test(moe): add GLM MegaMoE validation coverage

Add parser coverage for --moe-a2a-backend=megamoe and local unit coverage for MegaMoE gating, GLM fast-path routing, fallback routing, and sidecar weight preservation.

Use monkeypatch-style tests so the forward and weight-layout checks do not require loading GLM 5.2 or importing a real deep_gemm runtime.

Constraint: tests must not stage or depend on local skill/doc artifacts.

Feature-flag: --moe-a2a-backend=megamoe.

Conflict-hotspots: test/registered/unit/server_args/test_server_args.py, test/registered/unit/moe/test_glm_megamoe.py.

Scope-risk: runtime GPU e2e still needs target Blackwell/DeepGEMM environment.

Tested: PYTHONPYCACHEPREFIX=/private/tmp/sglang_pycache python3 -m py_compile test/registered/unit/server_args/test_server_args.py test/registered/unit/moe/test_glm_megamoe.py.

Tested: git diff --check.

Not-tested: PYTHONPATH=python python3 -m pytest test/registered/unit/moe/test_glm_megamoe.py -q; local Python has no pytest.

Not-tested: PYTHONPATH=python python3 -m unittest test.registered.unit.moe.test_glm_megamoe -v; local Python has no torch.

Not-tested: GLM 5.2 MegaMoE GPU e2e; local environment lacks target runtime and hardware.
This commit is contained in:
LuminolT
2026-07-06 10:44:05 +08:00
parent 57bf729af9
commit 93e3840578
2 changed files with 181 additions and 0 deletions
@@ -1,4 +1,5 @@
import json
import os
import tempfile
import unittest
from unittest.mock import MagicMock, patch
@@ -195,6 +196,46 @@ def test_nsa_index_skip_topk_offset_zero_means_unset():
assert override_args == {"index_topk_freq": 4}
def test_megamoe_a2a_backend_parser_sets_ep_to_tp():
import argparse
from sglang.srt.environ import envs
field = envs.SGLANG_OPT_FIX_MEGA_MOE_MEMORY
backup_present = field.name in os.environ
backup_value = os.environ.get(field.name)
backup_set_to_none = field._set_to_none
os.environ.pop(field.name, None)
field._set_to_none = False
field._cache_valid = False
try:
parser = argparse.ArgumentParser()
ServerArgs.add_cli_args(parser)
raw_args = parser.parse_args(
[
"--model-path",
"dummy",
"--tp-size",
"4",
"--moe-a2a-backend",
"megamoe",
]
)
args = ServerArgs.from_cli_args(raw_args)
assert args.moe_a2a_backend == "megamoe"
assert args.tp_size == 4
assert args.ep_size == 4
assert field.get() is True
finally:
if backup_present:
os.environ[field.name] = backup_value
else:
os.environ.pop(field.name, None)
field._set_to_none = backup_set_to_none
field._cache_valid = False
class TestLoadBalanceMethod(unittest.TestCase):
def test_non_pd_defaults_to_round_robin(self):
server_args = ServerArgs(model_path="dummy", disaggregation_mode="null")