Microsoft MVP성태의 닷넷 이야기
VC++: 80. 내 컴퓨터에서 C++ AMP 코드가 실행이 될까요? [링크 복사], [링크+제목 복사],
조회: 17367
글쓴 사람
정성태 (techsharer at outlook.com)
홈페이지
첨부 파일
 

내 컴퓨터에서 C++ AMP 코드가 실행이 될까요?

공식 홈페이지를 보면,

C++ AMP Overview
; https://docs.microsoft.com/en-us/cpp/parallel/amp/cpp-amp-overview

시스템 요구 사항이 다음과 같이 나옵니다.

  • Windows 7, Windows 8, Windows Server 2008 R2, or Windows Server 2012
  • DirectX 11 Feature Level 11.0 or later hardware
  • For debugging on the software emulator, Windows 8 or Windows Server 2012 is required. For debugging on the hardware, you must install the drivers for your graphics card.

이런 거 보고 바로 알 수 있으면 좋겠지만 역시나 그냥 프로그램 돌려서 되냐 안되냐를 알 수 있는 것이 속편합니다. ^^ 이에 대해서는 다음의 글에 나오는 "VerifyAmpDevices" 유틸리티를 쓰시면 됩니다.

Can I Run C++ AMP on My Device?
; http://blogs.msdn.com/b/nativeconcurrency/archive/2011/09/22/can-i-run-c-amp-on-my-device.aspx

VerifyAmpDevices 유틸리티는 아래의 글에 포함된 list_all_accelerators() 함수를 이용해 쉽게 만들 수 있습니다.

Using accelerator and accelerator_view Objects
; https://docs.microsoft.com/en-us/cpp/parallel/amp/using-accelerator-and-accelerator-view-objects

뭐 대충 이런 식입니다.

#include "stdafx.h"

#include <amp.h>
#include <vector>
#include <amprt.h>

#include <iostream>

using namespace Concurrency;

// https://docs.microsoft.com/en-us/cpp/parallel/amp/using-accelerator-and-accelerator-view-objects
void list_all_accelerators();

int _tmain(int argc, _TCHAR* argv[])
{
    list_all_accelerators();

    return 0;
}

void list_all_accelerators()
{
    std::vector<accelerator> accs = accelerator::get_all();

    for (int i = 0; i < accs.size(); i++) {
        std::wcout << accs[i].device_path << ", desc = ";
        std::wcout << accs[i].description << ", has_display = ";
        std::wcout << accs[i].get_has_display() << ", is_emulated = ";
        std::wcout << accs[i].is_emulated << "\n";

        //std::wcout << accs[i].dedicated_memory << "\n";
        //std::wcout << (accs[i].supports_cpu_shared_memory ?
        //  "CPU shared memory: true" : "CPU shared memory: false") << "\n";
        //std::wcout << (accs[i].supports_double_precision ?
        //  "double precision: true" : "double precision: false") << "\n";
        //std::wcout << (accs[i].supports_limited_double_precision ?
        //  "limited double precision: true" : "limited double precision: false") << "\n\n";
    }
}

제 컴퓨터에서 위의 코드를 실행하면 다음과 같은 출력 결과가 나옵니다.

PCI\...[장치경로]..., desc = NVIDIA GeForce GTX 660  , has_display = 1, is_emulated = 0
direct3d\warp, desc = Microsoft Basic Render Driver, has_display = 0, is_emulated = 1
direct3d\ref, desc = Software Adapter, has_display = 1, is_emulated = 1
cpu, desc = CPU accelerator, has_display = 0, is_emulated = 1

당연히 is_emulated = 0인 값의 장치가 가장 성능이 우수할 것입니다. 또한 has_display는 현재 모니터 장치와 연결된 것을 의미한다고 하는데, direct3d\ref가 has_display = 1을 갖는 의미가 뭔지는 해석이 안되는군요. ^^

참고로, 중간의 warp와 ref에 대해서는 다음의 글에 약간의 설명이 나옵니다.

concurrency::accelerator 
; http://www.danielmoth.com/Blog/concurrencyaccelerator.aspx

  • accelerator::direct3d_ref represents the reference rasterizer emulator that simulates a direct3d device on the CPU (in a very slow manner). This emulator is available on systems with Visual Studio installed and is useful for debugging. More on debugging in general in future posts. Example: accelerator acc(accelerator::direct3d_ref);
  • accelerator::direct3d_warp represents WARP which is the current CPU fallback. Example: accelerator acc(accelerator::direct3d_warp);

ref는 Visual Studio가 설치된 경우에 있으며 디버깅이 쉽다는 장점이 있다고 하니, 그렇다면 일반적인 컴퓨터라면 GPU 하나와 direct3d\ref, cpu 정도만 AMP 장치로 검색될 것입니다.

아래의 글도 읽어보면 좋을 듯 싶군요. ^^

Server environments and C++ AMP
; http://blogs.msdn.com/b/nativeconcurrency/archive/2012/06/21/server-environments-and-c-amp.aspx





[이 글에 대해서 여러분들과 의견을 공유하고 싶습니다. 틀리거나 미흡한 부분 또는 의문 사항이 있으시면 언제든 댓글 남겨주십시오.]







[최초 등록일: ]
[최종 수정일: 7/17/2021]

Creative Commons License
이 저작물은 크리에이티브 커먼즈 코리아 저작자표시-비영리-변경금지 2.0 대한민국 라이센스에 따라 이용하실 수 있습니다.
by SeongTae Jeong, mailto:techsharer at outlook.com

비밀번호

댓글 작성자
 



2022-05-04 04시52분
VS2022부터 C++ AMP 공식 지원중단
; https://blog.naver.com/drvoss/222653027219
정성태

... 16  17  18  19  20  21  22  23  24  25  26  27  28  29  [30]  ...
NoWriterDateCnt.TitleFile(s)
12890정성태12/23/20217131VC++: 152. Golang - (문자가 아닌) 바이트 위치를 반환하는 strings.IndexRune 함수
12889정성태12/22/20219564.NET Framework: 1123. C# - (SharpDX + DXGI) 화면 캡처한 이미지를 빠르게 JPG로 변환하는 방법파일 다운로드1
12888정성태12/21/20217668.NET Framework: 1122. C# - ImageCodecInfo 사용 시 System.Drawing.Image와 System.Drawing.Bitmap에 따른 Save 성능 차이파일 다운로드1
12887정성태12/21/20219800오류 유형: 777. OpenCVSharp4를 사용한 프로그램 실행 시 "The type initializer for 'OpenCvSharp.Internal.NativeMethods' threw an exception." 예외 발생
12886정성태12/20/20217615스크립트: 37. 파이썬 - uwsgi의 --enable-threads 옵션 [2]
12885정성태12/20/20217881오류 유형: 776. uwsgi-plugin-python3 환경에서 MySQLdb 사용 환경
12884정성태12/20/20216921개발 환경 구성: 620. Windows 10+에서 WMI root/Microsoft/Windows/WindowsUpdate 네임스페이스 제거
12883정성태12/19/20217824오류 유형: 775. uwsgi-plugin-python3 환경에서 "ModuleNotFoundError: No module named 'django'" 오류 발생
12882정성태12/18/20216934개발 환경 구성: 619. Windows Server에서 WSL을 위한 리눅스 배포본을 설치하는 방법
12881정성태12/17/20217395개발 환경 구성: 618. WSL Ubuntu 20.04에서 파이썬을 위한 uwsgi 설치 방법 (2)
12880정성태12/16/20217247VS.NET IDE: 170. Visual Studio에서 .NET Core/5+ 역어셈블 소스코드 확인하는 방법
12879정성태12/16/202113511오류 유형: 774. Windows Server 2022 + docker desktop 설치 시 WSL 2로 선택한 경우 "Failed to deploy distro docker-desktop to ..." 오류 발생
12878정성태12/15/20218541개발 환경 구성: 617. 윈도우 WSL 환경에서 같은 종류의 리눅스를 다중으로 설치하는 방법
12877정성태12/15/20217197스크립트: 36. 파이썬 - pymysql 기본 예제 코드
12876정성태12/14/20217027개발 환경 구성: 616. Custom Sources를 이용한 Azure Monitor Metric 만들기
12875정성태12/13/20216704스크립트: 35. python - time.sleep(...) 호출 시 hang이 걸리는 듯한 문제
12874정성태12/13/20216717오류 유형: 773. shell script 실행 시 "$'\r': command not found" 오류
12873정성태12/12/20217856오류 유형: 772. 리눅스 - PATH에 등록했는데도 "command not found"가 나온다면?
12872정성태12/12/20217678개발 환경 구성: 615. GoLang과 Python 빌드가 모두 가능한 docker 이미지 만들기
12871정성태12/12/20217735오류 유형: 771. docker: Error response from daemon: OCI runtime create failed
12870정성태12/9/20216306개발 환경 구성: 614. 파이썬 - PyPI 패키지 만들기 (4) package_data 옵션
12869정성태12/8/20218611개발 환경 구성: 613. git clone 실행 시 fingerprint 묻는 단계를 생략하는 방법
12868정성태12/7/20217184오류 유형: 770. twine 업로드 시 "HTTPError: 400 Bad Request ..." 오류 [1]
12867정성태12/7/20216855개발 환경 구성: 612. 파이썬 - PyPI 패키지 만들기 (3) entry_points 옵션
12866정성태12/7/202114268오류 유형: 769. "docker build ..." 시 "failed to solve with frontend dockerfile.v0: failed to read dockerfile ..." 오류
12865정성태12/6/20216940개발 환경 구성: 611. 파이썬 - PyPI 패키지 만들기 (2) long_description, cmdclass 옵션
... 16  17  18  19  20  21  22  23  24  25  26  27  28  29  [30]  ...