chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

ci: migrate scheduler tests to test/registered/scheduler/ (#16442)

Browse Source
This commit is contained in:
Alison Shao
2026-01-06 14:55:52 -08:00
committed by GitHub
parent 2e0527dd74
commit 399d5283f8
8 changed files with 27 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
test/srt/test_abort.py → test/registered/scheduler/test_abort.py
View File

@@ -7,6 +7,7 @@ import requests
from sglang.srt.environ import envs
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
@@ -16,6 +17,9 @@ from sglang.test.test_utils import (
run_and_check_memory_leak,
)
register_cuda_ci(est_time=131, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=51, suite="stage-b-test-small-1-gpu-amd")
class TestAbort(CustomTestCase):
def workload_func(self, base_url, model):

4
test/srt/test_chunked_prefill.py → test/registered/scheduler/test_chunked_prefill.py
View File

@@ -4,8 +4,12 @@ python3 -m unittest test_chunked_prefill.TestChunkedPrefill.test_mixed_chunked_p
import unittest
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import CustomTestCase, run_mmlu_test, run_mulit_request_test
register_cuda_ci(est_time=312, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=312, suite="stage-b-test-small-1-gpu-amd")
class TestChunkedPrefill(CustomTestCase):
def test_chunked_prefill(self):

4
test/srt/test_no_chunked_prefill.py → test/registered/scheduler/test_no_chunked_prefill.py
View File

@@ -1,5 +1,6 @@
import unittest
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_TEST,
CustomTestCase,
@@ -7,6 +8,9 @@ from sglang.test.test_utils import (
run_mmlu_test,
)
register_cuda_ci(est_time=108, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=108, suite="stage-b-test-small-1-gpu-amd")
class TestNoChunkedPrefill(CustomTestCase):

3
test/srt/test_no_overlap_scheduler.py → test/registered/scheduler/test_no_overlap_scheduler.py
View File

@@ -6,8 +6,11 @@ python3 test_overlap_schedule.py
import unittest
from sglang.test.ci.ci_register import register_cuda_ci
from sglang.test.test_utils import CustomTestCase, run_mmlu_test
register_cuda_ci(est_time=217, suite="stage-b-test-small-1-gpu")
class TestOverlapSchedule(CustomTestCase):
def test_no_radix_attention_chunked_prefill(self):

4
test/srt/test_prefill_adder.py → test/registered/scheduler/test_prefill_adder.py
View File

@@ -4,8 +4,12 @@ from unittest.mock import MagicMock, patch
from sglang.srt.managers.schedule_batch import Req
from sglang.srt.managers.schedule_policy import PrefillAdder
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import CustomTestCase
register_cuda_ci(est_time=1, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=2, suite="stage-b-test-small-1-gpu-amd")
class TestPrefillAdder(CustomTestCase):
def setUp(self):

4
test/srt/test_priority_scheduling.py → test/registered/scheduler/test_priority_scheduling.py
View File

@@ -5,6 +5,7 @@ import unittest
from typing import Any, List, Optional, Tuple
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import (
DEFAULT_SMALL_MODEL_NAME_FOR_TEST,
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
@@ -16,6 +17,9 @@ from sglang.test.test_utils import (
send_concurrent_generate_requests_with_custom_params,
)
register_cuda_ci(est_time=130, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=195, suite="stage-b-test-small-1-gpu-amd")
class TestPriorityScheduling(CustomTestCase):
@classmethod

4
test/srt/test_retract_decode.py → test/registered/scheduler/test_retract_decode.py
View File

@@ -6,6 +6,7 @@ import requests
from sglang.srt.environ import envs
from sglang.srt.utils import kill_process_tree
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.run_eval import run_eval
from sglang.test.test_utils import (
DEFAULT_MODEL_NAME_FOR_TEST,
@@ -16,6 +17,9 @@ from sglang.test.test_utils import (
)
from sglang.utils import is_in_ci
register_cuda_ci(est_time=259, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=450, suite="stage-b-test-small-1-gpu-amd")
class TestRetractDecode(CustomTestCase):
"""python -m unittest test_retract_decode.TestRetractDecode"""

13
test/srt/run_suite.py
View File

@@ -9,8 +9,6 @@ from sglang.test.ci.ci_utils import TestFile, run_unittest_files
# NOTE: please sort the test cases alphabetically by the test file name
suites = {
"per-commit-1-gpu": [
TestFile("test_abort.py", 131),
TestFile("test_chunked_prefill.py", 312),
TestFile("test_deterministic.py", 228),
TestFile("test_eval_fp8_accuracy.py", 250),
TestFile("test_evs.py", 20),
@@ -25,13 +23,8 @@ suites = {
TestFile("test_model_hooks.py", 6),
TestFile("test_modelopt_loader.py", 11),
TestFile("test_multi_tokenizer.py", 230),
TestFile("test_no_chunked_prefill.py", 108),
TestFile("test_no_overlap_scheduler.py", 217),
TestFile("test_page_size.py", 60),
TestFile("test_prefill_adder.py", 1),
TestFile("test_priority_scheduling.py", 130),
TestFile("test_request_queue_validation.py", 47),
TestFile("test_retract_decode.py", 259),
TestFile("test_score_api.py", 260),
TestFile("test_server_args.py", 9),
TestFile("test_skip_tokenizer_init.py", 77),
@@ -134,9 +127,7 @@ suite_amd = {
# TestFile("lora/test_lora_backend.py", 99), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/13107
# TestFile("lora/test_lora_cuda_graph.py", 250), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/13107
# TestFile("lora/test_lora_qwen3.py", 97), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/13107
TestFile("test_abort.py", 51),
TestFile("test_bench_typebaseddispatcher.py", 10),
TestFile("test_chunked_prefill.py", 312),
TestFile("test_eval_fp8_accuracy.py", 303),
TestFile("test_external_models.py", 45),
TestFile("test_input_embeddings.py", 38),
@@ -144,14 +135,10 @@ suite_amd = {
TestFile("test_jinja_template_utils.py", 1),
TestFile("test_model_hooks.py", 10),
TestFile("test_multi_tokenizer.py", 345),
TestFile("test_no_chunked_prefill.py", 108),
TestFile("test_page_size.py", 60),
TestFile("test_prefill_adder.py", 2),
TestFile("test_priority_scheduling.py", 195),
TestFile("test_profile_merger.py", 12),
TestFile("test_profile_merger_http_api.py", 15),
TestFile("test_request_queue_validation.py", 70),
TestFile("test_retract_decode.py", 450),
TestFile("test_rope_rocm.py", 3),
TestFile("test_server_args.py", 1),
TestFile("test_skip_tokenizer_init.py", 117),