athena merge requestshttps://gitlab.cern.ch/atlas/athena/-/merge_requests2021-11-30T21:44:33+01:00https://gitlab.cern.ch/atlas/athena/-/merge_requests/48667Powheg: update to version 05-022021-11-30T21:44:33+01:00Ewelina Maria LobodzinskaPowheg: update to version 05-02Powheg: update to version 05-02Powheg: update to version 05-02https://gitlab.cern.ch/atlas/athena/-/merge_requests/48599Update BeamHaloGeneratorAlg.cxx to fix a warning2021-11-30T18:20:15+01:00Andrii VerbytskyiUpdate BeamHaloGeneratorAlg.cxx to fix a warningFix a warning ATLASSIM-5512Fix a warning ATLASSIM-5512https://gitlab.cern.ch/atlas/athena/-/merge_requests/48156Removing loop parameter from default params2021-11-29T21:44:32+01:00Hannes MildnerRemoving loop parameter from default paramsNo longer setting the (now hidden) "loop" parameter in the param card.
For MadLoop, this parameter corresponds to the renormalization scale. For event generation, the renormalization scale is set via the run card, so this parameter shoul...No longer setting the (now hidden) "loop" parameter in the param card.
For MadLoop, this parameter corresponds to the renormalization scale. For event generation, the renormalization scale is set via the run card, so this parameter shouldn't matter for us anyway.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48543Crmc File Installation Fix, master branch (2021.11.25.)2021-11-26T15:43:56+01:00Attila KrasznahorkayCrmc File Installation Fix, master branch (2021.11.25.)I made way too many mistakes on this one. :frowning: This is a follow-up to !48209 and !48243.
This time around I made sure that the `find_file(...)` commands would start from scratch. Since as it turned out, [find_file](https://cmake.o...I made way too many mistakes on this one. :frowning: This is a follow-up to !48209 and !48243.
This time around I made sure that the `find_file(...)` commands would start from scratch. Since as it turned out, [find_file](https://cmake.org/cmake/help/latest/command/find_file.html) respects existing settings on a non-cache variable as well. So to make sure that it would search for all input files correctly, we have to explicitly unset any pre-existing value on that variable...
Mea culpa @ewelina. :frowning2: These updates caused way more trouble than they were worth...https://gitlab.cern.ch/atlas/athena/-/merge_requests/48524Create weight index map function for GenEvent2021-11-25T21:43:26+01:00Andrii VerbytskyiCreate weight index map function for GenEventCreate weight index map function for GenEvent.Create weight index map function for GenEvent.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48243Crmc Usage Update, master branch (2021.11.15.)2021-11-25T16:48:44+01:00Attila KrasznahorkayCrmc Usage Update, master branch (2021.11.15.)While working on !48240, I bumped into this error message:
```
-- Configuring done
-- Generating done
CMake Error:
Running
'/opt/ninja/1.10.2/Linux-x86_64/ninja' '-C' '/atlas/krasznaa/ATLASRECTS-6680/build' '-t' 'recompact'
fai...While working on !48240, I bumped into this error message:
```
-- Configuring done
-- Generating done
CMake Error:
Running
'/opt/ninja/1.10.2/Linux-x86_64/ninja' '-C' '/atlas/krasznaa/ATLASRECTS-6680/build' '-t' 'recompact'
failed with:
ninja: error: build.ninja:206928: multiple rules generate x86_64-centos7-gcc8-opt/share/epos.initl [-w dupbuild=err]
CMake Generate step failed. Build files cannot be regenerated correctly.
```
After some debugging I realised that this was because I was using CMake 3.20.2 in that build. And what I didn't realise in !48209 was that the `NO_CACHE` variable for [find_file(...)](https://cmake.org/cmake/help/latest/command/find_file.html) was only introduced with CMake 3.21. :frowning:
I was considering for a moment how I could change the code in the packages to work correctly with older CMake versions, but all solutions that I could come up with, looked pretty ugly. So I thought this would be the simplest. To just make it explicit that we need at least 3.21 for these two packages.
Hopefully this will work fine with all nightlies. :thinking: I believe @aundrus has switched all master nightlies to 3.21.3 by now... (Which is why the nightlies didn't freak out about !48209.)
Pinging @fwinkl for information.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48209Crmc Compatibility Update, master branch (2021.11.12.)2021-11-25T16:48:44+01:00Attila KrasznahorkayCrmc Compatibility Update, master branch (2021.11.12.)Made [Epos_i](Generators/Epos_i) and [QGSJet_i](Generators/QGSJet_i) smarter with installing files from `Crmc`. This is necessary to find those files in different versions of `Crmc`.
This is a follow-up to !48181. It is meant to fix the...Made [Epos_i](Generators/Epos_i) and [QGSJet_i](Generators/QGSJet_i) smarter with installing files from `Crmc`. This is necessary to find those files in different versions of `Crmc`.
This is a follow-up to !48181. It is meant to fix the current CMake configuration issue in the `master--dev4LCG` nightly.
https://atlas-computing.web.cern.ch/atlas-computing/links/distDirectory/gitwww/MASTERWebArea/ardoc_web_areamasterdev4LCG64BC7G11AthenaOpt/ARDOC_Log_2021-11-12T0635/ardoc_cmake_config.html
As a note to the shifter: Discipline around notifying me of MRs with non-trivial CMake changes is pretty lax these days. :frowning: I would've loved to be notified of !48181...
This is also tied to [SPI-2024](https://sft.its.cern.ch/jira/browse/SPI-2024), and https://gitlab.cern.ch/atlas/atlasexternals/-/merge_requests/890. But it can be merged in independently from the latter.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48314Add weight name to Pythia8B interface2021-11-25T15:42:20+01:00Andrii VerbytskyiAdd weight name to Pythia8B interfaceAdd weight name to Pythia8B interface.
Tag @jroggel @ewelina
This should add weight to P8B_A14NNPDF23LO_pp_JpsimumuZmumu in
https://its.cern.ch/jira/browse/AGENE-2104
@ewelina, do you know what is the reason why the Pythia8B interface...Add weight name to Pythia8B interface.
Tag @jroggel @ewelina
This should add weight to P8B_A14NNPDF23LO_pp_JpsimumuZmumu in
https://its.cern.ch/jira/browse/AGENE-2104
@ewelina, do you know what is the reason why the Pythia8B interface is so different from Pythia8?https://gitlab.cern.ch/atlas/athena/-/merge_requests/48457EvgenJobTransforms: deactivate QGSJet ART tests2021-11-24T15:44:13+01:00Ewelina Maria LobodzinskaEvgenJobTransforms: deactivate QGSJet ART testsdeactivate QGSJet ART tests, because this generator is removed from releasedeactivate QGSJet ART tests, because this generator is removed from releasehttps://gitlab.cern.ch/atlas/athena/-/merge_requests/48450Make IOVDbMetaDataTool public2021-11-23T15:43:37+01:00Frank WinklmeierMake IOVDbMetaDataTool publicChange all remaining private instances of `IOVDbMetaDataTool` to public to ensure consistency.
Also register incidents before retrieving MetaDataStore. This is somehow
necessary to ensure the correct order.
cc @mnowakChange all remaining private instances of `IOVDbMetaDataTool` to public to ensure consistency.
Also register incidents before retrieving MetaDataStore. This is somehow
necessary to ensure the correct order.
cc @mnowakhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/48443Sherpa_i add SHERPA_SHARE_PATH to avoid config. problems2021-11-23T15:43:19+01:00Ewelina Maria LobodzinskaSherpa_i add SHERPA_SHARE_PATH to avoid config. problemsSherpa_i add SHERPA_SHARE_PATH to avoid config. problems
this MR tries to fix Sherpa crashes in gcc11 ART testsSherpa_i add SHERPA_SHARE_PATH to avoid config. problems
this MR tries to fix Sherpa crashes in gcc11 ART testshttps://gitlab.cern.ch/atlas/athena/-/merge_requests/48417GeneratorFilters add common fragments for xAOD filters and unify names2021-11-23T15:43:08+01:00Ewelina Maria LobodzinskaGeneratorFilters add common fragments for xAOD filters and unify namesGeneratorFilters add common fragments for xAOD filters and unify names
this MR superseds https://gitlab.cern.ch/atlas/athena/-/merge_requests/48341GeneratorFilters add common fragments for xAOD filters and unify names
this MR superseds https://gitlab.cern.ch/atlas/athena/-/merge_requests/48341https://gitlab.cern.ch/atlas/athena/-/merge_requests/48335GeneratorFilters: Truth xAOD Generator Filter MultiElecMuTau2021-11-18T21:43:26+01:00Valentina TudoracheGeneratorFilters: Truth xAOD Generator Filter MultiElecMuTauAdd new filter xAODMultiElecMuTauFilter in Generators/GeneratorFilters, replicating the existing MultiElecMuTauFilter, but running on Truth xAOD instead of HepMC.Add new filter xAODMultiElecMuTauFilter in Generators/GeneratorFilters, replicating the existing MultiElecMuTauFilter, but running on Truth xAOD instead of HepMC.https://gitlab.cern.ch/atlas/athena/-/merge_requests/46581GeneratorObjectsTPCnv: Clear flow information when reading.2021-11-18T10:00:29+01:00Scott SnyderGeneratorObjectsTPCnv: Clear flow information when reading.For versions p2, p3, and p4, we weren't clearing flow information before
filling it. Since these objects are managed by a DataPool, than means
that the flow information is garbage when we read these versions.
Fixed.
See ATLASSIM-5403.For versions p2, p3, and p4, we weren't clearing flow information before
filling it. Since these objects are managed by a DataPool, than means
that the flow information is garbage when we read these versions.
Fixed.
See ATLASSIM-5403.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48250PowhegControl: fix Main31 parameters in some example joboptions (top processe...2021-11-16T15:44:02+01:00Timothee Theveneaux-PelzerPowhegControl: fix Main31 parameters in some example joboptions (top processes) - AGENE-2047This MR specifies the `Powheg:NFinal` main31 parameter for some of the top processes, in the example joboptions. Otherwise it seems Pythia8 crashes in release 22 (not in release 21).
Reminder: the purpose of these example jopboptions is ...This MR specifies the `Powheg:NFinal` main31 parameter for some of the top processes, in the example joboptions. Otherwise it seems Pythia8 crashes in release 22 (not in release 21).
Reminder: the purpose of these example jopboptions is to be used as test, so we should be able to run with them, but other parameters may need to be adjusted for an actual MC sample for physics.
Also I'm deleting 4 ttW example joboptions which were duplicated (they should all be in `share/example/processes` already).https://gitlab.cern.ch/atlas/athena/-/merge_requests/48239Brought MadGraphControl in master up-to-speed with 21.62021-11-16T15:43:34+01:00Hannes MildnerBrought MadGraphControl in master up-to-speed with 21.6Includes commits addressing the following issues
- Remove hacking of the "run.sh" script for LO gridpacks, instead add reweighting in standalone mode
- Fix an issue where MG would use the wrong lhapdf-config
- Fix case sensitivity issue ...Includes commits addressing the following issues
- Remove hacking of the "run.sh" script for LO gridpacks, instead add reweighting in standalone mode
- Fix an issue where MG would use the wrong lhapdf-config
- Fix case sensitivity issue for decay settings
- Completely removed above "run.sh" script and run gridpacks from directly from python instead of using intermediate shell script
- updated py2-specific solutions to py3
- Removed setting of "loop" parameter (corresponding to mu_R), which is hidden in newer MG versions and only affects standalone MadLoop but not event generation, for which mu_R is set from run_cardhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/48251PowhegControl: fix Main31 parameters in some example joboptions (top processe...2021-11-16T10:49:55+01:00Timothee Theveneaux-PelzerPowhegControl: fix Main31 parameters in some example joboptions (top processes) - AGENE-2047This MR specifies the `Powheg:NFinal` main31 parameter for some of the top processes, in the example joboptions. Otherwise it seems Pythia8 crashes in release 22 (not in release 21).
Reminder: the purpose of these example jopboptions is ...This MR specifies the `Powheg:NFinal` main31 parameter for some of the top processes, in the example joboptions. Otherwise it seems Pythia8 crashes in release 22 (not in release 21).
Reminder: the purpose of these example jopboptions is to be used as test, so we should be able to run with them, but other parameters may need to be adjusted for an actual MC sample for physics.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48181Epos_i: make the config. files location dependent on the version2021-11-12T17:51:08+01:00Ewelina Maria LobodzinskaEpos_i: make the config. files location dependent on the versionEpos_i: make the config. files location dependent on the version
The change is needed to fix thecrash in the nightly dev4LCG
See also https://sft.its.cern.ch/jira/browse/SPI-2024 - so urgent label added.Epos_i: make the config. files location dependent on the version
The change is needed to fix thecrash in the nightly dev4LCG
See also https://sft.its.cern.ch/jira/browse/SPI-2024 - so urgent label added.https://gitlab.cern.ch/atlas/athena/-/merge_requests/48155Faster reading from EVT files for MC2021-11-12T15:42:48+01:00Andrii VerbytskyiFaster reading from EVT files for MCFaster reading from EVT files for MC.
The improvement is reached replacing the ``HepMC::barcode_to_vertex`` (emulated in HepMC3) function with a simple map.
A fix for https://its.cern.ch/jira/browse/AGENE-2105
Tag @ewelina @jchapman @...Faster reading from EVT files for MC.
The improvement is reached replacing the ``HepMC::barcode_to_vertex`` (emulated in HepMC3) function with a simple map.
A fix for https://its.cern.ch/jira/browse/AGENE-2105
Tag @ewelina @jchapman @jroggelhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/47728Manual sweep of !39716 from 21.0 to master: Make McEventCollectionCnv_p5_test...2021-11-11T15:42:56+01:00John Derek ChapmanManual sweep of !39716 from 21.0 to master: Make McEventCollectionCnv_p5_test and McEventCollectionCnv_p4_test more robustMerge rewuest makes `McEventCollectionCnv_p5_test` and `McEventCollectionCnv_p4_test` more robust by:
* checking all properties of the `GenEvent`.
* `GenVertex` positions not at `(0,0,0)` to properly test TP conversion of those variable...Merge rewuest makes `McEventCollectionCnv_p5_test` and `McEventCollectionCnv_p4_test` more robust by:
* checking all properties of the `GenEvent`.
* `GenVertex` positions not at `(0,0,0)` to properly test TP conversion of those variables.
* some double-precision properties are only stored with single-precision in the persistent version, so must be cast to `float` before being compared.
The current HepMC2-based versions of `McEventCollectionCnv_p5` and `McEventCollectionCnv_p4` used in the master nightlies flip the order of the `GenParticles` associated with each `GenVertex` after T1->P1->T2 conversion. Running though this twice is sufficient to regain agreement (i.e. T1->P1->T2->P2->T3. T1 == T3 != T2).
The HepMC3-based versions of `McEventCollectionCnv_p5` and `McEventCollectionCnv_p4` do not have this issue (i.e. T1 == T2).
See merge request atlas/athena!39716
(Should be merged after !47934.)