Choose timezone
Your profile timezone:
ROOT Review planning
Present: The Root Team, Marco, Peter, Andreas, Bill(R), Pere, John, Andrea
Actions:
o Danilo+Fons: find out why the string header is not found by rootcling in the LHCb configuration
o Peter: Provide a callgrind profile of the workflow for which Root6 is slower
o Danilo: Check with david the status of the IBs
o Danilo: Start a build with Clang to see if this is a mismatch of compilers (clang in cling and Gcc)
Summary:
Axel:
o 6 Issues are not anymore blocking or were not related to the beta and have been moved to 6.00, 12 issues remaining.
o String header not found by LHCb builds. This is a relocation problem. See Actions.
Peter:
o Simulation seems to be a little faster with Root6
o The reco from digis is slower for Root6: maybe linked to trees in directories? A profile is needed (see actions)
o The physics performance is _the same_
Marco:
o Now in the situation of running with 14-16 failing tests
o Will provide items related to the tests one by one.
Bill:
o TFitterMinuit was removed, changes will be integrated in the IB
o Paul: progressing with the removal of Reflex in the corners where it is still present
Andrea:
o Inline asm: is a blocker for pycool. If the patch version of boost is not good for LHCb, we may not have a way out
o Still open issues with namespaces
o dbg and opt builds failing in a different way: probably linked to the presence of LLVMDEV variable set in dbg mode which causes asserts to be present.
Actions:
o Danilo+Fons: find out why the string header is not found by rootcling in the LHCb configuration
o Peter: Provide a callgrind profile of the workflow for which Root6 is slower
o Danilo: Check with david the status of the IBs
o Danilo: Start a build with Clang to see if this is a mismatch of compilers (clang in cling and Gcc)
Summary:
Axel:
o 6 Issues are not anymore blocking or were not related to the beta and have been moved to 6.00, 12 issues remaining.
o String header not found by LHCb builds. This is a relocation problem. See Actions.
Peter:
o Simulation seems to be a little faster with Root6
o The reco from digis is slower for Root6: maybe linked to trees in directories? A profile is needed (see actions)
o The physics performance is _the same_
Marco:
o Now in the situation of running with 14-16 failing tests
o Will provide items related to the tests one by one.
Bill:
o TFitterMinuit was removed, changes will be integrated in the IB
o Paul: progressing with the removal of Reflex in the corners where it is still present
Andrea:
o Inline asm: is a blocker for pycool. If the patch version of boost is not good for LHCb, we may not have a way out
o Still open issues with namespaces
o dbg and opt builds failing in a different way: probably linked to the presence of LLVMDEV variable set in dbg mode which causes asserts to be present.
There are minutes attached to this event.
Show them.