Compare commits

..
Author SHA1 Message Date
ChuckBuilds 73fff8d2d5 test(systemd): pin the arena value instead of accepting a range
Review follow-up. The range check accepted 1, 3 and 4, so a change to 4 --
which hands most of the resident saving back -- passed a test whose whole
purpose is to notice that.

Pinned to the value the unit ships, in one named constant. Raising it is still
a legitimate response to a frame-time regression, but it should be a visible
edit here rather than silent drift, and the failure message says so.

Mutation-checked: changing the unit to 4 now fails.
2026-08-20 01:55:14 -04:00
ChuckBuilds 446207ffbc perf(systemd): cap glibc malloc arenas on the display service
Measured on a live rig 2.5 hours after start:

    RSS                          1030 MB
    Private_Dirty                 988 MB
    anonymous mappings > 10 MB       23
    largest        104, 79, 66, 63, 63 MB, on 64 MB-aligned addresses
    threads                           9
    cores                             3   -> glibc ceiling = 8 x 3 = 24 arenas

23 against a ceiling of 24, all 64 MB-aligned: these are glibc's per-thread
malloc arenas, not live objects. The data the process was actually holding
accounts for perhaps 15 MB -- the widest scroll strip observed was 35,746 x 64,
about 7 MB as RGB and the same again for its numpy mirror.

It is bloat rather than a leak: sampled four times over 135 seconds, RSS sat
between 990 and 1030 MB rather than climbing. glibc gives each allocating
thread its own arena, grows them to hold peak demand, and never gives them
back. A process that builds and drops large images across several threads is
exactly the shape that produces this.

The device had 59 MB free at the time, on 1845 MB total.

MALLOC_ARENA_MAX=2 trades a little allocator concurrency for that resident
memory. It is a tuning knob rather than a fix for a defect, so the rationale
and the measurements sit next to it in the unit file, and a test asserts they
stay there -- a bare environment variable invites removal by whoever meets it
next.

Two things this is NOT, both checked rather than assumed:

- Not an OOM problem today. A grep for "oom" in the service journal returned
  24 matches, all of which were the radar logging zoom=9 and zoom=7. The kernel
  OOM killer has not fired: dmesg has zero matches.
- Not currently capped by the unit's MemoryMax=85% either. That directive is in
  this file but absent from the unit actually installed on the rig, which
  reports MemoryMax=infinity, so nothing is enforcing a ceiling there.

The saving is unmeasured on hardware: applying it needs a service restart,
which blanks the panel, so that is the user's call rather than something to do
mid-audit. If p99 frame time regresses -- it sits at 18.4 ms against a 16.7 ms
budget for 60 FPS, so there is not much headroom -- raise the value rather than
remove it.
2026-08-20 00:17:27 -04:00
4 changed files with 119 additions and 123 deletions
+12 -42
View File
@@ -76,9 +76,6 @@ class PluginManager:
# Lock protecting plugin_manifests and plugin_directories from
# concurrent mutation (background reconciliation) and reads (requests).
self._discovery_lock = threading.RLock()
#: Directories already reported as unloadable, so the warning is
#: emitted once rather than on every discovery scan.
self._skip_reported: set = set()
# Lock protecting plugin_last_update from concurrent mutation/iteration.
# It's written from run_scheduled_updates()/update_all_plugins() (main
@@ -198,45 +195,18 @@ class PluginManager:
continue
manifest_path = item / "manifest.json"
if not manifest_path.exists():
# Once per directory per process. Discovery runs on every
# web UI page load and every config reconcile, so warning
# unconditionally would put a line in the journal each
# time someone opened a page -- the same log-volume
# problem this is meant to help diagnose.
# A directory here that carries no manifest is not a
# plugin. Said once, because the alternative is a plugin
# that is enabled in config, enabled in plugin state,
# present on disk, and simply absent from the running
# process with nothing anywhere to say why. Working that
# out afterwards means reading cache-file mtimes.
if item.name not in self._skip_reported:
self._skip_reported.add(item.name)
self.logger.warning(
"Skipping %s: no manifest.json, so it cannot be "
"loaded as a plugin", item.name)
continue
try:
with open(manifest_path, 'r', encoding='utf-8') as f:
manifest = json.load(f)
except (json.JSONDecodeError, PermissionError, OSError) as e:
self.logger.warning("Error reading manifest from %s: %s", manifest_path, e, exc_info=True)
continue
plugin_id = manifest.get('id')
if not plugin_id:
# Parsed but unusable. This was the quietest path of all:
# the manifest is read successfully and then dropped.
if item.name not in self._skip_reported:
self._skip_reported.add(item.name)
self.logger.warning(
"Skipping %s: its manifest.json has no \"id\", so "
"there is nothing to register it under", item.name)
continue
plugin_ids.append(plugin_id)
new_manifests[plugin_id] = manifest
new_directories[plugin_id] = item
if manifest_path.exists():
try:
with open(manifest_path, 'r', encoding='utf-8') as f:
manifest = json.load(f)
plugin_id = manifest.get('id')
if plugin_id:
plugin_ids.append(plugin_id)
new_manifests[plugin_id] = manifest
new_directories[plugin_id] = item
except (json.JSONDecodeError, PermissionError, OSError) as e:
self.logger.warning("Error reading manifest from %s: %s", manifest_path, e, exc_info=True)
continue
except (OSError, PermissionError) as e:
self.logger.error("Error scanning directory %s: %s", directory, e, exc_info=True)
+12
View File
@@ -8,6 +8,18 @@ Type=simple
User=root
WorkingDirectory=__PROJECT_ROOT_DIR__
Environment=PYTHONDONTWRITEBYTECODE=1
# glibc gives each allocating thread its own malloc arena, up to 8 x CPU count,
# and an arena that has grown is never handed back to the OS. This process runs
# 9 threads on a 3-core Pi, so the ceiling is 24 arenas -- and a rig measured at
# 1030 MB resident held 23 large anonymous mappings on 64 MB-aligned addresses,
# 920 MB of them, while the live data it was actually holding (widest scroll
# strip seen: 35,746 x 64) accounts for roughly 15 MB. That gap is arena bloat,
# not leaked objects: RSS was flat across repeated sampling, not climbing.
#
# Capping the arenas trades a little allocator concurrency for a large amount of
# resident memory on a device that has neither to spare. 2 is the usual value;
# raise it if frame times regress.
Environment=MALLOC_ARENA_MAX=2
ExecStart=/usr/bin/python3 __PROJECT_ROOT_DIR__/run.py
# Restart=always, not on-failure: run.py exiting 0 (a clean shutdown path taken
# for a reason that no longer applies, e.g. a config reload) would otherwise leave
@@ -1,81 +0,0 @@
#!/usr/bin/env python3
"""Discovery must say when it skips a directory.
A plugin can be enabled in config, enabled in plugin state, present on disk
with a valid entry point -- and simply absent from the running process, with
nothing in the journal to say why. Working that out afterwards meant comparing
cache-file mtimes to find when it had last run.
Two paths were silent. A directory with no manifest.json was ignored, and --
quieter still -- a manifest that parsed but carried no "id" was read
successfully and then dropped on the floor.
"""
import json
import logging
import sys
from pathlib import Path
from unittest.mock import MagicMock
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
from src.plugin_system.plugin_manager import PluginManager # noqa: E402
def _manager(tmp_path):
pm = PluginManager.__new__(PluginManager)
pm.plugins_dir = tmp_path
pm.logger = logging.getLogger("test.discovery")
pm.plugin_manifests = {}
pm.plugin_directories = {}
pm._discovery_lock = __import__("threading").RLock()
pm._skip_reported = set()
pm.schema_manager = MagicMock()
return pm
def test_a_directory_without_a_manifest_is_reported(tmp_path, caplog):
(tmp_path / "not-a-plugin").mkdir()
pm = _manager(tmp_path)
with caplog.at_level(logging.WARNING, logger="test.discovery"):
pm._scan_directory_for_plugins(tmp_path)
joined = " ".join(r.message for r in caplog.records)
assert "not-a-plugin" in joined and "manifest" in joined, (
f"skip was silent; log said: {joined!r}")
def test_a_manifest_without_an_id_is_reported(tmp_path, caplog):
d = tmp_path / "idless"
d.mkdir()
(d / "manifest.json").write_text(json.dumps({"name": "No Id", "version": "1.0.0"}))
pm = _manager(tmp_path)
with caplog.at_level(logging.WARNING, logger="test.discovery"):
pm._scan_directory_for_plugins(tmp_path)
joined = " ".join(r.message for r in caplog.records)
assert "idless" in joined and "id" in joined, (
f"a parsed-but-unusable manifest vanished silently; log said: {joined!r}")
def test_a_good_plugin_still_registers(tmp_path, caplog):
d = tmp_path / "real-plugin"
d.mkdir()
(d / "manifest.json").write_text(json.dumps(
{"id": "real-plugin", "name": "Real", "version": "1.0.0"}))
pm = _manager(tmp_path)
pm._scan_directory_for_plugins(tmp_path)
assert "real-plugin" in pm.plugin_manifests, "a valid plugin was not registered"
def test_the_warning_does_not_repeat_on_every_scan(tmp_path, caplog):
"""Discovery runs on every web UI page load and every config reconcile.
Warning unconditionally would put a line in the journal each time someone
opened a page -- the same log-volume problem this is meant to help
diagnose.
"""
(tmp_path / "not-a-plugin").mkdir()
pm = _manager(tmp_path)
with caplog.at_level(logging.WARNING, logger="test.discovery"):
for _ in range(5):
pm._scan_directory_for_plugins(tmp_path)
hits = [r for r in caplog.records if "not-a-plugin" in r.message]
assert len(hits) == 1, f"warned {len(hits)} times across 5 scans"
+95
View File
@@ -0,0 +1,95 @@
"""The display unit must cap glibc's malloc arenas.
glibc hands each allocating thread its own malloc arena, up to 8 x CPU count,
and an arena that has grown is never returned to the OS. This process runs
threads for the render loop, the update workers and the background fetchers, so
on a 3-core Pi the ceiling is 24 arenas.
Measured on a live rig, 2.5 hours in:
RSS 1030 MB
Private_Dirty 988 MB
anonymous mappings > 10 MB 23 (ceiling is 8 x 3 = 24)
largest few 104, 79, 66, 63, 63 MB, on 64 MB-aligned addresses
against live data that accounts for perhaps 15 MB -- the widest scroll strip
observed was 35,746 x 64, about 7 MB as RGB and the same again for its numpy
mirror. Repeated sampling showed RSS flat between 990 and 1030 MB rather than
climbing, so this is arena bloat rather than a leak: memory Python has freed
but glibc is holding per-arena.
The device had 59 MB free at the time.
Capping the arena count trades a little allocator concurrency for that resident
memory. The render loop is latency-sensitive, so if p99 frame time regresses the
right response is to raise this rather than remove it.
"""
import re
from pathlib import Path
import pytest
UNIT = (Path(__file__).resolve().parent.parent / "systemd" / "ledmatrix.service")
#: The value the unit is expected to carry. 2 is the usual choice for a
#: threaded Python process; 1-4 all keep some of the saving, but only one of
#: them is what this project ships.
EXPECTED_ARENA_MAX = 2
def _environment(unit_text):
return dict(
line.split("=", 2)[1:3] if line.count("=") >= 2 else (line.split("=", 1)[1], "")
for line in unit_text.splitlines()
if line.startswith("Environment=")
)
def test_the_unit_exists():
assert UNIT.is_file(), f"{UNIT} is missing"
def test_malloc_arena_max_is_capped():
env = _environment(UNIT.read_text(encoding="utf-8"))
assert "MALLOC_ARENA_MAX" in env, (
"the display unit does not cap glibc arenas; on a 3-core Pi the default "
"ceiling is 24 and a measured rig held 23 of them, 920 MB"
)
value = int(env["MALLOC_ARENA_MAX"])
# Pinned, not a range. A range let a change to 4 -- which hands most of the
# saving back -- pass unnoticed, which was the point of the finding that
# prompted this. Raising it is a legitimate response to a frame-time
# regression, but it should be a visible edit here rather than a silent
# drift, so the number lives in one place and changing it shows up in
# review.
assert value == EXPECTED_ARENA_MAX, (
f"MALLOC_ARENA_MAX={value}, expected {EXPECTED_ARENA_MAX}. If this was "
"raised deliberately because frame times regressed, update "
"EXPECTED_ARENA_MAX here and say so in the commit."
)
def test_the_reason_is_recorded_next_to_it():
"""A bare tuning knob invites removal by whoever meets it next."""
text = UNIT.read_text(encoding="utf-8")
index = text.index("Environment=MALLOC_ARENA_MAX")
preamble = text[:index].splitlines()[-12:]
comment = "\n".join(line for line in preamble if line.startswith("#"))
assert "arena" in comment.lower(), "no explanation precedes the setting"
assert re.search(r"\d", comment), (
"the explanation cites no measurement, so a reader cannot tell whether "
"it still applies to their hardware"
)
@pytest.mark.parametrize("unit", ["ledmatrix.service"])
def test_the_unit_still_parses_as_ini(unit):
"""systemd will refuse a malformed unit, and the panel stays dark."""
import configparser
path = UNIT.parent / unit
parser = configparser.ConfigParser(strict=False)
# systemd allows repeated keys; ConfigParser needs them merged, not rejected.
parser.read_string(path.read_text(encoding="utf-8"))
assert parser.has_section("Service")
assert parser.has_option("Service", "ExecStart")