upgrade EvtGen interface for v03.00.00 #51029
Conversation
|
cms-bot internal usage |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49473 ERROR: Build errors found during clang-tidy run. |
| <use name="FWCore/Concurrency"/> | ||
| <use name="FWCore/ParameterSet"/> | ||
| <use name="FWCore/Framework"/> | ||
| <use name="SimG4Core/Application"/> |
There was a problem hiding this comment.
(as I commented already in the earlier issue) I think this dependence on SimG4Core should be avoided. I opened #51041 to move the ThreadHandoff to FWCore/Utilities.
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49482 ERROR: Build errors found during clang-tidy run. |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49483 ERROR: Build errors found during clang-tidy run. |
Right...will do |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49486 ERROR: Build errors found during clang-tidy run. |
|
@kawaho , please update this PR to avoid |
|
#51041 is now merged and available in IBs since CMSSW_17_0_X_2026-05-27-2300. |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49555 ERROR: Build errors found during clang-tidy run. |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49582 ERROR: Build errors found during clang-tidy run. |
|
code-checks with cms.week1.PR_fc95d40b/100.0-b40a4405c86ba87a5ae00696120942d1 |
|
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51029/49583
Code check has found code style and quality issues which could be resolved by applying following patch(s)
|
PR description:
Resolves #50935
This commits upgrade EvtGen interface to v03.00.00 in CMSSW, which
This commit depends on a new cmssw external build cms-sw/cmsdist#10580
PR validation:
Tested by producing a small TTTo2L2Nu Pythia+EvtGen sample locally with multiple threads, using a similar the same fragment as 1 but with PythiaConcurrentHadronizerFilter.
One remaining issue is that the EvtGen interface now calls ThreadHandoff implemented for GEANT4 to ensure a stream started in a certain thread stays in the same thread. Shall we move ThreadHandoff under
FWCore/Concurrencyinstead, so packages with similar concurrency implementation (one instance per thread/use of thread_local) can use the same module? @makortel @Dr15Jones, please advise.