sgl-project--sglang
94057c3d3e
PR Test (NPU) / check-changes (push) Has been cancelled
PR Test (NPU) / pr-gate (push) Has been cancelled
PR Test (NPU) / set-image-config (push) Has been cancelled
PR Test (NPU) / stage-b-test-1-npu-a2 (0) (push) Has been cancelled
PR Test (NPU) / stage-b-test-1-npu-a2 (1) (push) Has been cancelled
PR Test (NPU) / stage-b-test-2-npu-a2 (0) (push) Has been cancelled
PR Test (NPU) / stage-b-test-2-npu-a2 (1) (push) Has been cancelled
PR Test (NPU) / stage-b-test-4-npu-a3 (push) Has been cancelled
PR Test (NPU) / stage-b-test-16-npu-a3 (push) Has been cancelled
PR Test (NPU) / multimodal-gen-test-1-npu-a3 (push) Has been cancelled
PR Test (NPU) / multimodal-gen-test-2-npu-a3 (push) Has been cancelled
PR Test (Arm64) / pr-gate (push) Has been cancelled
PR Test (Arm64) / check-changes (push) Has been cancelled
PR Test (Arm64) / build-test (push) Has been cancelled
PR Test (sgl-router) / gate (push) Has been cancelled
PR Test (sgl-router) / tier-1 — lint (push) Has been cancelled
PR Test (sgl-router) / tier-2 — build + test (push) Has been cancelled
PR Test (sgl-router) / tier-3 — docker (placeholder) (push) Has been cancelled
PR Test (sgl-router) / tier-3 — k8s integration (push) Has been cancelled
PR Test (sgl-router) / tier-3 — e2e (push) Has been cancelled
PR Test (sgl-router) / finish (push) Has been cancelled
PR Test (NPU) / single-node-poc (map[name:qwen3_6_27b_w8a8_1p_in64k_out1k_50ms runner:linux-aarch64-a3-2 test_case:test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py test_type:perf]) (push) Has been cancelled
PR Test (NPU) / pr-test-npu-finish (push) Has been cancelled
PR Test (Xeon) / pr-gate (push) Has been cancelled
PR Test (Xeon) / check-changes (push) Has been cancelled
PR Test (Xeon) / build-test (, xeon-gnr, base-b-test-cpu) (push) Has been cancelled
PR Test (XPU) / check-changes (push) Has been cancelled
PR Test (XPU) / pr-gate (push) Has been cancelled
PR Test (XPU) / stage-a-test-1-gpu-xpu (push) Has been cancelled
PR Test (XPU) / wait-for-stage-a (push) Has been cancelled
PR Test (XPU) / stage-b-test-1-gpu-xpu (push) Has been cancelled
PR Test (XPU) / finish (push) Has been cancelled
CI Model Inventory / build-inventory (push) Has been cancelled
Lint / lint (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark Compilation Check (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark - Manual Policy (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark - Request Processing (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark Summary (push) Has been cancelled
PR Test (SMG) / build-wheel (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on windows (x86_64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on macos (x86_64 - auto) (push) Has been cancelled
PR Test (SMG) / python-unit-tests (push) Has been cancelled
PR Test (SMG) / unit-tests (push) Has been cancelled
PR Test (SMG) / benchmarks (push) Has been cancelled
PR Test (SMG) / chat-completions (push) Has been cancelled
PR Test (SMG) / chat-completions-4gpu (push) Has been cancelled
PR Test (SMG) / e2e (push) Has been cancelled
PR Test (SMG) / docker-build-test (push) Has been cancelled
PR Test (SMG) / k8s-integration (push) Has been cancelled
PR Test (SMG) / finish (push) Has been cancelled
PR Test (SMG) / summarize-benchmarks (push) Has been cancelled
Release SGLang Model Gateway Docker Image / publish (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on macos (aarch64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (aarch64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (x86_64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (aarch64 - musllinux_1_1) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (x86_64 - musllinux_1_1) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / Build SDist (push) Has been cancelled
Release SGLang Model Gateway to PyPI / Upload to PyPI (push) Has been cancelled
Release SGLang Kernels / build-cu129-matrix (aarch64, 12.9, 3.10, arm-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / build-cu129-matrix (x86_64, 12.9, 3.10, x64-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / release-cu129 (push) Has been cancelled
Release SGLang Kernels / build-cu130-matrix (aarch64, 13.0, 3.10, arm-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / build-cu130-matrix (x86_64, 13.0, 3.10, x64-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / release-cu130 (push) Has been cancelled
Release SGLang Kernels / build-rocm-matrix (3.10, 700) (push) Has been cancelled
Release SGLang Kernels / build-rocm-matrix (3.10, 720) (push) Has been cancelled
Release SGLang Kernels / release-rocm700 (push) Has been cancelled
Release SGLang Kernels / release-rocm720 (push) Has been cancelled
Release SGLang Kernels / build-musa43 (43, 3.10) (push) Has been cancelled
Release SGLang Kernels / release-musa43 (push) Has been cancelled
220 行
7.6 KiB
Python
220 行
7.6 KiB
Python
"""Unit tests for evict_policy.py"""
|
|
|
|
from sglang.test.ci.ci_register import register_cpu_ci
|
|
|
|
register_cpu_ci(est_time=6, suite="base-a-test-cpu")
|
|
register_cpu_ci(est_time=7, suite="base-c-test-cpu")
|
|
|
|
import unittest
|
|
from unittest.mock import MagicMock
|
|
|
|
from sglang.srt.mem_cache.evict_policy import (
|
|
FIFOStrategy,
|
|
FILOStrategy,
|
|
LFUStrategy,
|
|
LRUStrategy,
|
|
MRUStrategy,
|
|
PriorityStrategy,
|
|
SLRUStrategy,
|
|
)
|
|
|
|
|
|
def _make_node(**kwargs):
|
|
node = MagicMock()
|
|
node.last_access_time = kwargs.get("last_access_time", 0.0)
|
|
node.hit_count = kwargs.get("hit_count", 0)
|
|
node.creation_time = kwargs.get("creation_time", 0.0)
|
|
node.priority = kwargs.get("priority", 0)
|
|
return node
|
|
|
|
|
|
class TestLRUStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = LRUStrategy()
|
|
|
|
def test_priority_is_last_access_time(self):
|
|
node = _make_node(last_access_time=42.0)
|
|
self.assertEqual(self.strategy.get_priority(node), 42.0)
|
|
|
|
def test_older_access_evicted_first(self):
|
|
old = _make_node(last_access_time=1.0)
|
|
new = _make_node(last_access_time=10.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(old), self.strategy.get_priority(new)
|
|
)
|
|
|
|
|
|
class TestLFUStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = LFUStrategy()
|
|
|
|
def test_priority_is_hit_count_and_time(self):
|
|
node = _make_node(hit_count=5, last_access_time=3.0)
|
|
self.assertEqual(self.strategy.get_priority(node), (5, 3.0))
|
|
|
|
def test_lower_hit_count_evicted_first(self):
|
|
cold = _make_node(hit_count=1, last_access_time=10.0)
|
|
hot = _make_node(hit_count=100, last_access_time=1.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(cold), self.strategy.get_priority(hot)
|
|
)
|
|
|
|
def test_same_hit_count_older_access_evicted_first(self):
|
|
old = _make_node(hit_count=3, last_access_time=1.0)
|
|
new = _make_node(hit_count=3, last_access_time=10.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(old), self.strategy.get_priority(new)
|
|
)
|
|
|
|
|
|
class TestFIFOStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = FIFOStrategy()
|
|
|
|
def test_priority_is_creation_time(self):
|
|
node = _make_node(creation_time=7.0)
|
|
self.assertEqual(self.strategy.get_priority(node), 7.0)
|
|
|
|
def test_earlier_created_evicted_first(self):
|
|
first = _make_node(creation_time=1.0)
|
|
second = _make_node(creation_time=5.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(first), self.strategy.get_priority(second)
|
|
)
|
|
|
|
|
|
class TestMRUStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = MRUStrategy()
|
|
|
|
def test_priority_is_negated_access_time(self):
|
|
node = _make_node(last_access_time=5.0)
|
|
self.assertEqual(self.strategy.get_priority(node), -5.0)
|
|
|
|
def test_most_recently_used_evicted_first(self):
|
|
"""MRU evicts the most recently accessed node first (lowest priority value)."""
|
|
old = _make_node(last_access_time=1.0)
|
|
new = _make_node(last_access_time=10.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(new), self.strategy.get_priority(old)
|
|
)
|
|
|
|
|
|
class TestFILOStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = FILOStrategy()
|
|
|
|
def test_priority_is_negated_creation_time(self):
|
|
node = _make_node(creation_time=3.0)
|
|
self.assertEqual(self.strategy.get_priority(node), -3.0)
|
|
|
|
def test_last_created_evicted_first(self):
|
|
"""FILO evicts the most recently created node first."""
|
|
first = _make_node(creation_time=1.0)
|
|
second = _make_node(creation_time=5.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(second), self.strategy.get_priority(first)
|
|
)
|
|
|
|
|
|
class TestPriorityStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = PriorityStrategy()
|
|
|
|
def test_priority_is_tuple(self):
|
|
node = _make_node(priority=2, last_access_time=4.0)
|
|
self.assertEqual(self.strategy.get_priority(node), (2, 4.0))
|
|
|
|
def test_lower_priority_evicted_first(self):
|
|
low = _make_node(priority=1, last_access_time=10.0)
|
|
high = _make_node(priority=5, last_access_time=1.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(low), self.strategy.get_priority(high)
|
|
)
|
|
|
|
def test_same_priority_older_access_evicted_first(self):
|
|
old = _make_node(priority=3, last_access_time=1.0)
|
|
new = _make_node(priority=3, last_access_time=10.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(old), self.strategy.get_priority(new)
|
|
)
|
|
|
|
|
|
class TestSLRUStrategy(unittest.TestCase):
|
|
def setUp(self):
|
|
self.strategy = SLRUStrategy(protected_threshold=2)
|
|
|
|
def test_probationary_segment(self):
|
|
node = _make_node(hit_count=1, last_access_time=5.0)
|
|
self.assertEqual(self.strategy.get_priority(node), (0, 5.0))
|
|
|
|
def test_protected_segment(self):
|
|
node = _make_node(hit_count=2, last_access_time=5.0)
|
|
self.assertEqual(self.strategy.get_priority(node), (1, 5.0))
|
|
|
|
def test_highly_accessed_is_protected(self):
|
|
node = _make_node(hit_count=100, last_access_time=5.0)
|
|
self.assertEqual(self.strategy.get_priority(node), (1, 5.0))
|
|
|
|
def test_probationary_evicted_before_protected(self):
|
|
prob = _make_node(hit_count=1, last_access_time=10.0)
|
|
prot = _make_node(hit_count=5, last_access_time=1.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(prob), self.strategy.get_priority(prot)
|
|
)
|
|
|
|
def test_same_segment_older_access_evicted_first(self):
|
|
old = _make_node(hit_count=0, last_access_time=1.0)
|
|
new = _make_node(hit_count=0, last_access_time=10.0)
|
|
self.assertLess(
|
|
self.strategy.get_priority(old), self.strategy.get_priority(new)
|
|
)
|
|
|
|
def test_custom_threshold(self):
|
|
strategy = SLRUStrategy(protected_threshold=5)
|
|
below = _make_node(hit_count=4, last_access_time=1.0)
|
|
at = _make_node(hit_count=5, last_access_time=1.0)
|
|
self.assertEqual(strategy.get_priority(below), (0, 1.0))
|
|
self.assertEqual(strategy.get_priority(at), (1, 1.0))
|
|
|
|
def test_default_threshold_is_2(self):
|
|
default = SLRUStrategy()
|
|
self.assertEqual(default.protected_threshold, 2)
|
|
|
|
|
|
class TestEvictionOrdering(unittest.TestCase):
|
|
"""Integration-style test: sort a list of nodes by eviction priority."""
|
|
|
|
def test_lru_ordering(self):
|
|
strategy = LRUStrategy()
|
|
nodes = [
|
|
_make_node(last_access_time=5.0),
|
|
_make_node(last_access_time=1.0),
|
|
_make_node(last_access_time=3.0),
|
|
]
|
|
eviction_order = sorted(nodes, key=strategy.get_priority)
|
|
times = [n.last_access_time for n in eviction_order]
|
|
self.assertEqual(times, [1.0, 3.0, 5.0])
|
|
|
|
def test_slru_ordering(self):
|
|
strategy = SLRUStrategy(protected_threshold=2)
|
|
nodes = [
|
|
_make_node(hit_count=5, last_access_time=1.0), # protected, old
|
|
_make_node(hit_count=0, last_access_time=10.0), # probationary, new
|
|
_make_node(hit_count=0, last_access_time=2.0), # probationary, old
|
|
_make_node(hit_count=3, last_access_time=8.0), # protected, new
|
|
]
|
|
eviction_order = sorted(nodes, key=strategy.get_priority)
|
|
expected = [
|
|
(0, 2.0), # probationary old
|
|
(0, 10.0), # probationary new
|
|
(1, 1.0), # protected old
|
|
(1, 8.0), # protected new
|
|
]
|
|
actual = [strategy.get_priority(n) for n in eviction_order]
|
|
self.assertEqual(actual, expected)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|