EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[Bugfix][SHM] Use writer lock by default and remove redundant env (#4117)

Browse Source 
### What this PR does / why we need it?
This PR aims to remove env introduced by #3988 and use lock by default.
As described in https://github.com/vllm-project/vllm/issues/27858, we
have tested the writer lock method in various scenarios and the
performance is almost unaffected. Therefore, we believe that it would be
safe to enable the lock by default and remove the redundant env
`SHM_BARRIER` now.

After discussion, we decide to preserve env and set it as true by
default.

### Does this PR introduce _any_ user-facing change?
`SHM_BARRIER` is set as true by default.

### How was this patch tested?
by ci

---------

Signed-off-by: Zetong Li <slippersss@126.com>
This commit is contained in:
Zetong Li
2025-12-01 22:27:01 +08:00
committed by GitHub
parent 76d0ba4342
commit 52abd47f8c
2 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/patch/platform/__init__.py
View File

@@ -25,6 +25,6 @@ if os.getenv("DYNAMIC_EPLB", "false") == "true" or os.getenv(
"EXPERT_MAP_RECORD", "false") == "true":
import vllm_ascend.patch.platform.patch_multiproc_executor # noqa
if os.getenv("SHM_BARRIER", "false") == "true":
if os.getenv("SHM_BARRIER", "true") == "true":
import vllm_ascend.patch.platform.patch_core # noqa
import vllm_ascend.patch.platform.patch_message_queue # noqa

7
vllm_ascend/patch/worker/__init__.py
View File

@@ -15,6 +15,8 @@
# limitations under the License.
#
import os
from vllm.triton_utils import HAS_TRITON
if HAS_TRITON:
@@ -29,4 +31,7 @@ import vllm_ascend.patch.worker.patch_weight_loader # noqa
import vllm_ascend.patch.worker.patch_multimodal_merge # noqa
import vllm_ascend.patch.worker.patch_minicpm # noqa
import vllm_ascend.patch.worker.patch_deepseek_mtp # noqa
import vllm_ascend.patch.worker.patch_attention_layer # noqa
import vllm_ascend.patch.worker.patch_attention_layer # noqa
if os.getenv("SHM_BARRIER", "true") == "true":
import vllm_ascend.patch.platform.patch_message_queue # noqa