Mailing List picongpu-users@hzdr.de Message #288
From: Georgiana Mania georgiana.mania@desy.de <picongpu-users@hzdr.de>
Subject: Re: [PIConGPU-Users] [PIConGPU-Users] [PIConGPU-Users] [PIConGPU-Users] pic-build failure
Date: Mon, 17 Aug 2020 09:24:28 +0200
To: <picongpu-users@hzdr.de>

Hi Axel,

I've created the GitHub issue: https://github.com/ComputationalRadiationPhysics/picongpu/issues/3326

Thank you,

Georgiana

On 8/14/20 6:17 PM, Axel Huebl a.huebl@hzdr.de wrote:
Thank you,

Could be a bug, please post this output together with
cd .build
make VERBOSE=ON

output as a GitHub issue?

Thanks,
Axel

On August 14, 2020 12:04:24 AM PDT, "Georgiana Mania georgiana.mania@desy.de" <picongpu-users@hzdr.de> wrote:

Hi Axel,

Thank you for the quick response. I'm attaching the complete build log.

I haven't reported the issue on GitHub since I'm not convinced that there is a real problem with PIConGPU but rather my environment or my way of building is somehow faulty. If you consider that I should do that, please let me know and I will raise the issue today.

Best regards,

Georgiana

On 8/14/20 12:28 AM, Axel Huebl a.huebl@hzdr.de wrote:
Hi Georgiana,

Thanks for trying PIConGPU!

It looks like the boost program options library cannot be found. Can you please post the whole configuration output, especially the part before "Configuring done"?

(Ideally, report build errors on our GitHub issue tracker.)

Thanks,
Axel

On August 13, 2020 5:38:32 AM PDT, "Georgiana Mania georgiana.mania@desy.de" <picongpu-users@hzdr.de> wrote:

Hello all,


I would be interested in running a PIConGPU example on a cluster with the purpose of measuring the impact of the GPU configurations on execution time and memory footprint.

I installed PIConGPU using Spack following the online documentation. Unfortunately I cannot get it to build. I get the following error when invoking pic-build:

-- Configuring done
-- Generating done
-- Build files have been written to: /nfs/dust/atlas/user/maniageo/picongpu/.build
[  4%] Built target cupla
[ 20%] Built target cuda_memtest
[ 29%] Built target mpiInfo
[ 33%] Building CXX object CMakeFiles/picongpu-hostonly.dir/ArgsParser.cpp.o
/nfs/dust/atlas/user/maniageo/spack/opt/spack/linux-centos7-skylake_avx512/gcc-7.3.0/picongpu-0.4.3-mhq6gednwt3qidk6ncd7olohrbogiaf4/include/picongpu/ArgsParser.cpp:64:17: error: 'ArgsErrorCode' in 'class picongpu::ArgsParser' does not name a type
     ArgsParser::ArgsErrorCode ArgsParser::parse( int argc, char** argv )
                 ^~~~~~~~~~~~~
make[2]: *** [CMakeFiles/picongpu-hostonly.dir/ArgsParser.cpp.o] Error 1
make[1]: *** [CMakeFiles/picongpu-hostonly.dir/all] Error 2
make: *** [all] Error 2

ERROR: Could not successfully run make install in build directory:
       .build

Also, I'm not sure how to setup a .tpl file and some input data for the tests. I would also like to be able to set the blocksize and gridsize for my nvidia GPU but I wasn't able to find a straightforward way to do so. Could you please point me to some documentation which could help me?

Thank you very much!

Best regards,

Georgiana Mania


--
Sent from my smartphone.
Please excuse my brevity.

--
Sent from my smartphone.
Please excuse my brevity.
Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster