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

[CI] Reorganize stage-b 1-GPU tests for 5090 compatibility (#16826)

Browse Source
This commit is contained in:
Alison Shao
2026-01-15 15:23:35 -08:00
committed by GitHub
parent 8b22deef5b
commit 146b5fcc84
136 changed files with 277 additions and 404 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/registered/attention/test_create_kvindices.py
-1
View File
@@ -9,7 +9,6 @@ from sglang.test.test_utils import CustomTestCase
# Triton kernel unit test for KV indices creation
register_cuda_ci(est_time=10, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=7, suite="stage-b-test-small-1-gpu-5090")
register_amd_ci(est_time=10, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_mamba_unittest.py
-1
View File
@@ -16,7 +16,6 @@ from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
register_cuda_ci(est_time=9, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=9, suite="stage-b-test-small-1-gpu-amd")
register_cuda_ci(est_time=7, suite="stage-b-test-small-1-gpu-5090")
class TestMamba(unittest.TestCase):
test/registered/attention/test_radix_attention.py
-1
View File
@@ -15,7 +15,6 @@ from sglang.test.test_utils import (
# RadixAttention server integration tests
register_cuda_ci(est_time=100, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=79, suite="stage-b-test-small-1-gpu-5090")
register_amd_ci(est_time=100, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_radix_cache_unit.py
-1
View File
@@ -22,7 +22,6 @@ from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
# CPU-based unit test, runs quickly on any GPU runner
register_cuda_ci(est_time=5, suite="stage-b-test-small-1-gpu")
register_amd_ci(est_time=5, suite="stage-b-test-small-1-gpu-amd")
register_cuda_ci(est_time=7, suite="stage-b-test-small-1-gpu-5090")
import time
import unittest
test/registered/attention/test_swa_unittest.py
+1 -1
View File
@@ -9,7 +9,7 @@ from sglang.srt.mem_cache.swa_memory_pool import SWAKVPool, SWATokenToKVPoolAllo
from sglang.srt.mem_cache.swa_radix_cache import SWARadixCache
from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
register_cuda_ci(est_time=8, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=8, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=10, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_torch_native_attention_backend.py
-1
View File
@@ -19,7 +19,6 @@ from sglang.test.test_utils import (
# Torch native attention backend integration test with MMLU eval
register_cuda_ci(est_time=169, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=120, suite="stage-b-test-small-1-gpu-5090")
register_amd_ci(est_time=150, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_triton_attention_backend.py
+1 -1
View File
@@ -20,7 +20,7 @@ from sglang.test.test_utils import (
)
# Triton attention backend integration test with latency benchmark and MMLU eval
register_cuda_ci(est_time=200, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=200, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=1110, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_triton_attention_kernels.py
+1 -1
View File
@@ -23,7 +23,7 @@ from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
from sglang.test.test_utils import CustomTestCase, is_in_amd_ci
# Triton attention kernel unit tests (decode, extend, prefill)
register_cuda_ci(est_time=30, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=30, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=30, suite="stage-b-test-small-1-gpu-amd")
test/registered/attention/test_triton_sliding_window.py
+1 -1
View File
@@ -15,7 +15,7 @@ from sglang.test.test_utils import (
)
# Sliding window attention with Triton backend (Gemma-3 model)
register_cuda_ci(est_time=100, suite="stage-b-test-small-1-gpu")
register_cuda_ci(est_time=100, suite="stage-b-test-large-1-gpu")
register_amd_ci(est_time=100, suite="stage-b-test-small-1-gpu-amd")