search

antonmks / Alenka

GPU database engine
Other
1.17k stars 120 forks source link

Compilation time very long #110

Closed archenroot closed 7 years ago

archenroot commented 7 years ago

Hi buddy,

I am on Gentoo machine with:

I understand I am probably running with latest and therefore not currently supported nvcc, but compilation is taking about 30 minutes and not yet finished.

archenroot commented 7 years ago

What is confirmed CUDA version to work?

archenroot commented 7 years ago

Ok, finally the compilation finished, but it looks I have some problem with modernGPU library. Alenka is failing and I tried to compile the library itself and run some included tests, the error is: terminate called after throwing an instance of 'mgpu::cuda_exception_t' what(): unknown error

I will ask on modern GPU github issues page.

archenroot commented 7 years ago

The note at modernGPU says: moderngpu 2.0 is designed for CUDA 7.5+. It runs on all generations of NVIDIA GPUs from sm_20 (Fermi) and up.

The reference environment is 64-bit Linux with g++ 4.9.3 with nvcc 7.5.17. This library and code using it might not compile on Microsoft Visual Studio 2013 because of symbol length limitations in that compiler. Windows users should try upgrading to CUDA 8.0 and using Visual Studio 2015.

So I try to compile lower gcc itself, I already additionally installed 5.4.0, but will try 4.* as well.

archenroot commented 7 years ago

Ok, I fixed issue by reinstalling nvidia drivers, finally I am compiling with GCC 6, I override the limitation manually in CUDA SDK header file.

Compilation time is probably just fine, I might have a look Makefile and execute different commands with CMAKE in parallel, that is the only optimization possible in my head.