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

[Patch] Remove the patch of ECExampleConnector (#5976)

Browse Source 
### What this PR does / why we need it?
Part of #5304.

https://github.com/vllm-project/vllm/pull/30225 has been merged now. We
don't need this patch anymore.

- vLLM version: v0.13.0
- vLLM main:
2c24bc6996

Signed-off-by: gcanlin <canlinguosdu@gmail.com>
This commit is contained in:
Canlin Guo
2026-01-26 17:10:03 +08:00
committed by GitHub
parent b390e0ef78
commit 2d3b8a51f9
3 changed files with 0 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
vllm_ascend/patch/__init__.py
View File

@@ -42,18 +42,6 @@
# Future Plan: # Future Plan:
# Find a better way to support tensor alignment for 310p without this patch. # Find a better way to support tensor alignment for 310p without this patch.
# #
# ** 2. File: platform/patch_ec_connector.py**
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# 1. `vllm.distributed.ec_transfer.ec_connector.shared_storage_connector.ECSharedStorageConnector.start_load_caches`
# Why:
# it's hard code to cuda
# How：
# change the cuda to npu
# Related PR (if no, explain why):
# https://github.com/vllm-project/vllm/pull/30225
# Future Plan:
# Remove this patch when vllm merges the PR.
#
# ** 3. File: platform/patch_mamba_config.py** # ** 3. File: platform/patch_mamba_config.py**
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# 1. `vllm.model_executor.models.config.HybridAttentionMambaModelConfig.verify_and_update_config` # 1. `vllm.model_executor.models.config.HybridAttentionMambaModelConfig.verify_and_update_config`

1
vllm_ascend/patch/platform/__init__.py
View File

@@ -17,7 +17,6 @@
import os import os
import vllm_ascend.patch.platform.patch_distributed # noqa import vllm_ascend.patch.platform.patch_distributed # noqa
import vllm_ascend.patch.platform.patch_ec_connector # noqa
import vllm_ascend.patch.platform.patch_mamba_config # noqa import vllm_ascend.patch.platform.patch_mamba_config # noqa
import vllm_ascend.patch.platform.patch_sched_yield # noqa import vllm_ascend.patch.platform.patch_sched_yield # noqa
from vllm_ascend import envs from vllm_ascend import envs

30
vllm_ascend/patch/platform/patch_ec_connector.py
View File

@@ -1,30 +0,0 @@
import vllm.distributed.ec_transfer.ec_connector.example_connector
from safetensors.torch import load_file
from vllm.distributed.ec_transfer.ec_connector.example_connector import ECConnectorMetadata, ECExampleConnector
from vllm.logger import logger
class AscendECExampleConnector(ECExampleConnector):
def start_load_caches(self, encoder_cache, **kwargs) -> None:
metadata: ECConnectorMetadata = self._get_connector_metadata()
assert isinstance(metadata, ECConnectorMetadata)
assert encoder_cache is not None
if metadata is None:
logger.warning(
(
"In connector.start_load_caches, ",
"but the connector metadata is None",
)
)
return
# Load the EC for each mm data
for mm_data in metadata.mm_datas:
if mm_data.mm_hash in encoder_cache:
continue
filename = self._generate_filename_debug(mm_data.mm_hash)
ec_cache = load_file(filename)["ec_cache"].npu()
encoder_cache[mm_data.mm_hash] = ec_cache
logger.debug("Success load encoder cache for hash %s", mm_data.mm_hash)
vllm.distributed.ec_transfer.ec_connector.example_connector.ECExampleConnector = AscendECExampleConnector