Skip to Content.
Sympa Menu

phys-npps-mgmt-l - Re: [Phys-npps-mgmt-l] Draft program development slides

phys-npps-mgmt-l AT lists.bnl.gov

Subject: NPPS Leadership Team

List archive

Chronological Thread  
  • From: pinkenburg <pinkenburg AT bnl.gov>
  • To: Brett Viren <bv AT bnl.gov>, pinkenburg via Phys-npps-mgmt-l <phys-npps-mgmt-l AT lists.bnl.gov>
  • Subject: Re: [Phys-npps-mgmt-l] Draft program development slides
  • Date: Wed, 10 Jul 2019 09:05:30 -0400

Hi Brett,

yes we are at the forefront in computing in rcf :)

There was some more back and forth this morning and yes it is the compiler (clang even newer version still uses the ancient gcc headers). They pointed me to their compiler collection in cvmfs which works. I am not too sure how to go from here. This was an academic exercise and sPHENIX per se has no need for a newer compiler (not opposed to it but it complicates things for users)

Chris


On 7/10/2019 8:42 AM, Brett Viren wrote:
pinkenburg via Phys-npps-mgmt-l <phys-npps-mgmt-l AT lists.bnl.gov> writes:

Anything new will take years to develop. ejana is an empty shell and
still in the "does not even compile" stage, DD4hep also doesn't
compile under SL7. I had some back and forth with the authors (who
responded within minutes to the issue raised which is good) and it
looks like gcc 4.8 is not supported (not completely C++11
compliant). You need at least 4.9 but clang 7.0.0 which should be
compliant also did not work, not sure what this means.
GCC 4 is ancient. (As comparison, C++17 is the accepted standard in
neutrino software. GCC 7 and 8 are used.)

Does this DD4hep example point to a more general and fundamental problem
in the EIC world? One of reliance on ancient OS-provided compilers? Is
there a need for EIC to provide developer and user runtime environments
for which EIC has control over versions?

If the answers are on the positive side of "maybe", there might be
fruitful collaboration between NP and HEP through the use of Spack both
to build binaries and to share them via CVMFS.

-Brett.

--
*************************************************************

Christopher H. Pinkenburg ; pinkenburg AT bnl.gov
; http://www.phenix.bnl.gov/~pinkenbu

Brookhaven National Laboratory ; phone: (631) 344-5692
Physics Department Bldg 510 C ; fax: (631) 344-3253
Upton, NY 11973-5000

*************************************************************





Archive powered by MHonArc 2.6.24.

Top of Page