Home People Research Blog Courses Links Search Download
NCHC

Blog

Blog 最新文章

  1. Visual Studio 2019 16.9 支援使用 OpenMP LLVM
    2021/03/03 13:52
  2. Valve 推出完全支援 OpenXR 的 SteamVR 1.16.8
    2021/02/25 09:43
  3. 讓 Oculus Quest 2 變 PC VR:Oculus Link
    2021/02/22 11:25

Blog 最新回應

  1. 加入斜體文字...
    2021/02/07 21:06
  2. 加入斜體文字...
    2021/02/07 21:06
  3. 加入斜體文字...
    2021/02/07 21:06

Keyword 關鍵字

OpenMP Pandas OpenCL Vulkan 資料視覺化 資訊地圖 PHP C++14 VR OpenGL Kinect 3d print C++17 Python Docker git NiTE2 開放資料 Qt Java C++11 C++14 HoloLens 2 OpenXR Oculus Windows MR OpenCV C++ MR HTC Vive xml HTC Vive Focus 開放資料 Python OpenNI iFlyover OpenVR CubeX WebGL svn Oculus Quest 2 OpenNI2 Boost C++20 VR Pandas 3D立體 ASUS Xtion Docker CUDA Valve Index GitLab Oculus Rift S javascript HTC Vive Pro

類別:技術相關 » 技術研究
文章發表|我要回應|RSS訂閱

CUDA 2.2 Release 開放下載!

四月初的時候,有貼過 CUDA 2.2 Beta 正在進行「封測」,開放給有註冊的開發者下載;而在五月七號,nVidia 則是放出 CUDA 2.2 的正式版提供所有人下載了∼詳細細節可以參考官方論壇的《CUDA Toolkit and SDK v2.2 released》一文。

官方的新功能列表如下:

  • Officially adds support for Windows 7, Server 2003, Server 2008, Ubuntu 8.10, RHEL 5.3, and Fedora 10
  • Includes cuda-gdb (hardware GPU debugger) for RHEL5 32 and 64-bit (officially supported and tested), but it may work on more platforms than just those
  • Exclusive device mode in Linux: set some GPUs as exclusive-compute (can only own a single CUDA context) and some as non-compute (no CUDA contexts allowed) for easier management of clusters/MPI applications. See the manpage for nvidia-smi for how to set this and cudaSetValidDevices in the reference manual on how to best use this from CUDART.
  • Zero-copy support: transparently read from certain system memory from a kernel on GT200 or MCP79 systems. See this post for more information on how it works.
  • Asynchronous memcpy support on Vista/Server 2008/Win7
  • Texture from pitchlinear memory: use this to avoid an additional memcpy at times in some scenarios.
  • >4GB of pinned memory in a single allocation on most OSes
  • maximum pinned memory per allocation increased in Vista to ~1.5GB
  • pinned memory can be shared between contexts
  • Multi-device OpenGL interop performance between a Quadro display card and a separate compute card is dramatically improved.
  • Visual Profiler works on Vista
  • Visual Profiler supports additional counters for GT200 to measure number of memory transactions of a given size, instruction throughput, etc.
  • Blocking sync support for all platforms: allows the host thread to sleep and be awoken by driver when the GPU operation the host thread is waiting on is completed.
  • Quite a few additional math functions added due to forum requests (feel free to keep posting requests, we do pay attention)
  • __threadfence(): ensure that a thread's pending memory writes are visible to all threads before continuing. It is explicitly not a global sync, unlike how it appears to some.
  • Lots of bugfixes, of course; most importantly, killing a CUDA app should behave much, much better than it ever has before, especially when you're on a dedicated compute card

下載頁面:http://www.nvidia.com/object/cuda_get.html;或是直接下載:

  • 驅動程式
    • Windows 185.85
      • for Windows XP 32bit:桌上型/工作站、筆記型
      • for Windows XP 64bit:桌上型/工作站、筆記型
      • for Windows Vista 32bit:桌上型/工作站、筆記型
      • for Windows Vista 64bit:桌上型/工作站、筆記型
      • for Windows 7 32bit:桌上型/工作站、筆記型
      • for Windows 7 64bit:桌上型/工作站、筆記型
    • Linux 185.18.08:32bit OS、64bit OS
  • CUDA Toolkit 2.2
    • for Windows:32bit、64bit
    • for Fedora 9:32bit、64bit
    • for Fedora 10:32bit、64bit
    • for RHEL 4.7:32bit、64bit
    • for RHEL 5.3:32bit、64bit
    • for SLED 10 SP2:32bit、64bit
    • for SUSE 11:32bit、64bit
    • for SUSE 11.1:32bit、64bit
    • for Ubuntu 8.04:32bit、64bit
    • for Ubuntu 8.10:32bit、64bit
    • for MacOS 10.5
    • CUDA Toolkit 2.2 EULA
  • SDK 2.2
    • for Windows:32bit、64bit
    • for Linux
    • for Mac
  • Debugger:
    • CUDA-GDB for RHEL 5.3 32bit / 64bit
    • CUDA-GDB 2.2 User's Manual
  • Visual Profiler
    • for MacOS 10.5(其他平台包含在 Toolkit 內)
    • CUDA Visual Profiler 1.2 Readme
    • CUDA Visual Profiler EULA
  • Documentation
    • CUDA 2.2: Getting Started in Linux / Mac OS / Windows
    • CUDA 2.2 Release Notes for Linux / Mac OS / Windows
    • CUDA 2.2 API Reference Manual
    • CUDA 2.2 Programming Guide
張貼者:heresy於2009/05/12 10:00 上午有4則回應,瀏覽次數:2,874次
STAN 於 2009/06/19 09:13 上午 回應:
I get error when I am installing the CUDA 2.2 SDK in Open SUSE11.1. 1. The driver 185.18.08 is installed OK.2. The CUDA toolket 2.2 is installed OK. 3. The CUDA SDK installing with problem in "make". make[1]: *** [objdebug/rendercheck_gl.cpp.o] error1 make: *** [lib/librendercheckgl.so] error2please help. Thanks!!
作者作者heresy 於 2009/06/19 13:18 下午 回應:
Soory, I don't have OpenSUSE...
STAN 於 2009/06/20 00:38 上午 回應:
Could I know what OS you did install with CUDA 2.2?Thanks!
作者作者heresy 於 2009/06/22 09:45 上午 回應:
We use Windows x64 with CUDA.

-- TOP --

我要回應
* 身份  訪客 (暱稱:)
 本篇文章作者 (帳號:密碼:)
* 內容      
很高興 悲傷 震驚 疑惑 大笑 發瘋 傷心
* 留言密碼 (請輸入下方圖片中去除前、後位數的數字,共五碼。)
說明 1. * 表示必填欄位。
2. 不支援HTML Tag。
   

-- TOP --

© Visualization and Interactive Media Laboratory of NCHC, 2007 - 2021, All Rights Reserved. Contact E-mail