-
Notifications
You must be signed in to change notification settings - Fork 4.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[8_0_X] Support vector<string> data products in friendly class name #36580
Conversation
Added special cases to the friendly name generator to handle std::shared_ptr and std::unique_ptr as items being held by collections which are used as data products. This avoids having an extra underscore be in friendly class name which is not allowed.
This is needed for C++11 (and above) ABI, which is available starting GCC 5. Signed-off-by: David Abdurachmanov <[email protected]>
The gcc ABI for C++11 internally uses the namespace std::__cxx11 for the standard library containers. The friendly class name generator now knows about that naming in order to keep the branch names the same as before the ABI change.
A new Pull Request was created by @makortel (Matti Kortelainen) for CMSSW_8_0_X. It involves the following packages:
@cmsbuild, @smuzaffar, @Dr15Jones, @makortel can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
@cmsbuild, please test (although this will likely fail in absence of 8_0_X IBs) |
Yup. |
Note that if we really backport this, I'd consider backporting #36458 too. |
I think this is a good solution! About the urgency, I guess we can come back to it in January (provided that all the other years work fine for the MC production, which I think they do) |
re-enable 80X IBs , see cms-sw/cmssw#36580
please test |
-1 Failed Tests: RelVals RelVals
|
The dataset |
@smuzaffar can you please move that file to the IB EOS area? Thanks! |
|
@smuzaffar I see the bot added |
+1 Failed Tests: RelVals RelVals
Comparison SummaryThere are some workflows for which there are errors in the baseline: Summary:
|
please test |
-1 Failed Tests: RelVals RelVals
|
@cmsbuild, please test We're getting bitten by randomness in the MixingModule constructor for the initial file open that checks the contents (or something). |
+1 Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-8676f1/21682/summary.html Comparison SummaryThere are some workflows for which there are errors in the baseline: Summary:
|
The tests look ok. Comparisons show differences in
but they look more like "technical issue", and this PR could not really have that kind of effect. |
(just to remind that this PR was signed already) |
+1 |
PR description:
This PR backports
as a possible solution for problem described in https://its.cern.ch/jira/browse/CMSCOMPPR-22029. In short, there a 10_6_X release is used for GEN, SIM, and DIGI steps, and 8_0_33_UL is used for the HLT step. The SIM step adds a product
std::vector<std::string>
that the HLT step is unable to read because the friendly class name (that is part of the TTree branch name) is different. The friendly class name forvector<string>
in 8_0_X is currentlycharcharstdchar_traitsstdbasic_strings
, which can not be used for persistent data products because of the underscore.An alternative solution would be to redesign the workflow to give only the RAW input to the HLT step, and using two-file solution in the RECO step to read in DIGI and HLT outputs. I'm making a backport anyway in case it would be useful (since it was rather easy to do).
PR validation:
Local test of HLT step of https://its.cern.ch/jira/browse/CMSCOMPPR-22029 succeeds.
FWCore/Utilities
unit tests pass.