cudaErrorNoDevice GTS250 gentoo
第一个问题是
cudaGetDeviceCount 返回cudaErrorNoDevice：
这表明安装的 CUDA 驱动程序未检测到支持 CUDA 的设备。
cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX x86 Kernel Module 260.19.26 Sun Nov 28 22:38:24 PST 2010
GCC version: gcc version 4.4.5 (Gentoo 4.4.5 p1.2, pie-0.4.5)
lspci -v
...
02:00.0 VGA compatible controller: nVidia Corporation G92 [GeForce GTS 250] (rev a2) (prog-if 00 [VGA controller])
Flags: bus master, fast devsel, latency 0, IRQ 19
Memory at fb000000 (32-bit, non-prefetchable) [size=16M]
Memory at e0000000 (64-bit, prefetchable) [size=256M]
Memory at f8000000 (64-bit, non-prefetchable) [size=32M]
I/O ports at ec00 [size=128]
[virtual] Expansion ROM at fafe0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 3
Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [78] Express Endpoint, MSI 00
Capabilities: [100] Virtual Channel <?>
Capabilities: [128] Power Budgeting <?>
Capabilities: [600] Vendor Specific Information <?>
Kernel driver in use: nvidia
Kernel modules: nvidia
...
< code>nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Wed_Nov__3_16:14:08_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
CUDA 计算 SDK 3.2.16
，当我尝试制作 sdk 示例时，我遇到第二个问题：
make[1]: Entering directory `/home/style/NVIDIA_GPU_Computing_SDK/C/src/MersenneTwister'
nvcc fatal : Unsupported gpu architecture 'compute_20'
我尝试编辑 common.mk，但我不知道我到底应该编辑什么
感谢您的帮助
first question is
cudaGetDeviceCount return cudaErrorNoDevice:
This indicates that no CUDA-capable devices were detected by the installed CUDA driver.
cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX x86 Kernel Module 260.19.26 Sun Nov 28 22:38:24 PST 2010
GCC version: gcc version 4.4.5 (Gentoo 4.4.5 p1.2, pie-0.4.5)
lspci -v
...
02:00.0 VGA compatible controller: nVidia Corporation G92 [GeForce GTS 250] (rev a2) (prog-if 00 [VGA controller])
Flags: bus master, fast devsel, latency 0, IRQ 19
Memory at fb000000 (32-bit, non-prefetchable) [size=16M]
Memory at e0000000 (64-bit, prefetchable) [size=256M]
Memory at f8000000 (64-bit, non-prefetchable) [size=32M]
I/O ports at ec00 [size=128]
[virtual] Expansion ROM at fafe0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 3
Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [78] Express Endpoint, MSI 00
Capabilities: [100] Virtual Channel <?>
Capabilities: [128] Power Budgeting <?>
Capabilities: [600] Vendor Specific Information <?>
Kernel driver in use: nvidia
Kernel modules: nvidia
...
nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2010 NVIDIA Corporation
Built on Wed_Nov__3_16:14:08_PDT_2010
Cuda compilation tools, release 3.2, V0.2.1221
CUDA computing SDK 3.2.16
and when i try to make sdk samples i get second problem:
make[1]: Entering directory `/home/style/NVIDIA_GPU_Computing_SDK/C/src/MersenneTwister'
nvcc fatal : Unsupported gpu architecture 'compute_20'
i've tried to edit common.mk, but i don't know what exactly i should edit
Thanks for help
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（3）
验证您是否有A /DEV /NVIDIA0。如果不这样做，则可能需要重新启动X。
Verify if you have a /dev/nvidia0. If you do not, you may need to restart your X.
我没有看到显示您没有 CUDA 设备的错误。
回答你的第二个问题：
GeForce GTS 250 支持 CUDA 1.1 而不是 2.0。您需要编辑编译标志以将compute_20替换为compute_11。
I don't see the error where it says you have no CUDA devices.
In answer to your second question:
GeForce GTS 250 supports CUDA 1.1 not 2.0. You need to edit the compilation flags to replace compute_20 with compute_11.
由于我遇到了同样的问题，唯一对我有用的解决方案是重新安装驱动程序。请注意，该驱动程序未包含在 SDK 中。
As I had the same problem, the only solution that worked for me was to reinstall the driver. Notice that the driver is not included in the SDK.