ci: decouple stage and runner for cuda registry (#25197)

test/registered/unit/mem_cache/test_decode_radix_lock_ref.py

@@ -22,7 +22,7 @@ Usage:
 
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
-register_cuda_ci(est_time=10, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=10, stage="stage-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=10, suite="stage-b-test-1-gpu-small-amd")
 
 import unittest

test/registered/unit/mem_cache/test_mamba_unittest.py

@@ -23,7 +23,7 @@ from sglang.srt.server_args import ServerArgs, set_global_server_args_for_schedu
 from sglang.srt.utils import get_device
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
-register_cuda_ci(est_time=10, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=10, stage="stage-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=9, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/unit/mem_cache/test_nsa_pool_host_unit.py

@@ -12,7 +12,7 @@ from sglang.srt.mem_cache.memory_pool_host import (
 from sglang.srt.utils import is_cuda, is_hip, is_npu, is_xpu
 from sglang.test.ci.ci_register import register_cuda_ci
 
-register_cuda_ci(est_time=9, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=9, stage="stage-b", runner_config="1-gpu-small")
 
 
 class TestNSAHiCacheTransfer(unittest.TestCase):

test/registered/unit/mem_cache/test_radix_cache_slru_accuracy.py

@@ -13,7 +13,7 @@ from sglang.srt.mem_cache.memory_pool import MHATokenToKVPool, ReqToTokenPool
 from sglang.srt.mem_cache.radix_cache import RadixCache, RadixKey
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
-register_cuda_ci(est_time=8, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=8, stage="stage-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=8, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/unit/mem_cache/test_radix_cache_unit.py

@@ -21,7 +21,7 @@ from sglang.srt.mem_cache.common import available_and_evictable_str
 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=15, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=15, stage="stage-b", runner_config="1-gpu-small")
 register_amd_ci(est_time=5, suite="stage-b-test-1-gpu-small-amd")
 
 import random

test/registered/unit/mem_cache/test_swa_eviction_boundary.py

@@ -26,7 +26,7 @@ from sglang.srt.mem_cache.swa_radix_cache import SWARadixCache
 from sglang.srt.utils import get_device
 from sglang.test.ci.ci_register import register_amd_ci, register_cuda_ci
 
-register_cuda_ci(est_time=12, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=12, stage="stage-b", runner_config="1-gpu-large")
 register_amd_ci(est_time=10, suite="stage-b-test-1-gpu-small-amd")
 
 # ---------------------------------------------------------------------------

test/registered/unit/mem_cache/test_swa_lock_release_lifecycle.py

@@ -30,7 +30,7 @@ from sglang.srt.utils import get_device
 from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cuda_ci(est_time=12, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=12, stage="stage-b", runner_config="1-gpu-small")
 
 
 def _build_tree(

test/registered/unit/mem_cache/test_swa_unittest.py

@@ -21,7 +21,7 @@ from sglang.srt.utils import get_device
 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=9, suite="stage-b-test-1-gpu-large")
+register_cuda_ci(est_time=9, stage="stage-b", runner_config="1-gpu-large")
 register_amd_ci(est_time=10, suite="stage-b-test-1-gpu-small-amd")
 
 

test/registered/unit/mem_cache/test_unified_radix_cache_bench.py

@@ -38,7 +38,7 @@ from sglang.srt.server_args import ServerArgs, set_global_server_args_for_schedu
 from sglang.srt.utils import get_device
 from sglang.test.ci.ci_register import register_cuda_ci
 
-register_cuda_ci(est_time=25, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=25, stage="stage-b", runner_config="1-gpu-small")
 
 # ---------------------------------------------------------------------------
 # Constants

test/registered/unit/mem_cache/test_unified_radix_cache_unittest.py

@@ -48,7 +48,7 @@ from sglang.srt.utils import get_device
 from sglang.test.ci.ci_register import register_cuda_ci
 from sglang.test.test_utils import CustomTestCase
 
-register_cuda_ci(est_time=10, suite="stage-b-test-1-gpu-small")
+register_cuda_ci(est_time=10, stage="stage-b", runner_config="1-gpu-small")
 
 
 @dataclass(frozen=True)