OpenACC Hello World

▶ 在 windows 10 上搭建 OpenACC 环境

● 安装顺序：Visual Studio（PGI 18.04 不支持 VS 2017，PGI 19.04 已经支持到 VS 2019 啦！）；CUDA Toolkite（https://developer.nvidia.com/cuda-downloads）；PGI 编译器（https://www.pgroup.com/products/community.htm?*utm*_campaign=CE&utm_source=dev_nvidia_oacc&utm_medium=web_link&utm_term=*get_ce_text*）

● PGI 自带的 pgaccinfo.exe 抓取设备信息：

CUDA Driver Version:           

Device Number:
Device Name:                   GeForce GTX
Device Revision Number:        6.1
Global Memory Size:
Number of Multiprocessors:
Concurrent Copy and Execution: Yes
Total Constant Memory:
Total Shared Memory per Block:
Registers per Block:
Warp Size:
Maximum Threads per Block:
Maximum Block Dimensions:      , ,
Maximum Grid Dimensions:        x  x
Maximum Memory Pitch:          2147483647B
Texture Alignment:             512B
Clock Rate:                     MHz
Execution Timeout:             Yes
Integrated Device:             No
Can Map Host Memory:           Yes
Compute Mode:                  default
Concurrent Kernels:            Yes
ECC Enabled:                   No
Memory Clock Rate:              MHz
Memory Bus Width:               bits
L2 Cache Size:                  bytes
Max Threads Per SMP:
Async Engines:
Unified Addressing:            Yes
Managed Memory:                Yes
Concurrent Managed Memory:     No
Preemption Supported:          Yes
PGI Default Target:            -ta=tesla:cc60

● 编写代码

 #include <stdio.h>
 #include <openacc.h>

 int main()
 {
 #ifdef _OPENACC
     printf("%d device found!\n", acc_get_num_devices(acc_device_not_host));
 #else
     printf("OpenACC not support.\n");
 #endif
     getchar();
     return ;
 }

● 直接在 Visual Studio 2015 编译遇到了一些问题，最后是用 PGI 的命令行来进行编译的，这里只记录一下问题。先是要求添加 openacc.h 的路径（C:\Program Files\PGICE\win64\18.4\include），然后报错：

严重性    代码    说明                            项目            文件                                                行
错误      C1021  无效的预处理器命令“include_next”  OpenACCProject c:\program files\pgice\win64\18.4\include\sal.h         

　　这是因为在 <sal.h> 中第 28 行有  #include_next <sal.h>  ，意思是该 <sal.h> 中没有找到对应的头文件，要求预处理器去包含搜索路径的下一个 <sal.h>。但是 include_next 不是标准 C 的预处理器语句，在 Visual Studio 中无法执行（据说有办法解决，看起来有点麻烦，https://stackoverflow.com/questions/24638855/how-to-use-arduino-with-microsoft-visual-studio-due-to-include-next-precompil）。如果将该头文件中

 ...
 #if defined __PGI_TOOLS14
 #include <sal14.h>
 ...

　　前面强行添上  #include <sal4.h> （该目录下确实有 <sal14.h> 这个头文件）并删除 #include_next 行，则会引起 vadefs.h 中的相同类型的错误，再把 vadefs.h 中

 ...
 #if defined __PGI_TOOLS12
 #include <vadefs12.h>
 ...

　　前面强行添上  #include <C:/Program Files/PGICE/win64/18.4/include_acc/OT_14/vadefs.h> ，则没有了 #include_next 错误，但会有链接阶段报错：

严重性    代码        说明            项目            文件                                                行
错误      LNK1158    无法运行“rc.exe” OpenACCProject D:\Code\OpenACC\OpenACCProject\OpenACCProject\LINK      

　　说到底这是 MS 编译器不支持 OpenACC 的原因，缺少宏 _OPENACC 的定义

● 代码在 PGI 的命令行中正确的编译和执行

 PGI Community Edition 18.4
 Microsoft Windows [版本 10.0.17134.1]
 (c)  Microsoft Corporation。保留所有权利。

 C:\Users\cuan>D:

 D:\>cd D:\Code\OpenACC\OpenACCProject\OpenACCProject

 D:\Code\OpenACC\OpenACCProject\OpenACCProject>pgcc -o main-no-acc.exe main.c

 D:\Code\OpenACC\OpenACCProject\OpenACCProject>main-no-acc.exe
 OpenACC not support.

 D:\Code\OpenACC\OpenACCProject\OpenACCProject>pgcc -acc -o main.exe main.c

 D:\Code\OpenACC\OpenACCProject\OpenACCProject>main.exe
  device found!

 D:\Code\OpenACC\OpenACCProject\OpenACCProject>

▶ 在 Ubuntu16.04LTS 上搭建 OpenACC 环境

● 安装顺序：CUDA Toolkite（https://developer.nvidia.com/cuda-downloads）；PGI 编译器（https://www.pgroup.com/products/community.htm?*utm*_campaign=CE&utm_source=dev_nvidia_oacc&utm_medium=web_link&utm_term=*get_ce_text*）

cuan@CUAN:~/pgilinux-2019-194-x86-64$ sudo su
root@CUAN:/home/cuan/pgilinux-2019-194-x86-64# ./install

Welcome to the PGI Linux installer!

You are installing PGI 2019 version 19.4 for Linux.
Please note that all Trademarks and Marks are the properties
of their respective owners.

Press enter to continue...  # 回车

NVIDIA

...

Do you accept these terms? (accept,decline) # accept

...

1  Single system install
2  Network install

Please choose install option: #

...

Installation directory? [/opt/pgi] # /usr/local/pgi

...

JRE

Press enter to continue... # 回车

...

Do you accept these terms? (accept,decline) # accept

...

Do you wish to update/create links in the 2019 directory? (y/n) # y

...

MPI

...

Do you want to install Open MPI onto your system? (y/n) # y
Do you want to enable NVIDIA GPU support in Open MPI? (y/n) # y

...

License Key Management

...

Do you wish to obtain permanent license key or configure license service? (y/n) # y

...

Press enter to continue... # 回车

...

 1  Generate or update a license key for this computer
      ** Option 1 IS NOT AVAILABLE.
         Requires properly configurd curl and internet access.
 2  Configure and start a license service on this computer
 4  I have a functioning license key - what type is it? (beta)
 5  Quit

What do you want to do? #

Please enter the location of the license key file.
For example: /usr/local/pgi/license.dat
Answer? # /usr/local/pgi/license.dat

Should license services start when the system boots? (y/n) # y
Master daemon (lmgrd) is not running
Vendor daemon (pgroud) is not running

The PGI license tool can be re-started by running the script located at:

   /usr/local/pgi/linux86-64-llvm/19.4/bin/pgi_license_tool

The license key file is located at /usr/local/pgi/license.dat.

...

● 添加路径到 .bashrc，并执行 source .bashrc

#for pgi
export PATH=$PATH:/usr/local/pgi/linux86-/18.4/bin
export MANPATH=$MANPATH:/usr/local/pgi/linux86-/18.4/man
export LM_LICENSE_FILE=/usr/local/pgi/license.dat

● 编写代码同上，编译执行：

cuan@CUAN:~$ pgcc --version

pgcc 18.4- -bit target on x86- Linux -tp haswell
PGI Compilers and Tools
Copyright (c) , NVIDIA CORPORATION.  All rights reserved.
cuan@CUAN:~$ cd Temp/
cuan@CUAN:~/Temp$ pgcc -o main-no-acc.exe main.c
cuan@CUAN:~/Temp$ pgcc -acc -o main.exe main.c
cuan@CUAN:~/Temp$ ./main-no-acc.exe
OpenACC not support.

cuan@CUAN:~/Temp$ ./main.exe
 device found!

cuan@CUAN:~/Temp$ 

● 几个坑：

■ linux 上要先装了 gcc/g++ 才能安装 pgi， apt-get install build-essential

■ .bashrc 中加入 export 时，等号两边不能有空格

■ 误删了 .bashrc，从备份中去找， cp /etc/skel/.bashrc ~/

■ 在 Win10 WSL 上安装有点问题，核心问题是 Nvidia 的驱动安装不了，而 CUDA SDK 对其有依赖，强行安装 pgi 以后可以正常用 pgcc 编译，但是运行如上的代码时总是显示 0 device found! 它到底有没有调用 CUDA 核进行计算有待进一步探究，暂时折腾不动了

■ 删除 cygwin 用的 .bat，就是获得其文件的所有权。参考【https://www.cnblogs.com/litifeng/p/9996909.html】

 SET DIRECTORY_NAME="D:\cygwin64"
 C:\windows\system32\TAKEOWN /f %DIRECTORY_NAME% /r /d y
 C:\windows\system32\ICACLS %DIRECTORY_NAME% /grant administrators:F /t
 PAUSE