minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

Expand and update test coverage for AMD CI (#10044)

Browse Source
This commit is contained in:
Hubert Lu
2025-11-04 22:15:13 -08:00
committed by GitHub
parent 9f5e701879
commit 3694266051
4 changed files with 214 additions and 101 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/srt/hicache/test_hicache_storage.py
+2
View File
@@ -1,3 +1,4 @@
import time
import unittest
from types import SimpleNamespace
@@ -39,6 +40,7 @@ class TestHiCache(CustomTestCase):
@classmethod
def tearDownClass(cls):
kill_process_tree(cls.process.pid)
time.sleep(5)
def test_mmlu(self):
args = SimpleNamespace(
test/srt/models/test_vlm_models.py
+10 -9
View File
@@ -8,7 +8,7 @@ import sys
import unittest
from types import SimpleNamespace
from sglang.srt.utils import kill_process_tree
from sglang.srt.utils import is_hip, kill_process_tree
from sglang.test.test_utils import (
DEFAULT_TIMEOUT_FOR_SERVER_LAUNCH,
DEFAULT_URL_FOR_TEST,
@@ -17,15 +17,16 @@ from sglang.test.test_utils import (
popen_launch_server,
)
_is_hip = is_hip()
# VLM models for testing
MODELS = [
SimpleNamespace(model="google/gemma-3-27b-it", mmmu_accuracy=0.45),
SimpleNamespace(
model="Qwen/Qwen2.5-VL-3B-Instruct",
mmmu_accuracy=0.4,
),
SimpleNamespace(model="openbmb/MiniCPM-V-2_6", mmmu_accuracy=0.4),
]
if _is_hip:
MODELS = [SimpleNamespace(model="openbmb/MiniCPM-V-2_6", mmmu_accuracy=0.4)]
else:
MODELS = [
SimpleNamespace(model="google/gemma-3-27b-it", mmmu_accuracy=0.45),
SimpleNamespace(model="Qwen/Qwen2.5-VL-3B-Instruct", mmmu_accuracy=0.4),
SimpleNamespace(model="openbmb/MiniCPM-V-2_6", mmmu_accuracy=0.4),
]
# Set default mem_fraction_static to 0.8
DEFAULT_MEM_FRACTION_STATIC = 0.8
test/srt/run_suite.py
+14 -8
View File
@@ -125,6 +125,7 @@ suites = {
TestFile("test_torch_compile_moe.py", 210),
TestFile("test_torch_native_attention_backend.py", 123),
TestFile("test_torchao.py", 70),
TestFile("test_triton_attention_kernels.py", 4),
TestFile("test_triton_attention_backend.py", 150),
TestFile("test_triton_attention_kernels.py", 4),
TestFile("test_triton_moe_channel_fp8_kernel.py", 25),
@@ -372,9 +373,9 @@ suites = {
suite_amd = {
"per-commit-amd": [
TestFile("function_call/test_json_schema_constraint.py", 1),
TestFile("hicache/test_hicache.py", 116),
TestFile("hicache/test_hicache_mla.py", 127),
TestFile("hicache/test_hicache_storage.py", 127),
# TestFile("hicache/test_hicache.py", 116), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/12575
# TestFile("hicache/test_hicache_mla.py", 127), # Disabled temporarily, # Temporarily disabled, see https://github.com/sgl-project/sglang/issues/12574
# TestFile("hicache/test_hicache_storage.py", 127), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/12575
TestFile("lora/test_lora.py", 150),
TestFile("lora/test_lora_backend.py", 99),
TestFile("lora/test_lora_cuda_graph.py", 250),
@@ -385,8 +386,9 @@ suite_amd = {
TestFile("models/test_qwen_models.py", 82),
TestFile("models/test_reward_models.py", 132),
TestFile("models/test_transformers_models.py", 320),
TestFile("openai_server/basic/test_openai_embedding.py", 79),
TestFile("openai_server/basic/test_openai_server.py", 270),
TestFile("models/test_vlm_models.py", 437),
TestFile("openai_server/basic/test_openai_embedding.py", 141),
TestFile("openai_server/basic/test_openai_server.py", 149),
TestFile("openai_server/basic/test_protocol.py", 10),
TestFile("openai_server/basic/test_serving_chat.py", 10),
TestFile("openai_server/basic/test_serving_completions.py", 10),
@@ -409,14 +411,15 @@ suite_amd = {
TestFile("test_create_kvindices.py", 2),
TestFile("test_eval_fp8_accuracy.py", 303),
TestFile("test_function_call_parser.py", 10),
TestFile("test_fused_moe.py", 80),
TestFile("test_fused_moe.py", 30),
TestFile("test_harmony_parser.py", 20),
TestFile("test_input_embeddings.py", 38),
TestFile("test_io_struct.py", 8),
TestFile("test_jinja_template_utils.py", 1),
TestFile("test_metrics.py", 32),
TestFile("test_metrics_utils.py", 1),
TestFile("test_mla.py", 242),
TestFile("test_mla_deepseek_v3.py", 221),
# TestFile("test_mla_deepseek_v3.py", 221), # Temporarily disabled, see https://github.com/sgl-project/sglang/issues/12574
TestFile("test_no_chunked_prefill.py", 108),
TestFile("test_page_size.py", 60),
TestFile("test_penalty.py", 180),
@@ -433,9 +436,12 @@ suite_amd = {
TestFile("test_torch_compile.py", 169),
TestFile("test_torch_compile_moe.py", 210),
TestFile("test_torch_native_attention_backend.py", 123),
TestFile("test_triton_attention_backend.py", 250),
# TestFile("test_triton_attention_kernels.py", 4),
TestFile("test_triton_attention_backend.py", 150),
TestFile("test_triton_sliding_window.py", 250),
TestFile("test_wave_attention_kernels.py", 2),
# Disabled temporarily
# TestFile("test_vlm_input_format.py", 300),
# TestFile("models/test_embedding_models.py", 73), # Disabled temporarily, see https://github.com/sgl-project/sglang/issues/11127
# TestFile("openai_server/features/test_openai_server_hidden_states.py", 240),
# TestFile("rl/test_update_weights_from_tensor.py", 48),