Error building extension 'wkv_cuda' 报错

[LiangWei88]

已经按照说明安装, 但是用 cuda 时报错

`
Using C:\Users\Admin\AppData\Local\torch_extensions\torch_extensions\Cache\py310_cu117 as PyTorch extensions root...
Detected CUDA files, patching ldflags
Emitting ninja build file C:\Users\Admin\AppData\Local\torch_extensions\torch_extensions\Cache\py310_cu117\wkv_cuda\build.ninja...
Building extension module wkv_cuda...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
[1/2] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\bin\nvcc --generate-dependencies-with-compile --dependency-output operators.cuda.o.d -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcompiler /EHsc -Xcompiler /wd4190 -Xcompiler /wd4018 -Xcompiler /wd4275 -Xcompiler /wd4267 -Xcompiler /wd4244 -Xcompiler /wd4251 -Xcompiler /wd4819 -Xcompiler /MD -DTORCH_EXTENSION_NAME=wkv_cuda -DTORCH_API_INCLUDE_EXTENSION_H -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\torch\csrc\api\include -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\TH -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\include" -IE:\Programs\Python\Python310\Include -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 "-t 4" -std=c++17 --use_fast_math -O3 --extra-device-vectorization -c E:\RWKV_Role_Playing\Env\lib\site-packages\rwkv\cuda\operators.cu -o operators.cuda.o
FAILED: operators.cuda.o
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\bin\nvcc --generate-dependencies-with-compile --dependency-output operators.cuda.o.d -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcompiler /EHsc -Xcompiler /wd4190 -Xcompiler /wd4018 -Xcompiler /wd4275 -Xcompiler /wd4267 -Xcompiler /wd4244 -Xcompiler /wd4251 -Xcompiler /wd4819 -Xcompiler /MD -DTORCH_EXTENSION_NAME=wkv_cuda -DTORCH_API_INCLUDE_EXTENSION_H -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\torch\csrc\api\include -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\TH -IE:\RWKV_Role_Playing\Env\lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\include" -IE:\Programs\Python\Python310\Include -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 "-t 4" -std=c++17 --use_fast_math -O3 --extra-device-vectorization -c E:\RWKV_Role_Playing\Env\lib\site-packages\rwkv\cuda\operators.cu -o operators.cuda.o
nvcc fatal : Unknown option '-t 4'
ninja: build stopped: subcommand failed.
Traceback (most recent call last):
File "E:\RWKV_Role_Playing\Env\lib\site-packages\torch\utils\cpp_extension.py", line 1900, in _run_ninja_build
subprocess.run(
File "E:\Programs\Python\Python310\lib\subprocess.py", line 526, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "E:\RWKV_Role_Playing\webui.py", line 20, in
from modules.model_utils import ModelUtils
File "E:\RWKV_Role_Playing\modules\model_utils.py", line 6, in
from rwkv.model import RWKV
File "E:\RWKV_Role_Playing\Env\lib\site-packages\rwkv\model.py", line 29, in
load(
File "E:\RWKV_Role_Playing\Env\lib\site-packages\torch\utils\cpp_extension.py", line 1284, in load
return _jit_compile(
File "E:\RWKV_Role_Playing\Env\lib\site-packages\torch\utils\cpp_extension.py", line 1508, in _jit_compile
_write_ninja_file_and_build_library(
File "E:\RWKV_Role_Playing\Env\lib\site-packages\torch\utils\cpp_extension.py", line 1623, in _write_ninja_file_and_build_library
_run_ninja_build(
File "E:\RWKV_Role_Playing\Env\lib\site-packages\torch\utils\cpp_extension.py", line 1916, in _run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'wkv_cuda'
(Env)
`

[shengxia]

老实讲我也不太能看出来，不过“subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.”这一句，我好像在哪儿见过，是版本的问题，你方便说一下pytorch的版本吗？

[ms-KuroNeko]

问题根本不在Pytorch，问题在ninja

subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

ninja -v 报错
检查 ninja 是否正确安装并可运行

我 Ubuntu 同样的错误是在安装好 PythonDev 之后解决的
建议把 PythonDev GCC G++ ninja 都检查一遍
Windows检查一下 MSVC 编译器，实在不行装个 Visual Studio 配个 C++ 开发环境暴力解决

[1014670860]

按照楼上的说的, 安装了python-devtools, 还是不行,点解

[shengxia]

这个编译好像需要装cudnn（反正我之前没装报错了），你这个装了吗？

[lty365]

环境变量 配置 ，C++ CUDA 什么的

[shengxia]

虽然有点晚，但是我最近才闹明白这个该怎么看，你这个报错主要的地方是：
nvcc fatal : Unknown option '-t 4'
这一句，就是不识别这个编译参数，不过鉴于这是去年的事儿了，rwkv库也升级了好几版了，我也不确定现在这个版本还有没有这个编译参数了，你要不更新到最新版再试试，看看行不行。