xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-23 14:17:59 +08:00
Code Issues Packages Projects Releases Wiki Activity

[CI/Build] Skip cpu offloading test on AMD (#27690)

Browse Source 
Signed-off-by: zhewenli <zhewenli@meta.com>
This commit is contained in:
Zhewen Li 2025-10-29 05:55:51 -07:00 committed by GitHub
parent ad3ec89532
commit 9a0d2f0d92
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tests/v1/kv_offload/test_cpu_offloading.py
View File

@ -12,6 +12,7 @@ from tqdm import tqdm
from vllm import LLM, SamplingParams, TokensPrompt from vllm import LLM, SamplingParams, TokensPrompt
from vllm.config import KVEventsConfig, KVTransferConfig from vllm.config import KVEventsConfig, KVTransferConfig
from vllm.distributed.kv_events import BlockStored, KVEventBatch from vllm.distributed.kv_events import BlockStored, KVEventBatch
from vllm.platforms import current_platform
CPU_BLOCK_SIZES = [16, 48] CPU_BLOCK_SIZES = [16, 48]
@ -63,6 +64,9 @@ class MockSubscriber:
self.sub.close() self.sub.close()
@pytest.mark.skipif(
not current_platform.is_cuda(), reason="CPU offloading only supported on CUDA"
)
@pytest.mark.parametrize("cpu_block_size", CPU_BLOCK_SIZES) @pytest.mark.parametrize("cpu_block_size", CPU_BLOCK_SIZES)
def test_cpu_offloading(cpu_block_size: int) -> None: def test_cpu_offloading(cpu_block_size: int) -> None:
""" """