CUDA驱动程序版本对于CUDA运行时版本是不够的

I got the message:

我得到消息:

"cutilCheckMsg() CUTIL CUDA error : kernel launch failure : CUDA driver version is insufficient for CUDA runtime version."

“cutilCheckMsg() CUTIL CUDA错误:内核启动失败:CUDA驱动程序版本对于CUDA运行时版本是不够的。”

while trying to run an example source code. Also happens for the function cutilSafeCall.

在尝试运行示例源代码时。也发生在函数cutilSafeCall。

Doing on the following environment.

在以下环境中执行。

windows 7 64bits
visual studio 2008
CUDA developer driver, toolkit and sdk 3.1
Emulation mode

windows 7 64bits visual studio 2008 CUDA developer driver, toolkit和sdk 3.1仿真模式

Can anybody help please? :)

有人能帮助好吗?:)

6 个解决方案

#1

I saw the same at runtime with the latest driver on Mac OS 10.6.

我在运行时看到了Mac OS 10.6上最新的驱动程序。

cudaError_t error = cudaGetDevice(&device); printf("%s\n", cudaGetErrorString(error));

cudaError_t错误= cudaGetDevice(及器具);printf(" % s \ n " cudaGetErrorString(错误));

I went back to the developer site, downloaded the driver again and now it runs. http://developer.nvidia.com/object/cuda_3_1_downloads.html#MacOS

我回到开发站点，再次下载驱动程序，现在它运行了。http://developer.nvidia.com/object/cuda_3_1_downloads.html MacOS

#2

You need to ensure that your driver version matches or exceeds your CUDA Toolkit version.

您需要确保您的驱动程序版本匹配或超过CUDA工具包版本。

For 2.3 you need a 190.x driver, for 3.0 you need 195.x and for 3.1 you need 256.x (actually anything up to the next multiple of five is ok, e.g. 258.x for 3.1).

需要190。x驱动，3.0需要195。x和3.1需要256。x(实际上是5的下一个倍数是可以的，例如258。x 3.1)。

You can check your driver version by either running the deviceQueryDrv SDK sample or go into the NVIDIA Control Panel and choose System Information.

您可以通过运行deviceQueryDrv SDK示例或进入NVIDIA控制面板并选择系统信息来检查驱动程序版本。

Download an updated driver from www.nvidia.com/drivers.

从www.nvidia.com/drivers中下载更新的驱动。

#3

You can either download the latest driver OR use an older toolkit version to compile your code.

您可以下载最新的驱动程序，或者使用旧的工具包版本来编译代码。

#4

My cent,

我的分,

with Linux/Unix this error may be related to the selected GPU mode (Performance/Power Saving Mode), when you select (with nvidia-settings utiliy) the integrated Intel GPU and you execute the deviceQuery script... you get this error:

对于Linux/Unix，这个错误可能与所选的GPU模式(性能/省电模式)有关，当您选择集成的Intel GPU并执行deviceQuery脚本时(使用nvida -settings)…你得到这个错误:

-> CUDA driver version is insufficient for CUDA runtime version

-> CUDA驱动版本对于CUDA运行时版本是不够的

But this error is misleading, by selecting back the NVIDIA(Performance mode) with nvidia-settings utility the problem disappears.

但是这个错误是有误导性的，通过选择NVIDIA -settings工具的NVIDIA(性能模式)，问题就消失了。

It is not a version problem.

这不是版本问题。

Regards

问候

P.s: "Power Saving Mode" tells Optimus to activate the CPU integrated Intel GPU

P。s:“节能模式”告诉Optimus激活CPU集成Intel GPU。

#5

CUDA driver version is insufficient for CUDA runtime version: means your GPU can`t been manipulated by the CUDA runtime API, so you need to update your driver.

CUDA驱动程序版本对于CUDA运行时版本是不够的:意味着您的GPU不能被CUDA运行时API操作，所以您需要更新您的驱动程序。

#6

Maybe it is related to the TBB lib: Error OpenCV with CUDA using TBB for multiple GPUs

也许它与TBB lib: Error OpenCV和CUDA有关，使用TBB处理多个gpu

Try rebuilding it making sure you passed the following parameters to CMake (assuming you already installed "tbb" and "tbb-devel" packages:

尝试重新构建它，确保您向CMake传递了以下参数(假设您已经安装了“tbb”和“tbb-devel”包):

-D WITH_TBB=YES -D TBB_INCLUDE_DIRS=/usr/include/tbb

#1