biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Core] Support fully transparent sleep mode (#11743)

Browse Source 
Signed-off-by: youkaichao <youkaichao@gmail.com>
This commit is contained in:
youkaichao
2025-01-22 14:39:32 +08:00
committed by GitHub
parent 4004f144f3
commit 68ad4e3a8d
14 changed files with 877 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
setup.py
View File

@@ -301,6 +301,7 @@ class repackage_wheel(build_ext):
"vllm/vllm_flash_attn/vllm_flash_attn_c.abi3.so",
"vllm/vllm_flash_attn/flash_attn_interface.py",
"vllm/vllm_flash_attn/__init__.py",
"vllm/cumem_allocator.abi3.so",
# "vllm/_version.py", # not available in nightly wheels yet
]
file_members = filter(lambda x: x.filename in files_to_copy,
@@ -594,6 +595,7 @@ if _is_hip():
if _is_cuda():
ext_modules.append(
CMakeExtension(name="vllm.vllm_flash_attn.vllm_flash_attn_c"))
ext_modules.append(CMakeExtension(name="vllm.cumem_allocator"))
if _build_custom_ops():
ext_modules.append(CMakeExtension(name="vllm._C"))