CTPPS: bug fix #2 in run ranges for FEDs channels mapping #18732
Conversation
|
A new Pull Request was created by @forthommel (Laurent Forthomme) for CMSSW_8_0_X. It involves the following packages: CondFormats/CTPPSReadoutObjects @ghellwig, @arunhep, @cerminar, @cmsbuild, @franzoni, @ggovi, @mmusich, @davidlange6 can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
Could somebody please launch the tests to promptly fix the issues encountered with the pilot run of the 2016H re-reco? |
|
@cmsbuild please test |
|
The tests are being triggered in jenkins. |
|
Comparison job queued. |
|
Comparison is ready Comparison Summary:
|
|
hello @forthommel Thanks! |
|
Hi @franzoni, |
|
Hi @franzoni, |
| boost::shared_ptr<TotemDAQMapping> mapping(new TotemDAQMapping()); | ||
| boost::shared_ptr<TotemAnalysisMask> mask(new TotemAnalysisMask()); | ||
| std::shared_ptr<TotemDAQMapping> mapping(new TotemDAQMapping()); | ||
| std::shared_ptr<TotemAnalysisMask> mask(new TotemAnalysisMask()); |
There was a problem hiding this comment.
@forthommel here you could use std::make_shared to have only one memory allocation instead of two and because you can avoid typing the type twice by using the auto keyword
| if (startEventID.event() == 1) | ||
| startEventID = EventID(startEventID.run(), startEventID.luminosityBlock(), 0); | ||
| if (range.startEventID()==edm::EventID(1, 0, 1)) | ||
| range = edm::EventRange(edm::EventID(1, 0, 0), range.endEventID()); |
There was a problem hiding this comment.
just for me to understand: using event number '0' has what exact special meaning?
or phrased differently, what kind of magic happens in EventRange and EventID that prevents the kind of failure mentioned by @franzoni from happening again?
There was a problem hiding this comment.
Hi @ghellwig,
It appears that the nomenclature <run_number>:min used for the run ranges in the python configuration refers to the event <run_number>:0:1 instead of <run_number>:0:0 (first event effectively observed in the stream), thus preventing any channels mapping to be associated for this first, boundary, event.
Again, this is a temporary workaround while a frontier conditions-based mapping is being developed!
cmsbuild
added
comparison-pending
db-pending
tests-pending
and removed
comparison-available
db-approved
tests-approved
labels
|
please test |
|
+1 |
|
The tests are being triggered in jenkins. |
|
Comparison job queued. |
|
Comparison is ready Comparison Summary:
|
Follow-up of #18461 (comment) to ensure a proper handling of the begin/end conditions through the usage of a
edm::contains(instead of a home-brewed check).This patch was tested on the conditions defined in this HN thread and successfully passed it.
runTheMatrix.py -l limitedyielded8 8 7 5 4 1 1 1 tests passed, 1 0 0 0 0 0 0 0 failedat 483c886 (with the usual DAS error atstep1of4.22)