athena merge requestshttps://gitlab.cern.ch/atlas/athena/-/merge_requests2019-01-09T17:56:59+01:00https://gitlab.cern.ch/atlas/athena/-/merge_requests/20010ATR-19246 TriggerMT Decision Encoding in ESD Update2019-01-09T17:56:59+01:00Tim MartinATR-19246 TriggerMT Decision Encoding in ESD UpdateThis persists in the trigger output a `HLTSummary`, and then in the ESD a `xAOD::TrigDecision` object the bit-set which corresponds to fired triggers in the new MT prototype. cc @tbold @bwynne
It hacks a short-term hard-coded fudge i...This persists in the trigger output a `HLTSummary`, and then in the ESD a `xAOD::TrigDecision` object the bit-set which corresponds to fired triggers in the new MT prototype. cc @tbold @bwynne
It hacks a short-term hard-coded fudge into the trigger configuration service in order to obtain `ChainCounters`. cc @mark @fwinkl @stelzer
* Remove my recently added extension to the TrigDecision EDM. Disk-space wise, this was not going to work, sadly.
* Update downstream clients of `DecisionSummaryMakerAlg` within the trigger (signatureMoni, streaming, bit-encoding in HLTResult)
* Update `TrigDecisionMakerMT` to read all HLT summaries from one collection, `DecisionSummaryMakerAlg`'s `HLTSummary`
* Update transient HLT seeding information written by `L1Decoder`, call this `L1DecoderSummary` everywhere.
* Copy TrigCost end-of-event trigger fire to `DecisionSummaryMakerAlg`
* Remove the output from the `TriggerSummaryAlg` - this now just prints to terminal
* Hack the TrigConfSvc to allow hard-coded chains to be injected in the short term, for testing.
(note known problem with egamma test, will try and fix that now too - but in its own MR)https://gitlab.cern.ch/atlas/athena/-/merge_requests/20096Muon efficiency correction feature updates2020-04-11T18:10:15+02:00Johannes Junggeburthjohannes.josef.junggeburth@cern.chMuon efficiency correction feature updatesHi everybody,
this MR is going to prepare the next round of recommendations for the reconstruction, TTVA and isolation scale-factors. The tool will be augmented with the following features
```
*** Break down of the systematics pe...Hi everybody,
this MR is going to prepare the next round of recommendations for the reconstruction, TTVA and isolation scale-factors. The tool will be augmented with the following features
```
*** Break down of the systematics per source (I.e. truth-closure/ fit-template/ isolation WP/ etc.
*** Decorrelation of the systematics per bin as it's already done for the statistical error if it's possible
*** Cleaning up depreciated classes
*** Minimize the number histograms to be loaded in the memory.
```
Cheers,
Johannes
Adding: @goblirsc, @fsforza, @szambito and our well admired friend the @atlasbot
![image](/uploads/91c09a1e79fa15433e6e06d0cde9bbf2/image.png)https://gitlab.cern.ch/atlas/athena/-/merge_requests/20146Update packages:TrigEgammaEmulationTool TrigEgammaMatchingTool…2019-01-10T15:23:06+01:00Fernando MonticelliUpdate packages:TrigEgammaEmulationTool TrigEgammaMatchingTool…Update packages:TrigEgammaEmulationTool TrigEgammaMatchingTool TrigEgammaMonitoring from upstream/21.0 to upstream/master via pseudo-merge
Following in ATR-19005.
The result of the test_physics_pp_v7_primaries_build.sh are availa...Update packages:TrigEgammaEmulationTool TrigEgammaMatchingTool TrigEgammaMonitoring from upstream/21.0 to upstream/master via pseudo-merge
Following in ATR-19005.
The result of the test_physics_pp_v7_primaries_build.sh are available in /afs/cern.ch/user/f/fernando/public/ATR-19005/pseudomerge_fernando_31_Dec_TrigEgammaEmulationTool_to_master/.
There are arrors unrelated to these packages.
TrigEgammaAnalysisTools pseudomerge fialed. So not included in this merge
Most of the conflicts resolved in favor of 21.0-TrigMC. Except those related to changes in interfaces or in messaging, that were updated in master and not in 21.0-TrigMC. Many conflicts resolved on many files. If doubts, please opend discussion or follow up in ATR-19005.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20168Move OutputStream to output Sequence to help AthenaMT scheduling.2019-02-22T10:35:41+01:00Peter Van GemmerenMove OutputStream to output Sequence to help AthenaMT scheduling.This MR moves OutputStreams (that have the asAlg flag set, most of them) into the Athena Output Sequence to help with MT scheduling (@leggett , @tsulaia ).
It also adds a mutex for writing that is currently needed if more than one strea...This MR moves OutputStreams (that have the asAlg flag set, most of them) into the Athena Output Sequence to help with MT scheduling (@leggett , @tsulaia ).
It also adds a mutex for writing that is currently needed if more than one stream is used (because writing modifies the SG proxy of the objects in the ItemList, conversion service caches file name for object placement, to be fixed later).
Some tests for Registration Stream may need updating or they could be obsolete anyway (@cranshaw ).https://gitlab.cern.ch/atlas/athena/-/merge_requests/20261Simplification of HLT CF naming conventions to allow further cleanup and redu...2019-01-17T16:51:09+01:00Francesca PastoreSimplification of HLT CF naming conventions to allow further cleanup and reductionsI have collected all the naming conventions for the CF objects in one python file, so one can track the strategy to set the CF components names. I also included the changes proposed by Tomasz in !20089. Tagging @tbold: just to say that I...I have collected all the naming conventions for the CF objects in one python file, so one can track the strategy to set the CF components names. I also included the changes proposed by Tomasz in !20089. Tagging @tbold: just to say that I dint’t touch the NewJO part, since I don’t know what part is used and what is just a test. All tests passed. I made a detailed diff of the logs in the emulation test, to check that nothing changed in the flow. This is replacing !20089, so that that is obsolete and can be closed (if Tomasz agrees)https://gitlab.cern.ch/atlas/athena/-/merge_requests/20263Support GSC for FTK b-jet chains2019-01-17T20:47:35+01:00Carlo Varnicarlo.varni@cern.chSupport GSC for FTK b-jet chainsFollowing ATR-19193, some code changes in the configuration of FTK b-jet chains are needed in order to support GSC.
This has been tested in 21.3 by adding some FTK+GSC b-jet chains in the Physics v7 menu and with the following command:
...Following ATR-19193, some code changes in the configuration of FTK b-jet chains are needed in order to support GSC.
This has been tested in 21.3 by adding some FTK+GSC b-jet chains in the Physics v7 menu and with the following command:
`trigtest.pl --cleardir --test AthenaTrigRDO_MC_pp_v7_no_prescale_toxAOD_FTK --rundir AthenaTrigRDO_MC_pp_v7_no_prescale_toxAOD_FTK --conf TriggerTest.conf`
Additional code changes have to be applied for addressing ATR-19262, but I'll open another MR for that since it is a different issue.
Tagging @jmontejo, @aranzazu and @baineshttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20294ATR-19303: Example to configure simple example electron chain2019-01-30T03:02:17+01:00Catrin BerniusATR-19303: Example to configure simple example electron chainElectronDef.py and generateElectronChainDefs.py are now in a state such that a simple electron chain can be configured.
Tagging @tbold and @fpastore
To be addressed in a different MR:
* MenuChains.py: remove the hardcoded dict for ...ElectronDef.py and generateElectronChainDefs.py are now in a state such that a simple electron chain can be configured.
Tagging @tbold and @fpastore
To be addressed in a different MR:
* MenuChains.py: remove the hardcoded dict for hypo tool configuration
* Noticed that chains have to be named with HLT_ prefix due to Trigger/TrigHypothesis/TrigEgammaHypo/python/TrigL2ElectronHypoTool.py (TrigL2ElectronHypoToolFromName), once this is changed, can work with just chain name without HLT_ prefix
Note:
* ElectronDef.py might need further tidying up (in terms of removing getStep, getFastCaloStep or getElectronStep) but since in dev, will keep it for now as the code might become more complex as well, so these functions might be needed.
Might need to consider designing a base class with some functions that can be shared across signatures.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20306New style python configuration for RngComps and BCM_Digitization2019-03-15T11:02:11+01:00Rupert TombsNew style python configuration for RngComps and BCM_DigitizationAdd ComponentAccumulator styled services for RngComps core pseudo-random number generation services.
Modify UnifyProperties.py to allow combination of seeds.
Add ConfigFlags styled configuration in BCM_DigitizationConfigFlags.py.
Add ...Add ComponentAccumulator styled services for RngComps core pseudo-random number generation services.
Modify UnifyProperties.py to allow combination of seeds.
Add ConfigFlags styled configuration in BCM_DigitizationConfigFlags.py.
Add ComponentAccumulator styled tools and algorithms for BCM_Digitization in BCM_DigitizationConfigNew.py.
-New appended to avoid collision with old style code.
Add tests for RandomServices.py and BCM_DigitizationConfigNew.py (which depends on BCM_DigitizationConfigFlags.py).https://gitlab.cern.ch/atlas/athena/-/merge_requests/20367Adding precision tracking to bjets ATR-192632019-04-09T12:37:54+02:00Matous VozakAdding precision tracking to bjets ATR-19263Adding ambiguity solver algorithm with xAOD TrackParticle conversion alg to bjet menu. The ticket for the migration process can be found in ATR-19263 @jmasikAdding ambiguity solver algorithm with xAOD TrackParticle conversion alg to bjet menu. The ticket for the migration process can be found in ATR-19263 @jmasikhttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20378Ringer AthenaMT migration (FEX and HYPO)2019-03-06T16:31:51+01:00Joao Victor Da Fonseca PintoRinger AthenaMT migration (FEX and HYPO)The Athena MT ringer migration will affect:
TrigT2CaloCommon: Include doRinger flag in python configuration;
TrigT2CaloEgamma: Code harmonization, new Athena monitoring, New Ringer Fex follows the recomendation from AthenaMT note (co...The Athena MT ringer migration will affect:
TrigT2CaloCommon: Include doRinger flag in python configuration;
TrigT2CaloEgamma: Code harmonization, new Athena monitoring, New Ringer Fex follows the recomendation from AthenaMT note (const methods), support to bphysics chains for no hadronic chains (nohad) for future developments;
TrigMultiVarHypo: Code update, harmonization and migration to Athena MT follow the recommendations from AthenaMT note; (solver the issue in ATR-19281);
TrigUpgradeTest: Add ringer chain test (fex+hypo);
Tested using: Asetup master,r4,setup,here And Source test_ringer_chain.sh (see test file in TrigUpgradeTest)
Tagging: @fernando and @thrynova (ringer MT task) and @wsfreundJoao Victor Da Fonseca PintoJoao Victor Da Fonseca Pintohttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20394Improvement of pixel conditions (ATLASRECTS-4662)2019-02-27T03:02:29+01:00Soshi TsunoImprovement of pixel conditions (ATLASRECTS-4662)In current code, the conditions data is interfaced by a tool, then, if the other tool needs the conditions data, that tool calls the "interface tool". That is, a tool calls a tool to access conditions data. This makes complicated structu...In current code, the conditions data is interfaced by a tool, then, if the other tool needs the conditions data, that tool calls the "interface tool". That is, a tool calls a tool to access conditions data. This makes complicated structure of the code.
So, the suggestion raised in ATLASRECTS-4662 was to drop the interface tool, but directly handle the conditions data.
Based on this suggestion, we deleted the PixelDCSConditionsTool which was called from PixelLorentzAngleTool, PixelPropertiesTool, PixelConditionsSummaryTool. Now, they access the DCS data through the conditions algorithm directly.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20448Fixing some problems with multuple stream handling in AthenaMT.2019-01-26T03:02:21+01:00Peter Van GemmerenFixing some problems with multuple stream handling in AthenaMT.This MR addresses some short comings of I/O when running AthenaMT and mainly writing to multiple streams:
1) Multiple stream writing has to be serial (for now) as it modifies the TransientAddress of output objects. Therefore a mutex is ...This MR addresses some short comings of I/O when running AthenaMT and mainly writing to multiple streams:
1) Multiple stream writing has to be serial (for now) as it modifies the TransientAddress of output objects. Therefore a mutex is introduced into the ItemList interface to allow the AthenaOutputStream serialization of streaming.
2) A warning for missing AttributeList is suppressed unless it in the EventStore
3) Use of ROOT Implicit Multithreading can be disabled via PoolSvc
https://gitlab.cern.ch/atlas/athena/-/merge_requests/20461rework of the EventLoop Worker implementation2019-02-06T20:33:52+01:00Nils Erik Krumnackrework of the EventLoop Worker implementationThis ended up as a rather major rewrite of the `Worker` class. Essentially I did three things though:
* I moved some functionality that was common between the different child classes into the `Worker` class itself. Back when we still h...This ended up as a rather major rewrite of the `Worker` class. Essentially I did three things though:
* I moved some functionality that was common between the different child classes into the `Worker` class itself. Back when we still had to support PROOF which worked completely differently this functionality needed to be implemented separately for each worker, but there was no longer any reason for such a setup.
* I moved all the diagnostic, instrumentation and helper code out of the `Worker` class and into separate classes. For that I implemented a new `Module` mechanism that allows to run code at various places inside the worker job. Not only does that simplify and clarify the `Worker` implementation, it also makes it a lot easier to understand what each of these modules do.
* I removed the state machine from the `Worker` class, which was actually quite hard to maintain and understand at this point. Part of the state machine is moved to a new module `AlgorithmStateModule`, but most of the state machine logic I could just remove straight out. This is another simplification due to no longer supporting PROOF.
* I also dd a general update to follow more of the standard ATLAS coding conventions with status codes, message streams/macros, smart pointers, etc.
Let's already alert @akraszna to this. The shifters should take a look at it first, but given the scope of these changes I presume that at least some parts will be flagged for expert review.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20467Optimization of CPU time of FastCaloSimV2 (ATLASSIM-3980)2019-04-04T09:56:29+02:00Xiaozhong HuangOptimization of CPU time of FastCaloSimV2 (ATLASSIM-3980)The CaloCellContainer is set to be incomplete by default. Thus, a look up table is needed to to find the correct
index based on the hash value of the cell element. In fact, the CaloCellContainer is always ordered and complete
after cre...The CaloCellContainer is set to be incomplete by default. Thus, a look up table is needed to to find the correct
index based on the hash value of the cell element. In fact, the CaloCellContainer is always ordered and complete
after creation the look up table is not needed.
After modification, the CPU time/evt is reduced from 71ms to 53ms for single photons.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20696Using CachedValue to fix the Vertex_v1 for MT2019-01-30T03:02:02+01:00Christos Anastopouloschristos.anastopoulos@cern.chUsing CachedValue to fix the Vertex_v1 for MTUsing CachedValue for the relevant parts of ``Vertex_v1``
Mentioning @ssnyder , @emoyse , @goetz , @mhodgkin for their info.
This should now leave the ``definingParameters`` for the NeutralParticle to be handled for the package to...Using CachedValue for the relevant parts of ``Vertex_v1``
Mentioning @ssnyder , @emoyse , @goetz , @mhodgkin for their info.
This should now leave the ``definingParameters`` for the NeutralParticle to be handled for the package to be fully MT
compatible.
Since I am here removed certain spurious ^M from the header ...https://gitlab.cern.ch/atlas/athena/-/merge_requests/20705Steps for merging chains with common sequences in the HLT control flow2019-02-14T10:23:45+01:00Francesca PastoreSteps for merging chains with common sequences in the HLT control flowThese changes cover:
- factorisation of sequences in the emulation test
- better organization and factorisation of the fast calo sequence
- factorisation of electron menu code
- added test for running the TriggerMenuMT generation wit...These changes cover:
- factorisation of sequences in the emulation test
- better organization and factorisation of the fast calo sequence
- factorisation of electron menu code
- added test for running the TriggerMenuMT generation with gamma chains
- some cleaning of the CF generation (touching all signature tests)
- bug fix for running electron and photons in fullMenu
- bug fix for MET L1 RoIs
- added MET chains in fullMenu
Tagging: @tbold, @sgeorge, @tamartin, @bernius , @damazio, @dbakshighttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20753Add wasMerged in ComponentAccumulator2019-02-02T03:01:54+01:00Tomasz BoldAdd wasMerged in ComponentAccumulatorThis MR contains few improvements:
1) adds wasMerged functionality - we need to use temporary CAs in HLT to do proper merging of hypo tools
1.5) improves one deduplication failure message
2) adds option to enable verbose merging of pr...This MR contains few improvements:
1) adds wasMerged functionality - we need to use temporary CAs in HLT to do proper merging of hypo tools
1.5) improves one deduplication failure message
2) adds option to enable verbose merging of properites
3) cleanes trailing whitespaces
Tagging: @wlampl @kjamroghttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20792Move DataQualityTools to ReadHandles2019-02-05T03:02:21+01:00Peter OnyisiMove DataQualityTools to ReadHandleshttps://gitlab.cern.ch/atlas/athena/-/merge_requests/20876Migrate LArDigitization to use IAtRNGSvc for AthenaMT2019-02-23T03:02:01+01:00Hass AbouzeidMigrate LArDigitization to use IAtRNGSvc for AthenaMTChanging the method in which random numbers are retrieved for the LAr
Digitization code in order for reproducible output when running in
AthenaMT.
cc @jchapman
Marking as WIP until folks are reasonably happy.Changing the method in which random numbers are retrieved for the LAr
Digitization code in order for reproducible output when running in
AthenaMT.
cc @jchapman
Marking as WIP until folks are reasonably happy.https://gitlab.cern.ch/atlas/athena/-/merge_requests/20971Removing unsupported muon isolation WP and adding new supported ones2019-02-12T03:02:55+01:00Tomas DadoRemoving unsupported muon isolation WP and adding new supported onesRemoving unsupported muon isolation WP and adding new supported ones. Updating the validation configuration files as the previously used muon isolation WPs are no longer supported. Closes ANALYSISTO-757.
This MR is important as the ch...Removing unsupported muon isolation WP and adding new supported ones. Updating the validation configuration files as the previously used muon isolation WPs are no longer supported. Closes ANALYSISTO-757.
This MR is important as the changes are needed for successful tests of !20869, the tests for that MR are failing because in the CI tests in AT we ran with a configuration that is no longer supported, this is fixed in this MR.
Furthermore, removing RC variables from the tests as these fail with the most recent ptags.
Tagging @spalazzo,@mvanadia and @pjacka