US ATLAS HL-LHC computing discussion

["Dong, Zhihua" <zdong AT bnl.gov>]

I only used gcc6

gcc8 may have problem with cuda.

 

I got the whole tarball from Tadej,     CUDA part , I used  

find_package(CUDA)

Never specify C++ standard in cmake files for nvcc .  

 

Maybe we should  set

CMAKE_CXX_STANDARD  to 11 or 14

Does  cmake   -DCMAKE_CXX_STANDARD=11 .. work ?

 

 

 

 

 

Zhihua Dong

Computational Science Initiative

Brookhaven National Laboratory

631-344-4141

 

Be Veg Go Green 2 Save the Planet

www.SupremeMasterTV.com

 

 

From: Usatlas-hllhc-computing-l <usatlas-hllhc-computing-l-bounces AT lists.bnl.gov> on behalf of Charles Leggett <cgleggett AT lbl.gov>
Date: Thursday, August 22, 2019 at 4:18 PM
To: "Lin, Meifeng" <mlin AT bnl.gov>
Cc: "usatlas-hllhc-computing-l AT lists.bnl.gov" <usatlas-hllhc-computing-l AT lists.bnl.gov>
Subject: Re: [Usatlas-hllhc-computing-l] data files

 

hi Meifeng -

 

I have not yet gotten hold of them.

 

Also, does anyone know who wrote the cmake / build infrastructure? The nvcc compile line for the cuda kernels is using std=c++1z, and I need to change that to std=c++11, but I'm not sure where it's set (c++1z causes compilation errors with std lib mismatches when using gcc 8.2)

 

                 cheers,  Charles.

 

 

 

On Thu, Aug 22, 2019 at 1:09 PM Lin, Meifeng <mlin AT bnl.gov> wrote:

Hi Charles,

Have you resolved the issue yet? I think the original input files did come from cvmfs. We don't have
access so we made a copy in our own directory. It should be easy for you to get these input data
files.

Perhaps Tadej can point you to the right place?

Thanks,

Meifeng

On Fri, 2019-08-09 at 09:03 -0700, Charles Leggett wrote:
>
> Also, I'm having issues with the compilation. I have a centos7 system. For centos7, the available
> root versions are compiled with gcc8 (from showVersions root). The code can't be compiled with
> gcc8 as there seem to be some incompatibilities with the stl headers. I will investigate further.
>
> cheers,  Charles.
>
> On Fri, Aug 9, 2019 at 8:45 AM Charles Leggett <cgleggett AT lbl.gov> wrote:
> > Hi all -
> >
> >  I'm looking at the build instructions and it says:
> > > cmake ../FastCaloSimAnalyzer -DENABLE_XROOTD=off -DENABLE_GPU=on
> > -DINPUT_PATH="/hpcgpfs01/work/csi/cce/FastCaloSimInputs"
> >
> > is what's in /hpcgpfs01/work/csi/cce/FastCaloSimInputs available somewhere else like on cvmfs or
> > eos?
> >
> >          cheers, Charles.
> >
> >
> > --
> > Charles Leggett <CGLeggett AT lbl.gov>
> > Lawrence Berkeley National Lab
> > 1 Cyclotron Road, MS 50F-1650, Berkeley, CA 94720
> > tel: 510.495.2930
>
>
> _______________________________________________
> Usatlas-hllhc-computing-l mailing list
> Usatlas-hllhc-computing-l AT lists.bnl.gov
> https://lists.bnl.gov/mailman/listinfo/usatlas-hllhc-computing-l

_______________________________________________
Usatlas-hllhc-computing-l mailing list
Usatlas-hllhc-computing-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/usatlas-hllhc-computing-l

--

Charles Leggett <CGLeggett AT lbl.gov>
Lawrence Berkeley National Lab
1 Cyclotron Road, MS 50F-1650, Berkeley, CA 94720

tel: 510.495.2930