athena merge requestshttps://gitlab.cern.ch/atlas/athena/-/merge_requests2023-07-04T18:12:49+02:00https://gitlab.cern.ch/atlas/athena/-/merge_requests/64096Fix missing 2->2 vertices in Herwig7/HepMC32023-07-04T18:12:49+02:00Andrii VerbytskyiFix missing 2->2 vertices in Herwig7/HepMC3For some unknown reason, sometimes Herwig7 has missing vertices for n->m processes. In the past, it was seen for 2->1 and 1->2 and the FixHepMC algorithm had fixes for those situations. But recently it appears to be the case for the 2->2...For some unknown reason, sometimes Herwig7 has missing vertices for n->m processes. In the past, it was seen for 2->1 and 1->2 and the FixHepMC algorithm had fixes for those situations. But recently it appears to be the case for the 2->2 processes (leptoquark production).
The issue was reported here AGENE-2197, but unfortunately, it was not reproducible outside of Athena, therefore so far the way to go is to amend the FixHepMC and allow it to apply the fixes for the case of 2->2.
Tag @ewelina @dbuchin @dhirsch
Use ``urgent`` because the affected samples are needed fro an analysis in progress.https://gitlab.cern.ch/atlas/athena/-/merge_requests/63563Particlegun with HepMC32023-06-14T13:04:53+02:00Andrii VerbytskyiParticlegun with HepMC3Particlegun with HepMC3.
The particle gun is a cppyy-based python script. Because HepMC3 uses smart pointers, several adjustments are required from the point of view of syntax.
The adjustments are also tested locally dropping the Athen...Particlegun with HepMC3.
The particle gun is a cppyy-based python script. Because HepMC3 uses smart pointers, several adjustments are required from the point of view of syntax.
The adjustments are also tested locally dropping the Athena-specific code and using a script
```
import cppyy
cppyy.include('HepMC3/GenParticle.h')
cppyy.include('HepMC3/GenEvent.h')
cppyy.include('HepMC3/FourVector.h')
cppyy.include('HepMC3/GenVertex.h')
cppyy.include('HepMC3/Print.h')
cppyy.include('HepMC3/WriterAscii.h')
cppyy.load_library('libHepMC3')
from cppyy.gbl import HepMC3 as HepMC
from cppyy.gbl import std as std
from ParticleGun import ParticleGun
evt = HepMC.GenEvent()
pg = ParticleGun()
pg.fillEvent(evt)
w=HepMC.WriterAscii("1.txt")
w.write_event(evt)
```
The resulting file contains
```
HepMC::Version 3.02.06
HepMC::Asciiv3-START_EVENT_LISTING
E 0 2 4
U MEV MM
W 1.0000000000000000000000e+00
A -2 barcode -2
A -1 barcode -1
A 1 barcode 1
A 2 barcode 2
A 3 barcode 3
A 4 barcode 4
P 1 0 4 0.0000000000000000e+00 0.0000000000000000e+00 -7.0000000000000000e+03 7.0000000000000000e+03 0.0000000000000000e+00 2212
P 2 0 4 0.0000000000000000e+00 0.0000000000000000e+00 7.0000000000000000e+03 7.0000000000000000e+03 0.0000000000000000e+00 2212
V -1 0 [1,2]
P 3 -1 999 0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 1.0000000000000000e+00 1.0000000000000000e+00 11
P 4 3 999 0.0000000000000000e+00 0.0000000000000000e+00 0.0000000000000000e+00 1.0000000000000000e+00 1.0000000000000000e+00 1
HepMC::Asciiv3-END_EVENT_LISTING
```
Please note that in order to have a fully connected tree-like event, an additional ghost particle per each final state particle is needed.
Tag @ewelina @jchapman @zmarshalhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/62979GeneratorFilters: allow to keep all events, EvgenJobTransforms: fix on...2023-05-11T21:41:57+02:00Ewelina Maria LobodzinskaGeneratorFilters: allow to keep all events, EvgenJobTransforms: fix on...this MR should deal with two tasks:
1. GeneratorFilters: a property KeepAllEvents is defined in GenFilter, it is by default "false", however setting it to true in the JOs (e.g. filtSeq.xAODHTFilter.KeepAllEvents = True) allows to run a g...this MR should deal with two tasks:
1. GeneratorFilters: a property KeepAllEvents is defined in GenFilter, it is by default "false", however setting it to true in the JOs (e.g. filtSeq.xAODHTFilter.KeepAllEvents = True) allows to run a given filter, but to keep all events. It is needed to fill filter variables in the xAOD::EventInfo . It is only for HepMC3.
2. AGENE-2191: try to prevent the increase of run number when event number exceeds 2B. It will allow Generators to use full 32bit integer limit for event numbers. In addition for HepMC3 64bit limit is allowed, for HepMC2 job is crashed when the 32bit limit is exceeded.
For info @tadej @dhirsch @mgignac @hmildner @averbyts @asaibel @zmarshal @cgutschohttps://gitlab.cern.ch/atlas/athena/-/merge_requests/62641Fix for McEventCollectionCnv_p62023-04-26T13:07:02+02:00John Derek ChapmanFix for McEventCollectionCnv_p6!62300 introduced a small bug which prevented `GenEvent_p6` objects from containing more than one persistified `Attribute`. This MR fixes the bug and extends the corresponding unit test to monitor that TP conversions works as expected fo...!62300 introduced a small bug which prevented `GenEvent_p6` objects from containing more than one persistified `Attribute`. This MR fixes the bug and extends the corresponding unit test to monitor that TP conversions works as expected for `GenEvent` objects with multiple `Attributes`.
Tagging @averbyts, @ewelina, @tlari, @mbandierhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/62056Update Powheg version to 06-012023-04-03T21:42:05+02:00Ewelina Maria LobodzinskaUpdate Powheg version to 06-01updating version of Powheg used to 06-01
a label "urgent" is added, because we need a production release with this version asapupdating version of Powheg used to 06-01
a label "urgent" is added, because we need a production release with this version asaphttps://gitlab.cern.ch/atlas/athena/-/merge_requests/61914Workaround jet CA reordering by flattening the accumulator and returning depe...2023-04-19T19:25:18+02:00Tadej Novaktadej.novak@cern.chWorkaround jet CA reordering by flattening the accumulator and returning dependencies as isWorkaround jet CA reordering by introducing CA flattening:
- only for parallel sequences (we could limit ourselves even more)
- also add `popEventAlgo`
- there's one place for legacy where the CA is still unmerged, but I did not want to ...Workaround jet CA reordering by introducing CA flattening:
- only for parallel sequences (we could limit ourselves even more)
- also add `popEventAlgo`
- there's one place for legacy where the CA is still unmerged, but I did not want to refactor the config more
I did not test the output but derivations now run without complaints.
/cc @tbold @fwinkl @tamartin @wlampl @delsart @cdelitzs @jcatmore @jchapmanhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/61886EvtGen_i CMake Fix, 23.0 branch (2023.03.27.)2023-03-28T22:14:46+02:00Attila KrasznahorkayEvtGen_i CMake Fix, 23.0 branch (2023.03.27.)Removed `${Boost_INCLUDE_DIRS}` and `${Boost_LIBRARIES}`. Without the accompanying `find_package(Boost)` call, these just introduced a non-relocatable CVMFS dependency for the `EvtGen_iLib` library.
This issue was introduced in !61638. ...Removed `${Boost_INCLUDE_DIRS}` and `${Boost_LIBRARIES}`. Without the accompanying `find_package(Boost)` call, these just introduced a non-relocatable CVMFS dependency for the `EvtGen_iLib` library.
This issue was introduced in !61638. Not catching it there was understandable, nobody is to be blamed for that but me. However once the MR went in, the nightlies started complaining right away.
![image](/uploads/b4c7417a8d926e7e7656a55b593e7ddc/image.png)
(It's all the failures in the "CPack" step.) And this PROC should've really picked up on... :frowning:
@averbyts, the nightly build is only complaining about this one package. But could you check if you may have made the same mistake in other packages as well? After removing `find_package(Boost)`, you also need to remove the use of the variables that the call was setting up.https://gitlab.cern.ch/atlas/athena/-/merge_requests/60534Draft: Mark HepMC3 attributes as parsed to speedup the access2023-02-13T15:31:26+01:00Andrii VerbytskyiDraft: Mark HepMC3 attributes as parsed to speedup the accessMark HepMC3 attributes as parsed to speedup the access
Tag @jchapman @tadejMark HepMC3 attributes as parsed to speedup the access
Tag @jchapman @tadejhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/60533HepMcParticleLink: fix inefficient barcode processing when they are 02023-02-09T21:44:56+01:00Tadej Novaktadej.novak@cern.chHepMcParticleLink: fix inefficient barcode processing when they are 0Fix inefficient barcode processing in `HepMcParticleLink` when they are 0.
Probably fixes ATLASSIM-6384 fully together with !60530.
/cc @jchapman @averbyts @tlari @mbandier @martindlFix inefficient barcode processing in `HepMcParticleLink` when they are 0.
Probably fixes ATLASSIM-6384 fully together with !60530.
/cc @jchapman @averbyts @tlari @mbandier @martindlhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/60530GenEvent: use references2023-02-09T21:43:34+01:00Tadej Novaktadej.novak@cern.chGenEvent: use referencesUse references in `GenEvent` for hotspot operations. We're still hit by slow `attributes` retrieval. Will open a ticket for that soon.
Note that I did that blindly (some may be redundant), but I already see 2x the speedup in HITS merge....Use references in `GenEvent` for hotspot operations. We're still hit by slow `attributes` retrieval. Will open a ticket for that soon.
Note that I did that blindly (some may be redundant), but I already see 2x the speedup in HITS merge. Relates to ATLASSIM-6384.
/cc @averbyts @jchapman @tlari @mbandier @ewelinahttps://gitlab.cern.ch/atlas/athena/-/merge_requests/60107Update version.txt to 23.6.32023-01-26T15:43:32+01:00Ewelina Maria LobodzinskaUpdate version.txt to 23.6.3Update version.txt to 23.6.3Update version.txt to 23.6.3https://gitlab.cern.ch/atlas/athena/-/merge_requests/60093Generators: set correct units in Herwig7 interface2023-01-26T09:43:40+01:00Christian GutschowGenerators: set correct units in Herwig7 interfaceTells GenEvent about ATLAS' preferred units.
Needed for AGENE-1966
cc @ewelina @dhirsch @mgignacTells GenEvent about ATLAS' preferred units.
Needed for AGENE-1966
cc @ewelina @dhirsch @mgignachttps://gitlab.cern.ch/atlas/athena/-/merge_requests/60022Fix compilation error in xAODJetFilter2023-01-25T12:07:52+01:00Tadej Novaktadej.novak@cern.chFix compilation error in xAODJetFilterFix compilation error in `xAODJetFilter`.
/cc @averbyts, @abartonFix compilation error in `xAODJetFilter`.
/cc @averbyts, @abartonhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/59944Adding some fast barcodes2023-01-25T21:42:10+01:00Andrii VerbytskyiAdding some fast barcodes Adding some fast barcodes developed by @ssnyder in the cases when the events are created from scratch.
TAg @jchapman , @mgignac Adding some fast barcodes developed by @ssnyder in the cases when the events are created from scratch.
TAg @jchapman , @mgignachttps://gitlab.cern.ch/atlas/athena/-/merge_requests/59922fillBarcodesAttribute call in GenModule2023-01-20T21:42:40+01:00Matthew GignacfillBarcodesAttribute call in GenModuleBarcode (and vertex) maps are empty in HepMC3. This MR calls `HepMC::fillBarcodesAttribute` within GenModule, such that these are filled for all generators inheriting from GenModule. Since the `xAODTruthCnvAlg` has recently migrated to l...Barcode (and vertex) maps are empty in HepMC3. This MR calls `HepMC::fillBarcodesAttribute` within GenModule, such that these are filled for all generators inheriting from GenModule. Since the `xAODTruthCnvAlg` has recently migrated to loop over the barcode map to fill `TruthParticles` ([link](https://gitlab.cern.ch/atlas/athena/-/blob/master/Event/xAOD/xAODTruthCnv/src/xAODTruthCnvAlg.cxx#L331-333)), these empty maps resulted in empty xAOD truth containers.
Tagging @ewelina, @averbyts , @dhirschhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/59874Add barcode attribute in Epos_i for HepMC32023-01-18T15:43:03+01:00Matthew GignacAdd barcode attribute in Epos_i for HepMC3HepMC3 requires the barcode attribute to be set, which was previously missing and causing PMG validation to fail for EPOS samples (and in particular, failing since truth particles are written out in these samples, and subsequently rely o...HepMC3 requires the barcode attribute to be set, which was previously missing and causing PMG validation to fail for EPOS samples (and in particular, failing since truth particles are written out in these samples, and subsequently rely on `xAODTruthCnvAlg.cxx` which throws an error if the barcodes are missing).https://gitlab.cern.ch/atlas/athena/-/merge_requests/59763PyJobTransforms: correctly read LHE files in trfArgClasses.py2023-01-13T15:44:04+01:00Ewelina Maria LobodzinskaPyJobTransforms: correctly read LHE files in trfArgClasses.pyPyJobTransforms: python3 fix: correctly read LHE files in trfArgClasses.py, so that the number of events (and weights) are calculated and submitted to metadata correctly.
For info:
@mavogel @muschmid @dhirsch
(Mustafa: can you, please,...PyJobTransforms: python3 fix: correctly read LHE files in trfArgClasses.py, so that the number of events (and weights) are calculated and submitted to metadata correctly.
For info:
@mavogel @muschmid @dhirsch
(Mustafa: can you, please, have a more general look at the file, possibly some other read statements need decoding too...)https://gitlab.cern.ch/atlas/athena/-/merge_requests/59634GeneratorObjectsTPCnv:+AthAllocators Fix DataPool usage for HepMC3.2023-01-12T11:40:54+01:00Scott SnyderGeneratorObjectsTPCnv:+AthAllocators Fix DataPool usage for HepMC3.With HepMC3, when we make a shared_ptr from an object we're just allocated
from a DataPool, we need to ensure that there are no other shared_ptr's
referencing that object; otherwise the behavior is undefined (but can
manifest as the weak...With HepMC3, when we make a shared_ptr from an object we're just allocated
from a DataPool, we need to ensure that there are no other shared_ptr's
referencing that object; otherwise the behavior is undefined (but can
manifest as the weak references in GenParticle mysteriously expiring;
see ATR-26790).
Set up clear hooks on the DataPools so that the objects get cleared
when they're returned to the pools.https://gitlab.cern.ch/atlas/athena/-/merge_requests/59477Pythia8B_i fix to reseed the generator2022-12-21T15:43:07+01:00Ewelina Maria LobodzinskaPythia8B_i fix to reseed the generatorthe changes are needed to make the seeds for Pythia8B change,
otherwise we are getting the same events
for info @jchapmanthe changes are needed to make the seeds for Pythia8B change,
otherwise we are getting the same events
for info @jchapmanhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/59430Optimize the barcodes2022-12-21T15:42:02+01:00Andrii VerbytskyiOptimize the barcodesOptimize the sorting of particles according to barcodes.
https://its.cern.ch/jira/browse/ATLASSIM-6193
Tag @jchapman @christosOptimize the sorting of particles according to barcodes.
https://its.cern.ch/jira/browse/ATLASSIM-6193
Tag @jchapman @christos