如何在Linux下禁用或更改GPU的超时限制? [英] How to disable or change the timeout limit for the GPU under linux?

问题描述

有人知道如何在Ubuntu 12.10下禁用或更改CUDA内核的超时限制吗? (在Windows的当前版本中，可以在注册表中设置超时限制.)

Does anybody know how to disable or change the timeout limit for CUDA kernels under Ubuntu 12.10? (With current versions of Windows one can set the timeout limit in the registry.)

也请告诉我，是否不可能使用Ubuntu做到这一点.

Please tell me as well if there is no possibility to do this with Ubuntu.

我上一次搜索的唯一结果如下:

The only results of my previous search are the following:

• 在没有图形显示的情况下运行CUDA内核已连接到GPU
• 将内核拆分成较小的内核，以避免超过时间限制

这两种解决方案都不适合我，因为我需要图形显示 对于我的应用程序，此后已经没有任何线程在随后运行 另一个在内核执行期间.并拆分单线程将 花费数月的工作时间.请不要进一步询问这些细节 方向.

Both solutions are no option for me because I need the graphical display for my application and there are already no threads running sequently after another during the kernel is executed. And splitting the single threads will cost months of work. Please do not ask further for the details in these directions.

谢谢您的帮助.

推荐答案

您可以通过修改Xorg配置( ).该问题的答案中提供了一个示例: CUDA Visual Profiler'Interactive'X配置选项?

You can disable the watchdog by modifying your Xorg config (Option Interactive "0"). An example is available in the answer to this question: CUDA Visual Profiler 'Interactive' X config option?

这篇关于如何在Linux下禁用或更改GPU的超时限制?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！