[Feat] (TraceApp, CI): pass arbitrary env vars through the retrace lane

- PLAN-B.md §8.2 and appendix B add a batch of new runtime switches
  (MOBILEGL_PIPE_PUSH / _VERIFY / _STATS / _LEGACY_MEMOS / _TEXEL_RETAIN_MB /
  _INDEX_MIRROR_MB, plus MOBILEGL_IPC_* later), and §11 P0 wants them parsed beside
  the existing ones. Today every knob that has to reach an Android replay costs an
  edit in five files - run_android_retrace_local.py, trace-replay-ci.sh,
  TraceReplayActivity's request record, the JNI marshalling, and the setenv block in
  trace_replay_core.cpp. That per-knob tax is what this replaces: one extra,
  `--es mobilegl_env "K=V;K=V"`, carries all of them.
- Applied last, immediately before dlopen(libMobileGL.so), so it can also override
  the dedicated fields above it - MobileGL's config is read during the load, and an
  escape hatch that cannot beat the defaults is not one. An entry with no '=' unsets
  the variable, which is the only way to clear a default the marshalling sets.
- The existing per-knob flags stay: they carry semantics beyond a setenv (use_angle
  also selects a variant, the dump lists are joined, DirectVulkan forces the
  R11G11B10F fallback), and rewriting them as env strings would move that logic into
  the callers.
- Surface: --env / MOBILEGL_TRACE_ENV in trace-replay-ci.sh, repeatable --env
  KEY=VALUE in run_android_retrace_local.py, `mobilegl_env` intent extra,
  Request::envOverrides.
- The two-level parse now lives in trace_env_overrides.hpp, beside the semicolon
  splitter it shares with the texture and FBO dump lists, and
  tools/trace_replay/trace_env_overrides_test.cpp pins it: the empty entries a
  trailing ';' leaves behind must not become unsetenv(""), `K=` must stay a Set of
  the empty string rather than an Unset (a knob read with getenv() != nullptr sees
  those as opposite answers), and only the FIRST '=' may separate, or a value
  carrying '=' is truncated without a word of warning. The whole MOBILEGL_PIPE_*
  batch rides on this parse, and the only lane that exercised it end to end was an
  on-device retrace, which would have reported a splitting bug as "the knob had no
  effect".
- The check is built and RUN at build time and mobilegl_trace_replay depends on it,
  so `cmake --build ... --target mobilegl_trace_replay` - the exact command of
  test.yml's "Build trace replay" job, which never invokes ctest - runs it. It is
  assert-free on purpose: that lane configures Release, and <cassert> under NDEBUG
  would compile every check into a green run that checked nothing. Negative control:
  swapping find('=') for rfind('=') fails 2 checks, and keeping the splitter's empty
  entries fails 2 more.
This commit is contained in:
2026-09-05 20:49:57 -04:00
parent 8a239177ac
commit 38d4c2372c
9 changed files with 289 additions and 24 deletions
@@ -119,7 +119,7 @@ def render_summary():
shutil.copyfile(SUMMARY_DIR / SUMMARY_HTML, SUMMARY_DIR / "index.html")
def run_case(case, backend, extra_args=None, timeout_seconds=None):
def run_case(case, backend, extra_args=None, timeout_seconds=None, env_overrides=None):
backend_info = BACKENDS[backend]
apk = find_trace_apk()
trace_archive = FIXTURES / case["trace_archive"]
@@ -189,6 +189,11 @@ def run_case(case, backend, extra_args=None, timeout_seconds=None):
command.append("--avoid-angle-llvmpipe-explicit-lod-bias")
if case.get("coherent_as_flush"):
command.append("--coherent-as-flush")
# Generic environment passthrough: --env MOBILEGL_FOO=1 needs no per-knob plumbing in
# this script, in trace-replay-ci.sh, in the Activity, in the JNI marshalling or in the
# runner - one extra carries them all.
if env_overrides:
command.extend(["--env", ";".join(env_overrides)])
env = dict(**__import__("os").environ)
env["PYTHON"] = "python"
env["MSYS2_ARG_CONV_EXCL"] = "/data/*"
@@ -255,7 +260,13 @@ def run_benchmark_case(case, backend, args):
stale = RESULT_ROOT / f"{safe_case(case['name'])}-{backend}" / "benchmark.json"
if stale.exists():
stale.unlink()
rc = run_case(case, backend, extra_args=extra_args, timeout_seconds=args.benchmark_timeout_seconds)
rc = run_case(
case,
backend,
extra_args=extra_args,
timeout_seconds=args.benchmark_timeout_seconds,
env_overrides=args.env,
)
report = read_benchmark(case, backend, run_index)
if rc != 0 or report is None:
print(f"{label} run {run_index}/{args.benchmark_repeats}: FAILED (exit {rc})", flush=True)
@@ -287,6 +298,15 @@ def parse_args():
parser.add_argument("--backend", action="append", choices=sorted(BACKENDS), help="Backend to run; may be repeated.")
parser.add_argument("--all", action="store_true", help="Run every case in the APK workflow matrix.")
parser.add_argument("--keep-results", action="store_true", help="Do not clear the previous result root.")
parser.add_argument(
"--env",
action="append",
default=[],
metavar="KEY=VALUE",
help="Environment variable to set in the replay process, applied just before "
"libMobileGL.so is loaded; may be repeated. A KEY with no '=' unsets it. This "
"is the generic passthrough for MOBILEGL_* knobs that have no flag of their own.",
)
parser.add_argument(
"--benchmark",
action="store_true",
@@ -343,7 +363,7 @@ def main():
failures += run_benchmark_case(case, backend, args)
continue
print(f"=== Android retrace: {case['name']} / {backend} ===", flush=True)
rc = run_case(case, backend)
rc = run_case(case, backend, env_overrides=args.env)
try:
render_summary()
except Exception as error: