@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / da317197dd
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 15, 23:30 UTC

[Build] Fix cuda link target of cumem_allocator in CPU env (#12863)

Signed-off-by: YuhongGuo <yuhong.gyh@antgroup.com> Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>

Yuhong Guo committed Feb 11, 2025 at 21:55 UTC da317197dd9352a7718d9bf697c2a5aeb9d42b41
1 file changed +1 -1
CMakeLists.txt
+1 -1
@@ -192,7 +192,7 @@ set_gencode_flags_for_srcs(
192 if(VLLM_GPU_LANG STREQUAL "CUDA")
193 message(STATUS "Enabling cumem allocator extension.")
194 # link against cuda driver library
195 - list(APPEND CUMEM_LIBS cuda)
195 + list(APPEND CUMEM_LIBS CUDA::cuda_driver)
196 define_gpu_extension_target(
197 cumem_allocator
198 DESTINATION vllm