apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-03 15:56:49 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Adapt for iluvatar gpu (#2684)

Browse Source
This commit is contained in:
liddk1121
2025-07-07 16:53:14 +08:00
committed by GitHub
parent 2579e8fea8
commit 1b54a2831e
50 changed files with 4485 additions and 80 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
fastdeploy/output/token_processor.py
View File

@@ -122,9 +122,13 @@ class TokenProcessor(object):
if current_platform.is_xpu():
from fastdeploy.model_executor.ops.xpu import get_output
elif current_platform.is_iluvatar():
from fastdeploy.model_executor.ops.iluvatar import get_output
else:
from fastdeploy.model_executor.ops.gpu import (
get_output, get_output_ep, speculate_get_output)
from fastdeploy.model_executor.ops.gpu import (get_output,
get_output_ep,
speculate_get_output
)
rank_id = self.cfg.parallel_config.local_data_parallel_id
while True:
@@ -413,9 +417,12 @@ class WarmUpTokenProcessor(TokenProcessor):
if current_platform.is_xpu():
from fastdeploy.model_executor.ops.xpu import get_output
elif current_platform.is_iluvatar():
from fastdeploy.model_executor.ops.iluvatar import get_output
else:
from fastdeploy.model_executor.ops.gpu import (
get_output, speculate_get_output)
from fastdeploy.model_executor.ops.gpu import (get_output,
speculate_get_output
)
while self._is_running:
try: