wassname/vllm
1
0
Fork 0
mirror of https://github.com/wassname/vllm.git synced 2026-09-09 11:40:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Kernel][Hardware][Amd]Custom paged attention kernel for rocm (#8310)

Browse Source
This commit is contained in:
Charlie Fu
2024-09-13 17:01:11 -07:00
committed by GitHub
parent 851725202a
commit 1ef0d2efd0
8 changed files with 1371 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
setup.py
+3
View File
@@ -462,6 +462,9 @@ if _build_core_ext():
if _is_cuda() or _is_hip():
ext_modules.append(CMakeExtension(name="vllm._moe_C"))
if _is_hip():
ext_modules.append(CMakeExtension(name="vllm._rocm_C"))
if _build_custom_ops():
ext_modules.append(CMakeExtension(name="vllm._C"))