Nvcc error: Only when targeting sm_20 or higher (Bug #3827)


Added by hoa Hong over 10 years ago. Updated over 9 years ago.


Status:New Start date:2014-07-24
Priority:Normal Due date:
Assignee:hoa Hong % Done:

0%

Category:gpu (cuda)
Target version:-
Affected version:branch 'master' (3.0-dev) Operating System:Windows
Difficulty: HW Platform:x86
Pull request:

Description

Hi,

I'm building latest opencv 3.0 from source with CUDA SDK 5.5. My graphics card is Gigabyte NVidia GeForce 9800GTX+ with 1.1 computability. When building opencv_core module, i get this error:

"*nvcc : fatal error : nvcc supports '--relocatable-device-code=true (-rdc=true)' , '--device-c (-dc)' and '--device-link (-dlink)' only when targeting sm_20 or higher*"

CMake settings:
CUDA_ARCH_BIN: 1.1 1.2 1.3 2.0 2.2(2.0) 3.0.3.5
CUDA_ARCH_PTX: 3.0

Enviroment and tools:
Visual Studio C++ 2012
Windows 7

I changed CUDA_ARCH_BIN and CUDA_ARCH_PTX but i seems that doesn't affect.

Is it possible a bug?


Associated revisions

Revision f80e7530
Added by Vadim Pisarevsky almost 10 years ago

Merge pull request #3827 from MSOpenTech:license-26

History

Updated by Steven Puttemans over 10 years ago

Since 3.0 is a development branch OpenCV does not guarantee it will build without problems. I experience similar GPU building problems with 3.0 compute capability, seems this is indeed an unfixed bug!

Updated by hoa Hong over 10 years ago

Steven Puttemans wrote:

Since 3.0 is a development branch OpenCV does not guarantee it will build without problems. I experience similar GPU building problems with 3.0 compute capability, seems this is indeed an unfixed bug!

Do you know how to fix it?

Updated by Steven Puttemans over 10 years ago

Nope :) My GPU experience is limited to using the functions :)

Updated by Vadim Pisarevsky almost 10 years ago

  • Category changed from core to gpu (cuda)

Updated by Maksim Shabunin over 9 years ago

Issue has been transferred to GitHub: https://github.com/Itseez/opencv/issues/4795

Also available in: Atom PDF