CUDA Visual Profiler 'Interactive' X 配置选项? [英] CUDA Visual Profiler 'Interactive' X config option?

问题描述

在 Ubuntu 11.04 上的 computeprof 中启动应用程序时收到以下警告:

I get the following warning when launching an application in computeprof on Ubuntu 11.04:

所选计数器gld 指令 8 位"、gld 指令 16 位"、gld 指令 32 位"、gld 指令 64 位"、gld 指令 128 位"、gst 指令 8 位"、gst 指令 16 位"、gst 指令 32 位"、gst 指令 64 位"、gst 指令 128 位"可能会导致 GPU 内核运行时间超过驱动程序的看门狗超时限制.在这种情况下，驱动程序将终止 GPU 内核，从而导致应用程序错误，并且分析数据将不可用.选择这些计数器时，建议将 X Config 选项Interactive"设置为 false.

The selected counter(s) "gld instructions 8bit", "gld instructions 16bit", "gld instructions 32bit", "gld instructions 64bit", "gld instructions 128bit", "gst instructions 8bit", "gst instructions 16bit", "gst instructions 32bit", "gst instructions 64bit", "gst instructions 128bit" can cause GPU kernels to run longer than the driver's watchdog timeout limit. In this case the driver will terminate the GPU kernel resulting in an application error and the profiling data will not be available. Setting the X Config option 'Interactive' to false is recommended when these counters are selected.

我在哪里设置这个配置选项?

Where do I set this config option?

在 Ubuntu 上分析应用程序时，我也没有得到 64b 和 128b 加载和存储的任何计数，而我在 OS X 上运行分析器时看到它们.交互式"配置选项是否也会影响这一点?请注意，内核似乎没有超时；它们在默认的 30 秒超时之前完成.

I am also not getting any counts for 64b and 128b loads and stores when profiling an app on Ubuntu, whereas I see them when I run the profiler on OS X. Is the 'Interactive' config option also affecting this? Note that the kernels do not appear to be timing out; they complete well before the default timeout of 30 seconds.

这是 Ubuntu 上的 CUDA 4.0.17 和 OS X 上的 4.0.19.

This is CUDA 4.0.17 on Ubuntu and 4.0.19 on OS X.

推荐答案

这个选项可以在/etc/X11/xorg.conf中设置.只需将 Option "Interactive" "0" 添加到 GPU 的设备部分.这是我的设备部分的样子:

this option can be set in /etc/X11/xorg.conf. Just add Option "Interactive" "0" to the device section of your GPU. Here is what my device section looks like:

Section "Device"
    Identifier     "Device0"
    Driver         "nvidia"
    VendorName     "NVIDIA Corporation"
    BoardName      "GeForce GTX 460"
    Option         "Interactive" "0"
EndSection

我在 ubuntu 中运行相同版本的 CUDA，但我使用的是 OpenCL.使用这些指令之一进行分析会导致 CPU 使用率很高并且永远不会完成(我必须终止进程) - 即使禁用交互模式也是如此.不管怎样，你可以试一试.

I run the same version of CUDA in ubuntu but I'm using OpenCL. Profiling with one of these instructions causes high CPU usage and never finishes (I have to kill the process) - even when interactive mode is disabled. Anyways, you can give it a try.

还要确保验证 GPU 的计算能力并检查是否支持请求的计数:Visual Profiler 用户指南.虽然这似乎不是你问题的原因，因为它在 OS X 中工作.

Also make sure to verify the compute capability of your GPU and check if the requested count is supported : Visual Profiler User Guide. Although it seems that this is not the cause of you problem since it's working in OS X.

这篇关于CUDA Visual Profiler 'Interactive' X 配置选项?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！