Move deep gemm related arguments to sglang.srt.environ (#11547)

test/srt/test_hybrid_attn_backend.py

@@ -4,6 +4,7 @@ from types import SimpleNamespace
 
 import requests
 
+from sglang.srt.environ import envs
 from sglang.srt.utils import get_device_sm, kill_process_tree
 from sglang.test.few_shot_gsm8k import run_eval as run_eval_few_shot_gsm8k
 from sglang.test.test_utils import (
@@ -49,18 +50,20 @@ class TestHybridAttnBackendBase(CustomTestCase):
     def setUpClass(cls):
         # disable deep gemm precompile to make launch server faster
         # please don't do this if you want to make your inference workload faster
-        os.environ["SGL_JIT_DEEPGEMM_PRECOMPILE"] = "false"
-        os.environ["SGL_ENABLE_JIT_DEEPGEMM"] = "false"
-        if cls.speculative_decode:
-            model = DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST
-        else:
-            model = cls.model
-        cls.process = popen_launch_server(
-            model,
-            cls.base_url,
-            timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
-            other_args=cls.get_server_args(),
-        )
+        with (
+            envs.SGLANG_JIT_DEEPGEMM_PRECOMPILE.override(False),
+            envs.SGLANG_ENABLE_JIT_DEEPGEMM.override(False),
+        ):
+            if cls.speculative_decode:
+                model = DEFAULT_EAGLE_TARGET_MODEL_FOR_TEST
+            else:
+                model = cls.model
+            cls.process = popen_launch_server(
+                model,
+                cls.base_url,
+                timeout=DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
+                other_args=cls.get_server_args(),
+            )
 
     @classmethod
     def tearDownClass(cls):