cpu_seconds: (after.cpu - before.cpu) / 1e6,
|
||||
cumulative_family_memory_peak_bytes: after.peak,
|
||||
snapshot: result.value.snapshot,
|
||||
source: result.value.source,
|
||||
facets: result.value.facets,
|
||||
});
|
||||
}
|
||||
assert.equal(phases[0].snapshot, phases[1].snapshot);
|
||||
assert.equal(phases[0].source.evidence_id, phases[1].source.evidence_id);
|
||||
await call("close_binary");
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
try {
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({
|
||||
phases,
|
||||
failures,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Real MCP, no model/token measurement. CPU deltas cover the owned cgroup. Memory peaks are cumulative; the warm value is an upper bound on its isolated phase peak. Both reads share one immutable captured artifact/lease; this is not a cross-session cache benchmark.",
|
||||
}),
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({
|
||||
output,
|
||||
phases: phases.map(({ phase, elapsed_ms, family_cpu_seconds }) => ({
|
||||
phase,
|
||||
elapsed_ms,
|
||||
family_cpu_seconds,
|
||||
})),
|
||||
}),
|
||||
);
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,206 +0,0 @@
|
||||
// Real public MCP checks; script assertions are not real-model performance data.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
const verifier = createVerifierRun();
|
||||
const client = new Client({ name: "rea-call-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const calls = [],
|
||||
failures = [];
|
||||
const call = async (name, input) => {
|
||||
const result = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(result._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
calls.push({ name, input, value });
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
const read = async (name, input) => {
|
||||
const value = direct
|
||||
? await call(`facts_${name}`, input)
|
||||
: await call("execute_javascript", {
|
||||
code: `const r=await rea.${name}(${JSON.stringify(input)});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;`,
|
||||
});
|
||||
return direct ? value.data : value.value;
|
||||
};
|
||||
const home = (row, space, offset, width) =>
|
||||
row.space === space && row.offset === offset && row.size_bytes === width;
|
||||
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
assert.equal(catalog.tools.length, direct ? 17 : 4);
|
||||
const opened = await call("open_binary", { path: binary });
|
||||
assert.equal(
|
||||
opened.context.artifact_sha256,
|
||||
"2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8",
|
||||
);
|
||||
assert.equal(opened.context.analysis_model, "experimental-native-ir-v3");
|
||||
const wrapper = await read("snapshot", { entry: "0x65180" });
|
||||
assert.deepEqual(wrapper.facets.calls.by_opcode, [
|
||||
{ opcode: "CPUI_CALL", known_target_count: 8, unknown_target_count: 0 },
|
||||
{ opcode: "CPUI_CALLIND", known_target_count: 0, unknown_target_count: 0 },
|
||||
]);
|
||||
const selected = await read("calls", {
|
||||
snapshot: wrapper.snapshot,
|
||||
origin_addresses: ["0x65211"],
|
||||
});
|
||||
assert.equal(selected.total, 1);
|
||||
const parserCall = selected.items[0];
|
||||
assert.deepEqual(parserCall.target, { space: "ram", offset: "0x64060" });
|
||||
const arg = parserCall.arguments.find(
|
||||
(a) => a.storage && home(a.storage, "register", "0x38", 8),
|
||||
);
|
||||
assert.ok(arg, "recovered RDI passing storage");
|
||||
const callerValue = await read("values", {
|
||||
snapshot: wrapper.snapshot,
|
||||
ids: [arg.value_id],
|
||||
});
|
||||
assert.equal(callerValue.items[0].is_input, false);
|
||||
assert.ok(callerValue.items[0].definition);
|
||||
const parser = await read("snapshot", { entry: parserCall.target.offset });
|
||||
assert.deepEqual(parser.facets.calls.by_opcode, [
|
||||
{ opcode: "CPUI_CALL", known_target_count: 28, unknown_target_count: 0 },
|
||||
{ opcode: "CPUI_CALLIND", known_target_count: 0, unknown_target_count: 5 },
|
||||
]);
|
||||
const inputs = await read("values", {
|
||||
snapshot: parser.snapshot,
|
||||
is_input: true,
|
||||
fields: ["space", "offset", "size_bytes", "is_input"],
|
||||
limit: 128,
|
||||
});
|
||||
assert.equal(inputs.next_offset, null);
|
||||
const candidates = inputs.items.filter((v) =>
|
||||
home(v, arg.storage.space, arg.storage.offset, arg.storage.size_bytes),
|
||||
);
|
||||
assert.equal(candidates.length, 1);
|
||||
const updates = await read("operations", {
|
||||
snapshot: parser.snapshot,
|
||||
origin_addresses: ["0x64699", "0x64c92"],
|
||||
kind: "store",
|
||||
fields: ["inputs", "definitions"],
|
||||
limit: 8,
|
||||
});
|
||||
assert.equal(updates.total, 2);
|
||||
const dependencies = await read("slice", {
|
||||
snapshot: parser.snapshot,
|
||||
seeds: updates.items.map((o) => o.id),
|
||||
max_operations: 32,
|
||||
fields: ["opcode", "inputs", "constants"],
|
||||
});
|
||||
assert.ok(
|
||||
dependencies.value_frontier.some((v) => v.value_id === candidates[0].id),
|
||||
);
|
||||
const indirect = await read("calls", {
|
||||
snapshot: parser.snapshot,
|
||||
origin_addresses: ["0x644eb", "0x6450c", "0x64af3", "0x647d1", "0x64807"],
|
||||
fields: ["target", "arguments"],
|
||||
});
|
||||
assert.equal(indirect.total, 5);
|
||||
assert.ok(
|
||||
indirect.items.every(
|
||||
(c) =>
|
||||
c.opcode === "CPUI_CALLIND" && c.target === null && c.spec_available,
|
||||
),
|
||||
);
|
||||
const machine = await read("instructions", {
|
||||
windows: [{ address: "0x644eb", count: 1 }],
|
||||
fields: ["bytes", "mnemonic"],
|
||||
});
|
||||
assert.equal(machine.results[0].items[0].bytes, "ff5328");
|
||||
assert.equal(machine.results[0].items[0].mnemonic, "CALL");
|
||||
const formatter = await read("snapshot", { entry: "0x5efc0" });
|
||||
const formatCalls = await read("calls", {
|
||||
snapshot: formatter.snapshot,
|
||||
origin_addresses: ["0x5f031", "0x5f062"],
|
||||
});
|
||||
assert.deepEqual(
|
||||
formatCalls.items.map((c) => c.target?.offset),
|
||||
["0x2adc0", "0x27e50"],
|
||||
);
|
||||
const ops = await read("operations", {
|
||||
snapshot: formatter.snapshot,
|
||||
ids: formatCalls.items.map((c) => c.id),
|
||||
fields: ["inputs"],
|
||||
});
|
||||
const annotations = await read("values", {
|
||||
snapshot: formatter.snapshot,
|
||||
ids: ops.items.map((o) => o.inputs[0].value_id),
|
||||
});
|
||||
assert.ok(
|
||||
annotations.items.every((v) => v.space === "fspec" && v.is_annotation),
|
||||
);
|
||||
assert.deepEqual(
|
||||
annotations.items.map((v) => v.offset),
|
||||
["0x10000", "0x10001"],
|
||||
);
|
||||
const projected = await read("calls", {
|
||||
snapshot: formatter.snapshot,
|
||||
ids: [formatCalls.items[0].id, "missing"],
|
||||
fields: ["target"],
|
||||
limit: 1,
|
||||
});
|
||||
assert.deepEqual(projected.missing, ["missing"]);
|
||||
assert.equal(Object.hasOwn(projected.items[0], "arguments"), false);
|
||||
const cached = await read("snapshot", { entry: "0x65180" });
|
||||
assert.equal(cached.snapshot, wrapper.snapshot);
|
||||
assert.equal(cached.source.evidence_id, wrapper.source.evidence_id);
|
||||
if (!direct) {
|
||||
const composed = await call("execute_javascript", {
|
||||
code: `const s=await rea.snapshot({entry:"0x65180"});const c=await rea.calls({snapshot:s.data.snapshot,origin_addresses:["0x65211"]});const a=c.data.items[0].arguments.find(a=>a.storage?.space==="register"&&a.storage.offset==="0x38");const t=await rea.snapshot({entry:c.data.items[0].target.offset});const v=await rea.values({snapshot:t.data.snapshot,is_input:true,limit:128});return {candidate_inputs:v.data.items.filter(v=>v.space===a.storage.space&&v.offset===a.storage.offset&&v.size_bytes===a.storage.size_bytes).map(v=>v.id)};`,
|
||||
});
|
||||
assert.deepEqual(
|
||||
composed.value.candidate_inputs,
|
||||
candidates.map((v) => v.id),
|
||||
);
|
||||
assert.ok(composed.evidence_ids.length >= 2);
|
||||
}
|
||||
await call("close_binary", {});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({ direct, calls, failures, lifecycle }),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({
|
||||
output,
|
||||
direct,
|
||||
calls: calls.length,
|
||||
failures,
|
||||
cleanup_verified: true,
|
||||
}),
|
||||
);
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,93 +0,0 @@
|
||||
// Real public MCP lifecycle, with one explicit test-only tagged failure.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join, resolve } from "node:path";
|
||||
import { pathToFileURL } from "node:url";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import { PrivateRuntimeRoot } from "../../dist/process/PrivateRuntimeRoot.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
assert.ok(binary && output, "Usage: verify-cleanup-mcp.mjs BINARY OUTPUT");
|
||||
const run = createVerifierRun(),
|
||||
calls = [],
|
||||
failures = [],
|
||||
root = await PrivateRuntimeRoot.create({
|
||||
prefix: "rea-cleanup-mcp-verifier-",
|
||||
});
|
||||
const url = (path) => JSON.stringify(pathToFileURL(resolve(path)).href);
|
||||
const bootstrap = join(root.path, "injected-cleanup.mjs");
|
||||
await writeFile(
|
||||
bootstrap,
|
||||
`import {KunaTargetCommands} from ${url("dist/kuna/KunaTargetCommands.js")};
|
||||
import {ProviderCleanupError} from ${url("dist/domain/providerCleanupError.js")};
|
||||
import {err} from ${url("dist/domain/result.js")};
|
||||
const close=KunaTargetCommands.prototype.close;let injected=false;
|
||||
KunaTargetCommands.prototype.close=async function(){
|
||||
if(!injected){injected=true;return err(new ProviderCleanupError("kuna",["verifier-injected-cleanup-resource"],{reason:"Explicit verifier fault; not a real engine cleanup failure"}));}
|
||||
return close.call(this);
|
||||
};`,
|
||||
{ flag: "wx" },
|
||||
);
|
||||
const client = new Client({ name: "rea-cleanup-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [
|
||||
"--import",
|
||||
bootstrap,
|
||||
resolve("experiments/atomic-codemode/server.mjs"),
|
||||
],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const call = async (name, input) => {
|
||||
const result = await client.callTool({ name, arguments: input });
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
calls.push({ name, input, value });
|
||||
return value;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const opened = await call("open_binary", { path: binary });
|
||||
assert.equal(opened.ok, true);
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const read = direct
|
||||
? await call("facts_functions", { limit: 1, fields: [] })
|
||||
: await call("execute_javascript", {
|
||||
code: "return await rea.functions({limit:1,fields:[]});",
|
||||
});
|
||||
assert.equal(read.ok, true);
|
||||
if (!direct) assert.equal(read.value.ok, true);
|
||||
const failed = await call("close_binary", {});
|
||||
assert.equal(failed.ok, false);
|
||||
assert.equal(failed.context.lease, opened.context.lease);
|
||||
assert.equal(failed.error.code, "cleanup_incomplete");
|
||||
assert.deepEqual(failed.error.details.failure.details.resources, [
|
||||
"verifier-injected-cleanup-resource",
|
||||
]);
|
||||
assert.equal((await call("close_binary", {})).ok, true);
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
// Also drain the one-shot injection if a preceding assertion failed early.
|
||||
for (let retry = 0; retry < 2; retry++) {
|
||||
try {
|
||||
await call("close_binary", {});
|
||||
} catch {}
|
||||
}
|
||||
await client.close();
|
||||
await transport.close();
|
||||
await root.close();
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(output, JSON.stringify({ calls, failures, lifecycle }), {
|
||||
flag: "wx",
|
||||
});
|
||||
console.log(JSON.stringify({ output, calls: calls.length, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,102 +0,0 @@
|
||||
// Real SDK/provider lifecycle with an injected tagged cleanup failure.
|
||||
// The failure is simulated; native reads and the successful retry use real owners.
|
||||
import assert from "node:assert/strict";
|
||||
import { stat, writeFile } from "node:fs/promises";
|
||||
import { AtomicSession } from "./api.mjs";
|
||||
import { KunaFactOwner } from "./provider.mjs";
|
||||
import { KunaTargetCommands } from "../../dist/kuna/KunaTargetCommands.js";
|
||||
import { ProviderCleanupError } from "../../dist/domain/providerCleanupError.js";
|
||||
import { err } from "../../dist/domain/result.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
assert.ok(binary && output, "Usage: verify-cleanup.mjs BINARY OUTPUT");
|
||||
const run = createVerifierRun(),
|
||||
originalClose = KunaTargetCommands.prototype.close,
|
||||
observations = [],
|
||||
failures = [];
|
||||
const owners = [];
|
||||
try {
|
||||
for (const action of ["close", "replace"]) {
|
||||
const session = new AtomicSession(async (path) => {
|
||||
const owner = await KunaFactOwner.open(path);
|
||||
owners.push(owner);
|
||||
return owner;
|
||||
});
|
||||
await session.open(binary);
|
||||
const lease = session.context.lease;
|
||||
const inventory = await session.call("functions", { limit: 1, fields: [] });
|
||||
assert.equal(inventory.ok, true);
|
||||
const evidence = await session.call("evidence", {
|
||||
evidence_id: inventory.data.source.evidence_id,
|
||||
limit: 1,
|
||||
});
|
||||
assert.equal(evidence.ok, true);
|
||||
const staged = evidence.data.source.arguments[1],
|
||||
identity = await stat(staged);
|
||||
let injected = false;
|
||||
KunaTargetCommands.prototype.close = async function () {
|
||||
if (!injected) {
|
||||
injected = true;
|
||||
return err(
|
||||
new ProviderCleanupError("kuna", [staged], {
|
||||
reason: "Verifier-injected tagged cleanup failure before retry",
|
||||
}),
|
||||
);
|
||||
}
|
||||
return originalClose.call(this);
|
||||
};
|
||||
const attempt = () =>
|
||||
action === "close" ? session.close() : session.open(binary);
|
||||
await assert.rejects(
|
||||
attempt(),
|
||||
(error) =>
|
||||
error.code === "cleanup_incomplete" &&
|
||||
error.details.failure.details.resources.includes(staged),
|
||||
`${action} must surface tagged cleanup failure`,
|
||||
);
|
||||
assert.equal(injected, true);
|
||||
assert.equal(
|
||||
session.context.lease,
|
||||
lease,
|
||||
"Failed cleanup retains its owner",
|
||||
);
|
||||
assert.equal((await stat(staged)).ino, identity.ino);
|
||||
const retried = await attempt();
|
||||
assert.equal(retried.ok, true);
|
||||
if (action === "close") assert.equal(session.context, null);
|
||||
else assert.notEqual(session.context.lease, lease);
|
||||
await assert.rejects(stat(staged), { code: "ENOENT" });
|
||||
await session.close();
|
||||
observations.push({
|
||||
action,
|
||||
tagged_failure_reported: true,
|
||||
owner_and_staged_file_retained: true,
|
||||
retry_succeeded: true,
|
||||
staged_file_removed_after_retry: true,
|
||||
});
|
||||
KunaTargetCommands.prototype.close = originalClose;
|
||||
}
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
KunaTargetCommands.prototype.close = originalClose;
|
||||
for (const owner of owners) {
|
||||
try {
|
||||
await owner.close();
|
||||
} catch (error) {
|
||||
failures.push(`Final cleanup: ${String(error)}`);
|
||||
}
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(output, JSON.stringify({ observations, failures, lifecycle }), {
|
||||
flag: "wx",
|
||||
});
|
||||
console.log(JSON.stringify({ output, observations, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,189 +0,0 @@
|
||||
// Verify actual SDK-converted catalogs and named help without target analysis.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [output] = process.argv.slice(2);
|
||||
const run = createVerifierRun(),
|
||||
results = [],
|
||||
failures = [];
|
||||
for (const direct of [false, true]) {
|
||||
const client = new Client({
|
||||
name: "rea-discovery-verifier",
|
||||
version: "0.1.0",
|
||||
});
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env, REA_ATOMIC_ADAPTER: direct ? "direct" : "codemode" },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const call = async (query, view = "contracts") => {
|
||||
const reply = await client.callTool({
|
||||
name: "reference",
|
||||
arguments: { query, view },
|
||||
});
|
||||
return JSON.parse(reply.content[0].text);
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
const prefix = direct ? "facts_" : "rea.";
|
||||
const ids = [
|
||||
"switches",
|
||||
"snapshot",
|
||||
"operations",
|
||||
"relations",
|
||||
"values",
|
||||
"instructions",
|
||||
"bytes",
|
||||
"evidence",
|
||||
];
|
||||
const selected = await call(ids.map((name) => prefix + name).join(" "));
|
||||
assert.deepEqual(
|
||||
selected.methods.map((m) => m.method),
|
||||
ids,
|
||||
);
|
||||
assert.deepEqual(selected.unmatched_queries, []);
|
||||
assert.equal(selected.coverage.declared_methods, 15);
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
for (const method of selected.methods) {
|
||||
ajv.compile(method.input_schema);
|
||||
assert.equal(method.name, prefix + method.method);
|
||||
if (direct)
|
||||
assert.ok(catalog.tools.some((tool) => tool.name === method.name));
|
||||
}
|
||||
for (const [name, good, bad] of [
|
||||
["snapshot", { entry: "0xea6d0" }, { snapshot: "snap_other" }],
|
||||
[
|
||||
"switches",
|
||||
{
|
||||
snapshot: "snap_other",
|
||||
origin_addresses: ["0xea801"],
|
||||
fields: ["selector", "table"],
|
||||
},
|
||||
{ snapshot: "snap_other", fields: ["rows"] },
|
||||
],
|
||||
["values", { snapshot: "snap_other", ids: ["v1"] }, { entry: "0xea6d0" }],
|
||||
[
|
||||
"relations",
|
||||
{
|
||||
snapshot: "snap_other",
|
||||
kind: "cases",
|
||||
seeds: ["0xea801:fe"],
|
||||
case_values: ["0xb5"],
|
||||
case_indices: [181],
|
||||
},
|
||||
{ snapshot: "snap_other", kind: "cases", seeds: [8] },
|
||||
],
|
||||
[
|
||||
"relations",
|
||||
{
|
||||
snapshot: "snap_other",
|
||||
kind: "operands",
|
||||
seeds: ["0x1:0"],
|
||||
input_indices: [0, 155],
|
||||
},
|
||||
{
|
||||
snapshot: "snap_other",
|
||||
kind: "uses",
|
||||
seeds: ["0x1:0"],
|
||||
input_indices: [0],
|
||||
},
|
||||
],
|
||||
[
|
||||
"evidence",
|
||||
{ evidence_id: "ev_example", component: "recovered_ir" },
|
||||
{ evidence_id: "ev_example", component: "unknown" },
|
||||
],
|
||||
]) {
|
||||
const method = selected.methods.find((m) => m.method === name);
|
||||
const validators = [ajv.compile(method.input_schema)];
|
||||
if (direct)
|
||||
validators.push(
|
||||
ajv.compile(
|
||||
catalog.tools.find((t) => t.name === method.name).inputSchema,
|
||||
),
|
||||
);
|
||||
for (const validate of validators) {
|
||||
assert.equal(validate(good), true);
|
||||
assert.equal(validate(bad), false);
|
||||
}
|
||||
}
|
||||
const missing = await call(`${prefix}values unknown`);
|
||||
assert.deepEqual(
|
||||
missing.methods.map((m) => m.method),
|
||||
["values"],
|
||||
);
|
||||
assert.deepEqual(missing.unmatched_queries, ["unknown"]);
|
||||
const all = await call("");
|
||||
assert.equal(all.methods.length, 15);
|
||||
for (const method of all.methods) ajv.compile(method.input_schema);
|
||||
const index = await call("", "index");
|
||||
assert.equal(index.coverage.view, "index");
|
||||
assert.deepEqual(
|
||||
index.methods.map((m) => m.method),
|
||||
all.methods.map((m) => m.method),
|
||||
);
|
||||
assert.ok(
|
||||
index.methods.every(
|
||||
(m) =>
|
||||
!Object.hasOwn(m, "input_schema") && !Object.hasOwn(m, "returns"),
|
||||
),
|
||||
);
|
||||
assert.deepEqual(
|
||||
index.methods.map((m) => ({
|
||||
method: m.method,
|
||||
name: m.name,
|
||||
description: m.description,
|
||||
})),
|
||||
all.methods.map((m) => ({
|
||||
method: m.method,
|
||||
name: m.name,
|
||||
description: m.description,
|
||||
})),
|
||||
);
|
||||
assert.ok(
|
||||
Buffer.byteLength(JSON.stringify(index)) <
|
||||
Buffer.byteLength(JSON.stringify(all)),
|
||||
);
|
||||
assert.ok(JSON.stringify(selected).length < JSON.stringify(all).length);
|
||||
results.push({
|
||||
direct,
|
||||
selected,
|
||||
all_reference_bytes: Buffer.byteLength(JSON.stringify(all)),
|
||||
index_bytes: Buffer.byteLength(JSON.stringify(index)),
|
||||
selected_reference_bytes: Buffer.byteLength(JSON.stringify(selected)),
|
||||
actual_catalog_and_named_schema_checks_passed: true,
|
||||
});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({
|
||||
results,
|
||||
failures,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Static contract/catalog checks; no native target analysis or model cost measurement",
|
||||
}),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(JSON.stringify({ output, lanes: results.length, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,88 +0,0 @@
|
||||
// Real MCP failure delivery, including source access after an oversized error.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
const verifier = createVerifierRun();
|
||||
const client = new Client({ name: "rea-executor-verifier", version: "0.1" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const observations = [],
|
||||
failures = [];
|
||||
const call = async (name, input = {}) => {
|
||||
const result = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(result._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
observations.push({
|
||||
name,
|
||||
input,
|
||||
value,
|
||||
encoded_utf8_bytes: Buffer.byteLength(result.content[0].text),
|
||||
});
|
||||
return value;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
for (const code of [
|
||||
'throw new Error("x".repeat(40000))',
|
||||
'throw new Error("\\0".repeat(6000))',
|
||||
]) {
|
||||
const result = await call("execute_javascript", { code });
|
||||
assert.equal(result.ok, false);
|
||||
assert.equal(result.error.code, "output_budget");
|
||||
assert.equal(result.error.details.original_message_omitted, true);
|
||||
assert.ok(observations.at(-1).encoded_utf8_bytes < 32768);
|
||||
}
|
||||
const ordinary = await call("execute_javascript", {
|
||||
code: 'throw new Error("x".repeat(10000))',
|
||||
});
|
||||
assert.equal(ordinary.error.code, "javascript_error");
|
||||
assert.ok(ordinary.error.message.includes("x".repeat(10000)));
|
||||
const opened = await call("open_binary", { path: binary });
|
||||
assert.equal(opened.ok, true);
|
||||
const result = await call("execute_javascript", {
|
||||
code: 'const r=await rea.strings({query:"incorrect header check"});if(!r.ok)throw new Error(JSON.stringify(r));throw new Error("x".repeat(40000));',
|
||||
});
|
||||
assert.equal(result.error.code, "output_budget");
|
||||
assert.equal(result.context.lease, opened.context.lease);
|
||||
assert.equal(result.evidence_ids.length, 1);
|
||||
const source = await call("execute_javascript", {
|
||||
code: `const r=await rea.evidence({evidence_id:${JSON.stringify(result.evidence_ids[0])},limit:64});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;`,
|
||||
});
|
||||
assert.equal(source.ok, true);
|
||||
assert.equal(source.value.source.evidence_id, result.evidence_ids[0]);
|
||||
assert.ok(source.value.source.bytes > 0);
|
||||
assert.ok(source.value.text.length > 0);
|
||||
assert.equal((await call("close_binary")).ok, true);
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(output, JSON.stringify({ observations, failures, lifecycle }), {
|
||||
flag: "wx",
|
||||
});
|
||||
console.log(JSON.stringify({ output, failures, cleanup_verified: true }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,227 +0,0 @@
|
||||
// Real public retained-graph reads; producer fidelity is not machine equivalence.
|
||||
import assert from "node:assert/strict";
|
||||
import { createHash } from "node:crypto";
|
||||
import { readFile, writeFile } from "node:fs/promises";
|
||||
import { resolve, join } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import { readNativeGraphEvidence } from "./native-ir-contract.mjs";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output, goldenPath, partialPath] = process.argv.slice(2);
|
||||
assert.equal(
|
||||
process.versions.node,
|
||||
(await readFile(resolve(".nvmrc"), "utf8")).trim(),
|
||||
"Use the repository's pinned Node before running native analysis",
|
||||
);
|
||||
const golden = JSON.parse(await readFile(goldenPath, "utf8")).functions[0];
|
||||
const graph = golden.atomic_facts;
|
||||
const partial = JSON.parse(await readFile(partialPath, "utf8")).functions[0]
|
||||
.atomic_projection;
|
||||
const formerFrontier = partial.coverage.frontier
|
||||
.slice(0, 3)
|
||||
.map((row) => row.definition);
|
||||
const run = createVerifierRun(),
|
||||
calls = [],
|
||||
failures = [];
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const client = new Client({
|
||||
name: "rea-full-graph-verifier",
|
||||
version: "0.1.0",
|
||||
});
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const call = async (name, input) => {
|
||||
const reply = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
const value = JSON.parse(reply.content[0].text);
|
||||
calls.push({
|
||||
name,
|
||||
input,
|
||||
value,
|
||||
content_bytes: Buffer.byteLength(reply.content[0].text),
|
||||
});
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
const read = async (name, input) => {
|
||||
const value = direct
|
||||
? await call(`facts_${name}`, input)
|
||||
: (
|
||||
await call("execute_javascript", {
|
||||
code: `return await rea.${name}(${JSON.stringify(input)});`,
|
||||
})
|
||||
).value;
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value.data;
|
||||
};
|
||||
let sourceReview;
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools(),
|
||||
ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
await call("open_binary", { path: binary });
|
||||
const capture = await read("snapshot", { entry: "0xea6d0" }),
|
||||
snapshot = capture.snapshot;
|
||||
assert.equal(capture.facets.recovered_ir.availability, "available");
|
||||
assert.equal(capture.facets.operations.count, 21349);
|
||||
assert.equal(capture.facets.values.count, 35755);
|
||||
assert.equal(capture.coverage.blocks, 1229);
|
||||
assert.equal(capture.coverage.omitted_definition_count, 0);
|
||||
assert.equal(capture.coverage.runtime_successors, "unknown");
|
||||
const definitions = await read("operations", {
|
||||
snapshot,
|
||||
ids: formerFrontier,
|
||||
fields: ["opcode", "output"],
|
||||
});
|
||||
assert.deepEqual(
|
||||
definitions.items.map((op) => op.id),
|
||||
formerFrontier,
|
||||
);
|
||||
for (const op of definitions.items) {
|
||||
const expected = graph.operations.find((row) => row.id === op.id);
|
||||
assert.equal(op.opcode, expected.opcode);
|
||||
assert.equal(op.output, expected.output);
|
||||
}
|
||||
const blocks = await read("blocks", {
|
||||
snapshot,
|
||||
ids: [1154, 1159],
|
||||
fields: ["predecessors", "successors", "operations"],
|
||||
});
|
||||
for (const block of blocks.items) {
|
||||
const expected = graph.blocks.find((row) => row.id === block.id);
|
||||
for (const field of ["predecessors", "successors", "operations"])
|
||||
assert.deepEqual(block[field], expected[field]);
|
||||
}
|
||||
const edges = await read("relations", {
|
||||
snapshot,
|
||||
kind: "incoming",
|
||||
seeds: [1154, 1159],
|
||||
limit: 128,
|
||||
});
|
||||
const expectedEdges = [1154, 1159].flatMap((seed) =>
|
||||
graph.blocks.flatMap((block) =>
|
||||
block.successors
|
||||
.filter((edge) => edge.target === seed)
|
||||
.map((edge) => ({
|
||||
...edge,
|
||||
from: block.id,
|
||||
to: edge.target,
|
||||
operation: block.operations.at(-1) ?? null,
|
||||
kind: edge.polarity ? "conditional" : "recovered",
|
||||
})),
|
||||
),
|
||||
);
|
||||
assert.deepEqual(edges.items, expectedEdges);
|
||||
const operands = await read("relations", {
|
||||
snapshot,
|
||||
kind: "operands",
|
||||
seeds: ["0xea998:14649"],
|
||||
input_indices: [0, 105, 155],
|
||||
});
|
||||
assert.equal(operands.items.length, 3);
|
||||
const retained = await read("calls", {
|
||||
snapshot,
|
||||
origin_addresses: ["0xf115c", "0xf238b"],
|
||||
fields: ["arguments", "target"],
|
||||
});
|
||||
assert.equal(retained.items.length, 2);
|
||||
assert.equal(retained.evidence_id, capture.source.evidence_id);
|
||||
const raw = await read("evidence", {
|
||||
evidence_id: capture.source.evidence_id,
|
||||
limit: 128,
|
||||
});
|
||||
const decoded = await read("evidence", {
|
||||
evidence_id: capture.source.evidence_id,
|
||||
component: "recovered_ir",
|
||||
limit: 256,
|
||||
});
|
||||
assert.equal(raw.view, undefined);
|
||||
assert.equal(decoded.view.encoding, "gzip-base64");
|
||||
assert.equal(decoded.source.sha256, capture.source.sha256);
|
||||
const text = await readFile(
|
||||
join(
|
||||
process.env.REA_ATOMIC_EVIDENCE_DIR,
|
||||
`${capture.source.evidence_id}.txt`,
|
||||
),
|
||||
"utf8",
|
||||
);
|
||||
assert.equal(
|
||||
createHash("sha256").update(text).digest("hex"),
|
||||
capture.source.sha256,
|
||||
);
|
||||
assert.ok(Buffer.byteLength(text) <= 8 * 1024 * 1024);
|
||||
const fn = JSON.parse(text).functions[0];
|
||||
const view = readNativeGraphEvidence(text);
|
||||
assert.deepEqual(JSON.parse(view.text), graph);
|
||||
assert.equal(
|
||||
decoded.view.sha256,
|
||||
createHash("sha256").update(view.text).digest("hex"),
|
||||
);
|
||||
assert.equal(decoded.view.bytes, Buffer.byteLength(view.text));
|
||||
assert.equal(decoded.text, view.text.slice(0, 256));
|
||||
const { atomic_facts: _ignored, ...ordinary } = fn;
|
||||
const { atomic_facts: _oldGraph, ...goldenOrdinary } = golden;
|
||||
assert.deepEqual(ordinary, goldenOrdinary);
|
||||
const next = await read("evidence", {
|
||||
evidence_id: capture.source.evidence_id,
|
||||
component: "recovered_ir",
|
||||
offset: decoded.next_offset,
|
||||
limit: 64,
|
||||
});
|
||||
assert.deepEqual(next.view, decoded.view);
|
||||
assert.equal(
|
||||
next.text,
|
||||
view.text.slice(decoded.next_offset, decoded.next_offset + 64),
|
||||
);
|
||||
assert.equal(
|
||||
(await read("snapshot", { entry: "0xea6d0" })).snapshot,
|
||||
snapshot,
|
||||
);
|
||||
sourceReview = {
|
||||
source_bytes: Buffer.byteLength(text),
|
||||
decoded_graph_bytes: Buffer.byteLength(view.text),
|
||||
logical_graph_equals_diagnostic_golden: true,
|
||||
ordinary_record_and_calls_equal_golden: true,
|
||||
former_projection_frontier_readable: formerFrontier,
|
||||
complete_evidence_views_verified: true,
|
||||
};
|
||||
await call("close_binary", {});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({
|
||||
direct,
|
||||
node_version: process.version,
|
||||
calls,
|
||||
failures,
|
||||
sourceReview,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Real public MCP plus complete retained producer-gold comparison; no model or runtime/machine-equivalence claim.",
|
||||
}),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({ output, calls: calls.length, failures, sourceReview }),
|
||||
);
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,219 +0,0 @@
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output, scenario = "zlib"] = process.argv.slice(2);
|
||||
assert.ok(["zlib", "sqlite-partial", "sqlite-encoded"].includes(scenario));
|
||||
const verifier = createVerifierRun();
|
||||
const client = new Client({ name: "rea-codemode-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
let stderr = "";
|
||||
transport.stderr?.on("data", (part) => {
|
||||
stderr = (stderr + part).slice(-8192);
|
||||
});
|
||||
const calls = [];
|
||||
const call = async (name, arguments_ = {}) => {
|
||||
const response = await client.callTool(
|
||||
{ name, arguments: arguments_ },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(response._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(response.content[0].text);
|
||||
calls.push({ name, arguments: arguments_, response: value });
|
||||
assert.equal(value.ok ?? true, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
let failure;
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
assert.deepEqual(catalog.tools.map((tool) => tool.name).sort(), [
|
||||
"close_binary",
|
||||
"execute_javascript",
|
||||
"open_binary",
|
||||
"reference",
|
||||
]);
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
const opened = await call("open_binary", { path: binary });
|
||||
if (scenario === "sqlite-encoded") {
|
||||
for (const entry of ["0xb90e0", "0xb3d00"]) {
|
||||
const result = await call("execute_javascript", {
|
||||
code: `
|
||||
const s=await rea.snapshot({entry:${JSON.stringify(entry)}});if(!s.ok)throw new Error(JSON.stringify(s));
|
||||
const c=await rea.code({snapshot:s.data.snapshot,limit:4});if(!c.ok)throw new Error(JSON.stringify(c));
|
||||
const operations=await rea.operations({snapshot:s.data.snapshot,limit:1,fields:["opcode"]});
|
||||
const graph=await rea.blocks({snapshot:s.data.snapshot,limit:1});
|
||||
return {snapshot:s.data,code:c.data,operations,graph};`,
|
||||
});
|
||||
assert.equal(result.value.snapshot.facets.code.availability, "available");
|
||||
assert.ok(result.value.snapshot.source.bytes <= 8 * 1024 * 1024);
|
||||
assert.ok(result.value.code.total_lines > 1000);
|
||||
assert.ok(result.value.code.text.length > 0);
|
||||
assert.equal(
|
||||
result.value.code.evidence_id,
|
||||
result.value.snapshot.source.evidence_id,
|
||||
);
|
||||
if (entry === "0xb3d00") {
|
||||
assert.equal(result.value.graph.ok, true);
|
||||
assert.equal(result.value.operations.ok, true);
|
||||
assert.equal(result.value.operations.data.total, 5164);
|
||||
assert.equal(
|
||||
result.value.snapshot.facets.operations.scope,
|
||||
"complete-for-recovered-graph",
|
||||
);
|
||||
} else {
|
||||
assert.equal(result.value.snapshot.coverage.blocks, null);
|
||||
assert.equal(
|
||||
result.value.snapshot.facets.operations.scope,
|
||||
"bounded-call-rooted-definitions",
|
||||
);
|
||||
assert.equal(result.value.operations.ok, true);
|
||||
assert.equal(result.value.graph.ok, false);
|
||||
assert.match(
|
||||
result.value.graph.error.details.native_error,
|
||||
/encoded document/u,
|
||||
);
|
||||
assert.equal(
|
||||
result.value.graph.error.details.evidence_id,
|
||||
result.value.snapshot.source.evidence_id,
|
||||
);
|
||||
}
|
||||
}
|
||||
} else if (scenario === "sqlite-partial") {
|
||||
assert.equal(opened.context.analysis_model, "experimental-native-ir-v3");
|
||||
const result = await call("execute_javascript", {
|
||||
code: `
|
||||
const read=async(p)=>{const r=await p;if(!r.ok)throw new Error(JSON.stringify(r));return r.data;};
|
||||
const names=await read(rea.functions({query:"sqlite3Fts3ErrMsg",exact:true}));
|
||||
const entries=await read(rea.functions({query:"sqlite3VdbeExec",exact:true}));
|
||||
const entry=entries.items.find(fn=>fn.address==="0xea6d0");if(!entry)throw new Error("fixture entry missing");
|
||||
const snap=await read(rea.snapshot({entry:entry.address}));
|
||||
const code=await read(rea.code({snapshot:snap.snapshot,limit:4}));
|
||||
const graph=await rea.blocks({snapshot:snap.snapshot,limit:1});
|
||||
const evidence=await read(rea.evidence({evidence_id:snap.source.evidence_id,limit:512}));
|
||||
return {names,snap,code,graph,evidence_source:evidence.source};`,
|
||||
});
|
||||
assert.deepEqual(result.value.names.items.map((fn) => fn.address).sort(), [
|
||||
"0x268f0",
|
||||
"0x5efc0",
|
||||
]);
|
||||
assert.equal(result.value.snap.facets.code.availability, "available");
|
||||
assert.equal(
|
||||
result.value.snap.facets.recovered_ir.availability,
|
||||
"unavailable",
|
||||
);
|
||||
assert.equal(result.value.snap.coverage.blocks, null);
|
||||
assert.equal(
|
||||
result.value.snap.facets.operations.scope,
|
||||
"bounded-call-rooted-definitions",
|
||||
);
|
||||
assert.ok(
|
||||
result.value.code.text.length > 0 && result.value.code.total_lines > 1000,
|
||||
);
|
||||
assert.equal(result.value.graph.ok, false);
|
||||
assert.equal(result.value.graph.error.code, "capability_unavailable");
|
||||
assert.equal(
|
||||
result.value.graph.error.details.evidence_id,
|
||||
result.value.snap.source.evidence_id,
|
||||
);
|
||||
assert.equal(
|
||||
result.value.evidence_source.sha256,
|
||||
result.value.snap.source.sha256,
|
||||
);
|
||||
} else {
|
||||
await call("reference", { query: "relations" });
|
||||
const result = await call("execute_javascript", {
|
||||
code: `
|
||||
const found=await rea.strings({query:"incorrect header check",limit:1});
|
||||
if(!found.ok) throw new Error(JSON.stringify(found));
|
||||
const literal=found.data.items[0].address;
|
||||
const refs=await rea.references({addresses:[literal]});
|
||||
const entry=refs.data.results[0].items[0].fromFunction.address;
|
||||
const snap=await rea.snapshot({entry}); if(!snap.ok)throw new Error(JSON.stringify(snap));
|
||||
const s=snap.data.snapshot;
|
||||
const seed=await rea.operations({snapshot:s,constant:literal});
|
||||
const uses=await rea.relations({snapshot:s,kind:"uses",seeds:seed.data.items.map(x=>x.id)});
|
||||
const consumers=await rea.operations({snapshot:s,ids:uses.data.items.map(x=>x.use)});
|
||||
const incoming=await rea.relations({snapshot:s,kind:"incoming",seeds:consumers.data.items.map(x=>x.block)});
|
||||
const slice=await rea.slice({snapshot:s,seeds:incoming.data.items.map(x=>x.operation).filter(Boolean),fields:["statement"]});
|
||||
return {literal,entry,snapshot:s,consumer_blocks:[...new Set(consumers.data.items.map(x=>x.block))],selected:slice.data.operations.length,projected:slice.data.operations.every(op=>!("inputs" in op)&&typeof op.origin_address==="string"),guards:slice.data.operations.map(x=>x.statement),coverage:slice.data.coverage,evidence_id:slice.data.evidence_id};`,
|
||||
});
|
||||
assert.equal(result.value.selected, 19);
|
||||
assert.equal(result.value.projected, true);
|
||||
assert.equal(result.value.coverage.status, "partial");
|
||||
assert.ok(result.value.guards.some((s) => s.includes("/ #0x1f")));
|
||||
if (opened.context.analysis_model === "experimental-native-ir-v3") {
|
||||
const native = await call("execute_javascript", {
|
||||
code: `const s=${JSON.stringify(result.value.snapshot)};
|
||||
const read=async(p)=>{const r=await p;if(!r.ok)throw new Error(JSON.stringify(r));return r.data;};
|
||||
const blocks=await read(rea.blocks({snapshot:s,ids:${JSON.stringify(result.value.consumer_blocks)}}));
|
||||
const controls=await read(rea.relations({snapshot:s,kind:"controllers",seeds:blocks.items.map(b=>b.id)}));
|
||||
const guards=await read(rea.operations({snapshot:s,block_ids:controls.items.map(r=>r.controller),kind:"branch",fields:["inputs"]}));
|
||||
const phi=await read(rea.operations({snapshot:s,kind:"phi",fields:["inputs","output"],limit:2}));
|
||||
const values=await read(rea.values({snapshot:s,ids:phi.items.flatMap(o=>o.inputs.map(i=>i.value_id)).filter(Boolean)}));
|
||||
return {blocks,controls,guards,phi,values};`,
|
||||
});
|
||||
assert.equal(native.value.controls.total, 2);
|
||||
assert.deepEqual(
|
||||
native.value.guards.items.map((op) => op.origin_address).sort(),
|
||||
["0xd9ec", "0xda32"],
|
||||
);
|
||||
assert.ok(
|
||||
native.value.phi.items.every((op) =>
|
||||
op.inputs.every(
|
||||
(input) => typeof input.phi_predecessor_edge === "string",
|
||||
),
|
||||
),
|
||||
);
|
||||
assert.equal(
|
||||
native.value.blocks.coverage.control_flow,
|
||||
"complete-for-recovered-graph",
|
||||
);
|
||||
assert.ok(native.value.values.items.length > 0);
|
||||
}
|
||||
const machine = await call("execute_javascript", {
|
||||
code: 'const r=await rea.instructions({windows:[{address:"0xd97e",count:1},{address:"0xe26d",count:2},{address:"0xdeadbeef",count:1}]});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;',
|
||||
});
|
||||
assert.equal(machine.value.results[0].items[0].bytes, "0f87e9080000");
|
||||
assert.match(machine.value.results[0].items[0].text, /e26d/u);
|
||||
assert.equal(machine.value.results[1].items[1].address, "0xe270");
|
||||
assert.match(machine.value.results[1].items[1].text, /d9c9/u);
|
||||
assert.equal(machine.value.results[2].ok, false);
|
||||
}
|
||||
await call("close_binary");
|
||||
} catch (error) {
|
||||
failure = error.stack ?? String(error);
|
||||
} finally {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({ calls, failure, stderr, lifecycle }, null, 2),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({
|
||||
output,
|
||||
calls: calls.length,
|
||||
failure,
|
||||
cleanup_verified: true,
|
||||
}),
|
||||
);
|
||||
if (failure) process.exitCode = 1;
|
||||
@@ -1,146 +0,0 @@
|
||||
// Selected real SQLite wide-PHI facts through either public MCP adapter.
|
||||
import assert from "node:assert/strict";
|
||||
import { readFile, writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output, goldenPath] = process.argv.slice(2);
|
||||
const golden = JSON.parse(await readFile(goldenPath, "utf8")).functions[0];
|
||||
const id = "0xea998:14649",
|
||||
indices = [0, 1, 77, 154, 155];
|
||||
const expected = golden.atomic_projection.operations.find((op) => op.id === id);
|
||||
assert.equal(expected.inputs.length, 156);
|
||||
const values = new Map(golden.atomic_projection.values.map((v) => [v.id, v]));
|
||||
const run = createVerifierRun(),
|
||||
calls = [],
|
||||
failures = [];
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const client = new Client({ name: "rea-operands-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const call = async (name, input) => {
|
||||
const reply = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
const value = JSON.parse(reply.content[0].text);
|
||||
calls.push({
|
||||
name,
|
||||
input,
|
||||
value,
|
||||
content_bytes: Buffer.byteLength(reply.content[0].text),
|
||||
});
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
};
|
||||
const read = async (name, input) => {
|
||||
const value = direct
|
||||
? await call(`facts_${name}`, input)
|
||||
: (
|
||||
await call("execute_javascript", {
|
||||
code: `return await rea.${name}(${JSON.stringify(input)});`,
|
||||
})
|
||||
).value;
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value.data;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
await call("open_binary", { path: binary });
|
||||
const capture = await read("snapshot", { entry: "0xea6d0" });
|
||||
const snapshot = capture.snapshot;
|
||||
const summary = await read("operations", { snapshot, ids: [id] });
|
||||
assert.equal(summary.items[0].input_count, 156);
|
||||
assert.equal(summary.items[0].definition_count, 156);
|
||||
assert.equal(Object.hasOwn(summary.items[0], "inputs"), false);
|
||||
const slice = await read("slice", {
|
||||
snapshot,
|
||||
seeds: [id],
|
||||
max_operations: 1,
|
||||
});
|
||||
assert.equal(slice.operations[0].input_count, 156);
|
||||
assert.ok(Buffer.byteLength(JSON.stringify(slice)) < 32768);
|
||||
const legacy = await read("operations", {
|
||||
snapshot,
|
||||
ids: [id],
|
||||
fields: ["inputs"],
|
||||
});
|
||||
assert.deepEqual(legacy.items[0].inputs, expected.inputs);
|
||||
const page = await read("relations", {
|
||||
snapshot,
|
||||
kind: "operands",
|
||||
seeds: [id],
|
||||
offset: 154,
|
||||
limit: 10,
|
||||
});
|
||||
assert.equal(page.total, 156);
|
||||
assert.equal(page.next_offset, null);
|
||||
assert.deepEqual(
|
||||
page.items.map((r) => r.input_index),
|
||||
[154, 155],
|
||||
);
|
||||
const selected = await read("relations", {
|
||||
snapshot,
|
||||
kind: "operands",
|
||||
seeds: [id, id, "0xffff:0"],
|
||||
input_indices: [...indices, 0, 156],
|
||||
});
|
||||
assert.deepEqual(selected.unretained_seeds, ["0xffff:0"]);
|
||||
assert.deepEqual(
|
||||
selected.items,
|
||||
indices.map((i) => ({
|
||||
operation: id,
|
||||
...expected.inputs[i],
|
||||
definition: values.get(expected.inputs[i].value_id).definition,
|
||||
})),
|
||||
);
|
||||
assert.equal(selected.total, indices.length);
|
||||
assert.equal(selected.next_offset, null);
|
||||
assert.equal(selected.evidence_id, capture.source.evidence_id);
|
||||
const operandValues = await read("values", {
|
||||
snapshot,
|
||||
ids: selected.items.map((r) => r.value_id),
|
||||
});
|
||||
assert.deepEqual(
|
||||
operandValues.items,
|
||||
[...new Set(selected.items.map((r) => r.value_id))].map((id) =>
|
||||
values.get(id),
|
||||
),
|
||||
);
|
||||
assert.equal(
|
||||
(await read("snapshot", { entry: "0xea6d0" })).snapshot,
|
||||
snapshot,
|
||||
);
|
||||
await call("close_binary", {});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({
|
||||
direct,
|
||||
calls,
|
||||
failures,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Real public MCP facts versus retained native producer gold; no runtime or model-token claim.",
|
||||
}),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(JSON.stringify({ output, calls: calls.length, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,179 +0,0 @@
|
||||
// Matched public MCP reads, without a model or changes to the native engine.
|
||||
import assert from "node:assert/strict";
|
||||
import { readFile, writeFile } from "node:fs/promises";
|
||||
import { resolve, join } from "node:path";
|
||||
import { setTimeout as delay } from "node:timers/promises";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import { familyCounters } from "./family-counters.mjs";
|
||||
import { readNativeGraphEvidence } from "./native-ir-contract.mjs";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output, route, ordering, goldenPath] = process.argv.slice(2);
|
||||
assert.ok(["direct", "code"].includes(route));
|
||||
assert.ok(["sequential", "overlap"].includes(ordering));
|
||||
assert.equal(
|
||||
process.versions.node,
|
||||
(await readFile(resolve(".nvmrc"), "utf8")).trim(),
|
||||
);
|
||||
assert.ok(
|
||||
process.env.REA_ATOMIC_EVIDENCE_DIR,
|
||||
"Retain exact source for fidelity checks",
|
||||
);
|
||||
const golden = JSON.parse(await readFile(goldenPath, "utf8")).functions[0];
|
||||
const windows = [
|
||||
{ address: "0xf2360", count: 18 },
|
||||
{ address: "0xf2410", count: 22 },
|
||||
{ address: "0x288f0", count: 3 },
|
||||
];
|
||||
const run = createVerifierRun();
|
||||
const client = new Client({ name: "rea-overlap-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: {
|
||||
...process.env,
|
||||
REA_ATOMIC_ADAPTER: route === "direct" ? "direct" : "code",
|
||||
},
|
||||
stderr: "pipe",
|
||||
});
|
||||
const failures = [],
|
||||
phases = [],
|
||||
requests = [];
|
||||
const epoch = performance.now();
|
||||
async function tool(name, input) {
|
||||
const reply = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
const value = JSON.parse(reply.content[0].text);
|
||||
assert.equal(value.ok, true, JSON.stringify(value));
|
||||
return value;
|
||||
}
|
||||
async function read(name, input) {
|
||||
const record = { method: name, started_ms: performance.now() - epoch };
|
||||
requests.push(record);
|
||||
const response =
|
||||
route === "direct"
|
||||
? await tool(`facts_${name}`, input)
|
||||
: (
|
||||
await tool("execute_javascript", {
|
||||
code: `return await rea.${name}(${JSON.stringify(input)});`,
|
||||
})
|
||||
).value;
|
||||
record.finished_ms = performance.now() - epoch;
|
||||
record.elapsed_ms = record.finished_ms - record.started_ms;
|
||||
assert.equal(response.ok, true, JSON.stringify(response));
|
||||
return response.data;
|
||||
}
|
||||
async function measured(phase, work) {
|
||||
const before = await familyCounters(),
|
||||
started = performance.now();
|
||||
const value = await work();
|
||||
const elapsed_ms = performance.now() - started,
|
||||
after = await familyCounters();
|
||||
phases.push({
|
||||
phase,
|
||||
elapsed_ms,
|
||||
family_cpu_seconds: (after.cpu - before.cpu) / 1e6,
|
||||
cumulative_family_memory_peak_bytes: after.peak,
|
||||
});
|
||||
return value;
|
||||
}
|
||||
let observations;
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const opened = await tool("open_binary", { path: binary });
|
||||
assert.equal(
|
||||
opened.context.artifact_sha256,
|
||||
"2e6eef9a727f081f0d453b4e5e6cbd8b9ef8b6f86cbf7681cbad444d3b0b55c8",
|
||||
);
|
||||
const [snapshot, instructions] = await measured("fresh-lease", async () => {
|
||||
if (ordering === "sequential")
|
||||
return [
|
||||
await read("snapshot", { entry: "0xea6d0" }),
|
||||
await read("instructions", { windows }),
|
||||
];
|
||||
const pending = read("snapshot", { entry: "0xea6d0" });
|
||||
// Deliberately overlap requests, not native processes: the session owns admission.
|
||||
// Attach rejection handling immediately even if the second request fails.
|
||||
const pair = Promise.all([
|
||||
pending,
|
||||
(async () => {
|
||||
await delay(100);
|
||||
return read("instructions", { windows });
|
||||
})(),
|
||||
]);
|
||||
return pair;
|
||||
});
|
||||
assert.equal(snapshot.facets.recovered_ir.availability, "available");
|
||||
assert.equal(snapshot.coverage.operations, 21349);
|
||||
assert.equal(snapshot.coverage.omitted_definition_count, 0);
|
||||
assert.equal(instructions.results.length, windows.length);
|
||||
for (const item of instructions.results) assert.equal(item.ok, true);
|
||||
const warm = await measured("same-lease-reuse", async () => [
|
||||
await read("snapshot", { entry: "0xea6d0" }),
|
||||
await read("instructions", { windows }),
|
||||
]);
|
||||
assert.deepEqual(warm, [snapshot, instructions]);
|
||||
const source = await readFile(
|
||||
join(
|
||||
process.env.REA_ATOMIC_EVIDENCE_DIR,
|
||||
`${snapshot.source.evidence_id}.txt`,
|
||||
),
|
||||
"utf8",
|
||||
);
|
||||
const { atomic_facts: _archive, ...ordinary } =
|
||||
JSON.parse(source).functions[0];
|
||||
const { atomic_facts: expectedGraph, ...expectedOrdinary } = golden;
|
||||
assert.deepEqual(
|
||||
JSON.parse(readNativeGraphEvidence(source).text),
|
||||
expectedGraph,
|
||||
);
|
||||
assert.deepEqual(ordinary, expectedOrdinary);
|
||||
const first = requests[0],
|
||||
second = requests[1];
|
||||
assert.equal(second.started_ms < first.finished_ms, ordering === "overlap");
|
||||
observations = {
|
||||
snapshot,
|
||||
instructions,
|
||||
source_fidelity: "full-graph-and-ordinary-fields-match-golden",
|
||||
};
|
||||
await tool("close_binary", {});
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
try {
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({
|
||||
route,
|
||||
ordering,
|
||||
phases,
|
||||
requests,
|
||||
observations,
|
||||
failures,
|
||||
lifecycle,
|
||||
scope:
|
||||
"Real MCP, same pinned engine and fresh lease/cache per run; no model/token measurement. Request overlap is observed at the client; AtomicSession serializes provider admission. Source elapsed includes command supervision/retention, not isolated native execution. CPU deltas cover each whole phase, not overlapping per-request attribution. Peak includes file cache and is cumulative; warm peak is an upper bound. OS caches and host contention are uncontrolled.",
|
||||
}),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(JSON.stringify({ output, route, ordering, phases, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,137 +0,0 @@
|
||||
// Real public selected-definition reads; full diagnostic comparison is separate.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
const run = createVerifierRun(),
|
||||
calls = [],
|
||||
failures = [];
|
||||
const client = new Client({
|
||||
name: "rea-projection-verifier",
|
||||
version: "0.1.0",
|
||||
});
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const call = async (name, input) => {
|
||||
const result = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(result._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
calls.push({ name, input, value });
|
||||
return value;
|
||||
};
|
||||
const read = async (name, input) => {
|
||||
const result = direct
|
||||
? await call(`facts_${name}`, input)
|
||||
: await call("execute_javascript", {
|
||||
code: `return await rea.${name}(${JSON.stringify(input)});`,
|
||||
});
|
||||
assert.equal(result.ok, true, JSON.stringify(result));
|
||||
const response = direct ? result : result.value;
|
||||
assert.equal(response.ok, true, JSON.stringify(response));
|
||||
return response.data;
|
||||
};
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
assert.equal((await call("open_binary", { path: binary })).ok, true);
|
||||
const captured = await read("snapshot", { entry: "0xea6d0" });
|
||||
assert.equal(captured.facets.recovered_ir.availability, "unavailable");
|
||||
assert.equal(
|
||||
captured.facets.operations.scope,
|
||||
"bounded-call-rooted-definitions",
|
||||
);
|
||||
assert.equal(captured.facets.operations.count, 4096);
|
||||
assert.equal(captured.coverage.control_flow, "not-collected");
|
||||
assert.ok(captured.facets.operations.definition_frontier_count > 0);
|
||||
const snapshot = captured.snapshot;
|
||||
const selected = await read("calls", {
|
||||
snapshot,
|
||||
origin_addresses: ["0xf115c", "0xf238b"],
|
||||
fields: ["arguments", "target", "target_value_id"],
|
||||
});
|
||||
assert.equal(selected.total, 2);
|
||||
const values = await read("values", {
|
||||
snapshot,
|
||||
ids: [
|
||||
...new Set(
|
||||
selected.items.flatMap((r) => r.arguments.map((a) => a.value_id)),
|
||||
),
|
||||
],
|
||||
});
|
||||
assert.match(values.coverage.value_scope, /retained call-rooted/);
|
||||
const definitions = values.items.map((v) => v.definition).filter(Boolean);
|
||||
const bodies = await read("operations", {
|
||||
snapshot,
|
||||
ids: definitions,
|
||||
fields: ["opcode", "inputs", "output"],
|
||||
});
|
||||
assert.equal(bodies.total, definitions.length);
|
||||
assert.ok(bodies.items.every((r) => r.opcode === "CPUI_MULTIEQUAL"));
|
||||
assert.equal(bodies.evidence_id, captured.source.evidence_id);
|
||||
const slices = await read("slice", {
|
||||
snapshot,
|
||||
seeds: selected.items.map((r) => r.id),
|
||||
max_operations: 48,
|
||||
stop_kinds: ["load"],
|
||||
fields: ["opcode", "inputs", "output", "constants"],
|
||||
});
|
||||
assert.ok(slices.operations.length > 5);
|
||||
assert.equal(slices.coverage.memory_alias_dependencies, "unknown");
|
||||
const uses = await read("relations", {
|
||||
snapshot,
|
||||
kind: "uses",
|
||||
seeds: [bodies.items[0].id],
|
||||
});
|
||||
assert.match(uses.relation_coverage.scope, /selected-operations-only/);
|
||||
if (!direct) {
|
||||
const unavailable = await call("execute_javascript", {
|
||||
code: `return await rea.relations({snapshot:${JSON.stringify(snapshot)},kind:"controllers",seeds:[0]});`,
|
||||
});
|
||||
assert.equal(unavailable.value.ok, false);
|
||||
assert.equal(unavailable.value.error.code, "capability_unavailable");
|
||||
assert.equal(
|
||||
unavailable.value.error.details.evidence_id,
|
||||
captured.source.evidence_id,
|
||||
);
|
||||
}
|
||||
assert.equal(
|
||||
(await read("snapshot", { entry: "0xea6d0" })).snapshot,
|
||||
snapshot,
|
||||
);
|
||||
assert.equal((await call("close_binary", {})).ok, true);
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(run);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({ direct, calls, failures, lifecycle }),
|
||||
{
|
||||
flag: "wx",
|
||||
},
|
||||
);
|
||||
console.log(JSON.stringify({ output, calls: calls.length, failures }));
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,224 +0,0 @@
|
||||
// Real adapter/reference coverage, including a function beyond full IR capacity.
|
||||
import assert from "node:assert/strict";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { resolve } from "node:path";
|
||||
import { Client } from "@modelcontextprotocol/client";
|
||||
import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
|
||||
import Ajv2020 from "ajv/dist/2020.js";
|
||||
import {
|
||||
createVerifierRun,
|
||||
completeVerifierRun,
|
||||
assertVerifierCleanupComplete,
|
||||
} from "../../scripts/lib/verifier-run.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
const verifier = createVerifierRun(),
|
||||
calls = [],
|
||||
failures = [];
|
||||
const direct = process.env.REA_ATOMIC_ADAPTER === "direct";
|
||||
const client = new Client({ name: "rea-reference-verifier", version: "0.1.0" });
|
||||
const transport = new StdioClientTransport({
|
||||
command: process.execPath,
|
||||
args: [resolve("experiments/atomic-codemode/server.mjs")],
|
||||
env: { ...process.env },
|
||||
stderr: "pipe",
|
||||
});
|
||||
const call = async (name, input) => {
|
||||
const result = await client.callTool(
|
||||
{ name, arguments: input },
|
||||
{ timeout: 180000 },
|
||||
);
|
||||
assert.equal(result._meta?.provenance?.untrusted_content, true);
|
||||
const value = JSON.parse(result.content[0].text);
|
||||
calls.push({ name, input, value });
|
||||
return value;
|
||||
};
|
||||
const read = async (name, input) => {
|
||||
const r = direct
|
||||
? await call(`facts_${name}`, input)
|
||||
: await call("execute_javascript", {
|
||||
code: `const r=await rea.${name}(${JSON.stringify(input)});if(!r.ok)throw new Error(JSON.stringify(r));return r.data;`,
|
||||
});
|
||||
assert.equal(r.ok, true, JSON.stringify(r));
|
||||
return direct ? r.data : r.value;
|
||||
};
|
||||
const references = (addresses, scope = "function", extra = {}) =>
|
||||
read("references", {
|
||||
addresses,
|
||||
direction: "outgoing",
|
||||
scope,
|
||||
kinds: ["call"],
|
||||
fields: [],
|
||||
limit: 128,
|
||||
...extra,
|
||||
});
|
||||
try {
|
||||
await client.connect(transport);
|
||||
const catalog = await client.listTools();
|
||||
const ajv = new Ajv2020({ strict: false });
|
||||
for (const tool of catalog.tools) ajv.compile(tool.inputSchema);
|
||||
assert.equal((await call("open_binary", { path: binary })).ok, true);
|
||||
const wrapper = (await references(["0x65180"])).results[0];
|
||||
assert.equal(wrapper.total, 9);
|
||||
assert.equal(wrapper.coverage.function_scope, "established-by-rows");
|
||||
assert.ok(
|
||||
wrapper.items.some((r) => r.from === "0x6534b" && r.to === "0x29370"),
|
||||
);
|
||||
assert.deepEqual(
|
||||
Object.keys(wrapper.items[0]).sort(),
|
||||
["source_row", "from", "to", "kind"].sort(),
|
||||
);
|
||||
const snapshot = await read("snapshot", { entry: "0x65180" });
|
||||
const recovered = await read("calls", {
|
||||
snapshot: snapshot.snapshot,
|
||||
fields: ["target"],
|
||||
limit: 128,
|
||||
});
|
||||
assert.equal(recovered.total, 8);
|
||||
assert.ok(recovered.items.every((r) => r.origin_address !== "0x6534b"));
|
||||
const exact = (await references(["0x65180"], "address")).results[0];
|
||||
assert.equal(exact.total, 0);
|
||||
assert.equal(exact.coverage.excluded_scope_rows, 9);
|
||||
assert.equal(exact.source.evidence_id, wrapper.source.evidence_id);
|
||||
const sites = await references(["0x65211", "0xf238b"], "address");
|
||||
assert.deepEqual(
|
||||
sites.results.map((r) => r.items.map((f) => [f.from, f.to])),
|
||||
[[["0x65211", "0x64060"]], [["0xf238b", "0x296c0"]]],
|
||||
);
|
||||
const pages = [];
|
||||
let source;
|
||||
for (let offset = 0; offset !== null;) {
|
||||
const row = (await references(["0xea6d0"], "function", { offset }))
|
||||
.results[0];
|
||||
assert.equal(row.ok, true);
|
||||
assert.equal(row.total, 396);
|
||||
if (source) assert.equal(row.source.evidence_id, source);
|
||||
source = row.source.evidence_id;
|
||||
pages.push(...row.items);
|
||||
offset = row.next_offset;
|
||||
}
|
||||
assert.equal(pages.length, 396);
|
||||
assert.equal(new Set(pages.map((p) => p.source_row)).size, 396);
|
||||
assert.equal(new Set(pages.map((p) => p.to)).size, 171);
|
||||
assert.equal(pages.filter((p) => p.to === "0x296c0").length, 16);
|
||||
const unknown = (await references(["0xdeadbeef"])).results[0];
|
||||
assert.equal(unknown.coverage.function_scope, "unknown");
|
||||
assert.equal(unknown.coverage.discovery, "unknown");
|
||||
if (direct) {
|
||||
// The SDK rejects invalid direct arguments before the target-content handler.
|
||||
const rejected = await client.callTool({
|
||||
name: "facts_references",
|
||||
arguments: {
|
||||
addresses: ["0x1000"],
|
||||
direction: "incoming",
|
||||
scope: "function",
|
||||
},
|
||||
});
|
||||
calls.push({ name: "facts_references:invalid-scope", value: rejected });
|
||||
assert.equal(rejected.isError, true);
|
||||
assert.match(
|
||||
JSON.stringify(rejected.content),
|
||||
/Incoming references require exact address scope/,
|
||||
);
|
||||
} else {
|
||||
const bad = await call("execute_javascript", {
|
||||
code: 'return await rea.references({addresses:["0x1000"],direction:"incoming",scope:"function"});',
|
||||
});
|
||||
assert.equal(bad.value.ok, false);
|
||||
assert.equal(bad.value.error.code, "invalid_input");
|
||||
}
|
||||
if (!direct) {
|
||||
const composed = await call("execute_javascript", {
|
||||
code: 'let offset=0,n=0,formatter=0;const targets=new Set();do{const r=await rea.references({addresses:["0xea6d0"],direction:"outgoing",scope:"function",kinds:["call"],fields:[],offset,limit:128});if(!r.ok)throw new Error(JSON.stringify(r));const p=r.data.results[0];if(!p.ok)throw new Error(JSON.stringify(p));for(const x of p.items){n++;targets.add(x.to);if(x.to==="0x296c0")formatter++;}offset=p.next_offset;}while(offset!==null);return {references:n,distinct_targets:targets.size,formatter};',
|
||||
});
|
||||
assert.equal(composed.ok, true);
|
||||
assert.deepEqual(composed.value, {
|
||||
references: 396,
|
||||
distinct_targets: 171,
|
||||
formatter: 16,
|
||||
});
|
||||
assert.deepEqual(composed.evidence_ids, [source]);
|
||||
const vm = await read("snapshot", { entry: "0xea6d0" });
|
||||
assert.equal(vm.facets.code.availability, "available");
|
||||
assert.equal(vm.facets.recovered_ir.availability, "unavailable");
|
||||
assert.match(vm.facets.recovered_ir.error.details.native_error, /capacity/);
|
||||
assert.equal(vm.facets.calls.availability, "available");
|
||||
assert.equal(vm.facets.calls.count, 416);
|
||||
assert.deepEqual(vm.facets.calls.by_opcode, [
|
||||
{ opcode: "CPUI_CALL", known_target_count: 395, unknown_target_count: 0 },
|
||||
{
|
||||
opcode: "CPUI_CALLIND",
|
||||
known_target_count: 0,
|
||||
unknown_target_count: 21,
|
||||
},
|
||||
]);
|
||||
assert.equal(vm.facets.calls.coverage.machine_call_equivalence, "unknown");
|
||||
assert.equal(vm.facets.values.scope, "call-rooted-definitions");
|
||||
const selectedCalls = await read("calls", {
|
||||
snapshot: vm.snapshot,
|
||||
origin_addresses: ["0xf238b", "0xf115c"],
|
||||
fields: ["target", "target_value_id", "arguments"],
|
||||
});
|
||||
assert.equal(selectedCalls.total, 2);
|
||||
assert.ok(selectedCalls.items.every((c) => c.target?.offset === "0x296c0"));
|
||||
assert.ok(selectedCalls.coverage.operations_examined > 12000);
|
||||
const ids = [
|
||||
...new Set(
|
||||
selectedCalls.items
|
||||
.flatMap((c) => [
|
||||
c.target_value_id,
|
||||
...c.arguments.map((a) => a.value_id),
|
||||
])
|
||||
.filter(Boolean),
|
||||
),
|
||||
];
|
||||
const boundary = await read("values", { snapshot: vm.snapshot, ids });
|
||||
assert.equal(boundary.missing.length, 0);
|
||||
assert.equal(boundary.coverage.control_flow, "not-collected");
|
||||
for (const row of selectedCalls.items)
|
||||
assert.ok(
|
||||
boundary.items.some(
|
||||
(v) =>
|
||||
v.id === row.target_value_id &&
|
||||
v.space === "fspec" &&
|
||||
v.is_annotation,
|
||||
),
|
||||
);
|
||||
const unavailable = await call("execute_javascript", {
|
||||
code: `return await rea.blocks({snapshot:${JSON.stringify(vm.snapshot)},limit:1});`,
|
||||
});
|
||||
assert.equal(unavailable.value.ok, false);
|
||||
assert.equal(unavailable.value.error.code, "capability_unavailable");
|
||||
assert.equal(
|
||||
(await read("snapshot", { entry: "0xea6d0" })).snapshot,
|
||||
vm.snapshot,
|
||||
);
|
||||
}
|
||||
assert.equal((await call("close_binary", {})).ok, true);
|
||||
} catch (error) {
|
||||
failures.push(error.stack ?? String(error));
|
||||
} finally {
|
||||
try {
|
||||
await client.close();
|
||||
await transport.close();
|
||||
} catch (error) {
|
||||
failures.push(String(error));
|
||||
}
|
||||
}
|
||||
const lifecycle = await completeVerifierRun(verifier);
|
||||
assertVerifierCleanupComplete(lifecycle.process_lineage);
|
||||
await writeFile(
|
||||
output,
|
||||
JSON.stringify({ direct, calls, failures, lifecycle }),
|
||||
{ flag: "wx" },
|
||||
);
|
||||
console.log(
|
||||
JSON.stringify({
|
||||
output,
|
||||
direct,
|
||||
calls: calls.length,
|
||||
failures,
|
||||
cleanup_verified: true,
|
||||
}),
|
||||
);
|
||||
if (failures.length) process.exitCode = 1;
|
||||
@@ -1,70 +0,0 @@
|
||||
// Real provider regression for shared SDK cache ownership, without MCP's copies.
|
||||
import assert from "node:assert/strict";
|
||||
import { createHash } from "node:crypto";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { AtomicSession } from "./api.mjs";
|
||||
|
||||
const [binary, output] = process.argv.slice(2);
|
||||
assert.ok(binary && output, "Usage: verify-sdk.mjs ZLIB OUTPUT");
|
||||
const session = new AtomicSession();
|
||||
const call = async (name, input) => {
|
||||
const result = await session.call(name, input);
|
||||
assert.equal(result.ok, true, JSON.stringify(result));
|
||||
return result.data;
|
||||
};
|
||||
let report;
|
||||
try {
|
||||
await session.open(binary);
|
||||
const first = await call("strings", { query: "incorrect header check" });
|
||||
assert.equal(first.total, 1);
|
||||
const original = structuredClone(first);
|
||||
assert.throws(() => {
|
||||
first.items[0].value = "changed by SDK caller";
|
||||
}, TypeError);
|
||||
assert.throws(() => {
|
||||
first.items[0].string.byte_length = 1;
|
||||
}, TypeError);
|
||||
// This array belongs to the reply, so changing it must also be harmless.
|
||||
first.items.length = 0;
|
||||
const next = await call("strings", { query: "incorrect header check" });
|
||||
assert.deepEqual(next, original);
|
||||
const functions = await call("functions", { query: "inflate", exact: true });
|
||||
assert.ok(functions.total >= 1);
|
||||
assert.throws(() => {
|
||||
functions.items[0].aliases.push("foreign alias");
|
||||
}, TypeError);
|
||||
const nextFunctions = await call("functions", {
|
||||
addresses: functions.items.map((item) => item.address),
|
||||
});
|
||||
assert.deepEqual(nextFunctions.items, functions.items);
|
||||
let offset = 0,
|
||||
text = "";
|
||||
do {
|
||||
const page = await call("evidence", {
|
||||
evidence_id: next.source.evidence_id,
|
||||
offset,
|
||||
limit: 16384,
|
||||
});
|
||||
assert.throws(() => {
|
||||
page.source.arguments[0] = "foreign operation";
|
||||
}, TypeError);
|
||||
text += page.text;
|
||||
offset = page.next_offset;
|
||||
} while (offset !== null);
|
||||
assert.equal(
|
||||
createHash("sha256").update(text).digest("hex"),
|
||||
next.source.sha256,
|
||||
);
|
||||
report = {
|
||||
context: session.context,
|
||||
source: next.source,
|
||||
cache_unchanged_after_caller_mutation: true,
|
||||
complete_source_digest_verified: true,
|
||||
scope:
|
||||
"Real SDK/provider reads. No model/token or runtime feasibility claim.",
|
||||
};
|
||||
} finally {
|
||||
await session.close();
|
||||
}
|
||||
await writeFile(output, JSON.stringify(report));
|
||||
console.log(JSON.stringify({ output, ...report }));
|
||||
Loaded 100 of 131 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user