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 17, 08:00 UTC
Update setup.py (#282)
Co-authored-by: neubig <neubig@gmail.com>
Cody Yu committed
Jun 27, 2023 at 14:34 UTC
2b7d3aca2e1dd25fe26424f57c051af3b823cd71
1 file changed
+2
-3
setup.py
+2
-3
@@ -20,10 +20,9 @@ ABI = 1 if torch._C._GLIBCXX_USE_CXX11_ABI else 0
20
CXX_FLAGS += [f"-D_GLIBCXX_USE_CXX11_ABI={ABI}"]
21
NVCC_FLAGS += [f"-D_GLIBCXX_USE_CXX11_ABI={ABI}"]
22
23
-if not torch.cuda.is_available():
23
+if CUDA_HOME is None:
24
raise RuntimeError(
25
- f"Cannot find CUDA at CUDA_HOME: {CUDA_HOME}. "
26
- "CUDA must be available in order to build the package.")
25
+ f"Cannot find CUDA_HOME. CUDA must be available in order to build the package.")
26
27
28
def get_nvcc_cuda_version(cuda_dir: str) -> Version: