ATLASRECTS-7818 : Clean up the math in AlignmentErrorTool

This merge request cleans up two floating point computations that were performed in a clumsy manner in the past version.

The cleaned up computations should have the same results as in the previous version, although a small numeric error is likely to appear.

This is a follow-up MR from !67056 (merged) containing only the math clean-up discussed in this comment.

This merge request affects 1 package:

• MuonSpectrometer/MuonAlignment/MuonAlignError/MuonAlignErrorTool

This merge request affects 1 file:

• MuonSpectrometer/MuonAlignment/MuonAlignError/MuonAlignErrorTool/src/AlignmentErrorTool.cxx

Adding @rosati ,@stavrop ,@apsallid ,@jojungge ,@kluit ,@pscholer as watchers

added MuonSpectrometer main review-pending-level-1 labels

added Run2-DataReco-output-changed Run3-DataReco-output-changed labels

CI Result FAILURE (hash 655046de)

	Athena
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 1250]

I have looked at the failing tests and below is what I found.

For RecoRun2Data, I have run:

acmd.py diff-root --nan-equal /cvmfs/atlas-nightlies.cern.ch/repo/data/data-art/WorkflowReferences/main/q442/v33/myESD.pool.root /eos/atlas/atlascerngroupdisk/proj-sit/gitlabci/MR67168_655046de9714aedef259c7bf5c156571f4178a6f/RecoRun2Data/myESD.pool.root

and:

acmd.py diff-root --nan-equal /cvmfs/atlas-nightlies.cern.ch/repo/data/data-art/WorkflowReferences/main/q442/v33/myAOD.pool.root /eos/atlas/atlascerngroupdisk/proj-sit/gitlabci/MR67168_655046de9714aedef259c7bf5c156571f4178a6f/RecoRun2Data/myAOD.pool.root

And indeed I find only small differences that can be explained with the numeric error.

For RecoRun3Data_Checks, I have run:

acmd.py diff-root --nan-equal /cvmfs/atlas-nightlies.cern.ch/repo/data/data-art/WorkflowReferences/main/q449/v42/myESD.pool.root /eos/atlas/atlascerngroupdisk/proj-sit/gitlabci/MR67168_655046de9714aedef259c7bf5c156571f4178a6f/RecoRun3Data_Checks/myESD.pool.root

and:

acmd.py diff-root --nan-equal /cvmfs/atlas-nightlies.cern.ch/repo/data/data-art/WorkflowReferences/main/q449/v42/myAOD.pool.root /eos/atlas/atlascerngroupdisk/proj-sit/gitlabci/MR67168_655046de9714aedef259c7bf5c156571f4178a6f/RecoRun3Data_Checks/myAOD.pool.root

But in this case I don't really understand the differences I see: they are not related to the change I introduced. For example the ESD diff is saying:

Py:diff-root         INFO comparing [3032] leaves over entries...
Py:diff-root         INFO 001.BCM_RDO_Container_p0_BCM_CompactDOs.0.0.m_word1 134247709 -> 32336 => diff= [49.97591898%]
Py:diff-root         INFO 001.BCM_RDO_Container_p0_BCM_CompactDOs.0.0.m_word2 238589867 -> 155341813 => diff= [10.56630607%]
Py:diff-root         INFO ::sync-old 001.BCM_RDO_Container_p0_BCM_CompactDOs.0.m_chanId
Py:diff-root         INFO ::sync-new 001.BCM_RDO_Container_p0_BCM_CompactDOs.0.1.m_word1
Py:diff-root         INFO ::sync-old skipping entry

I am probably misunderstanding what I should be comparing. Anyone has an idea?

As far as I can see, you are comparing the right files. I am surprised about the BCM_ROD differences you mention, I don't find them in the CI output.

Changes look fine from L1 perspective.

Putting this to review-user-action-required while the differences are investigated.

Cheers, Steffen (L1)

added review-user-action-required label and removed review-pending-level-1 label

added RC Attention Required label

added offline-sw-review-approved label

resolved all threads

OK

Lets do the review as std.

Root references and digest references will be updated before merge so the shifter can ignore these .

removed review-user-action-required label

added review-pending-level-1 label

Jenkins please retry a build

Build area was cleaned as per request posted in the DB. The full software build will be performed

This merge request affects 1 package:

• MuonSpectrometer/MuonAlignment/MuonAlignError/MuonAlignErrorTool

This merge request affects 1 file:

• MuonSpectrometer/MuonAlignment/MuonAlignError/MuonAlignErrorTool/src/AlignmentErrorTool.cxx

Adding @kluit ,@apsallid ,@pscholer ,@stavrop ,@jojungge ,@rosati as watchers

CI Result FAILURE (hash 655046de)

	Athena
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 1346]

changed title from Clean up the math in AlignmentErrorTool to ATLASRECTS-7818 : Clean up the math in AlignmentErrorTool

added 97 commits

• 655046de...7a751f72 - 95 commits from branch atlas:main
• 2c15106b - Clean up the math in AlignmentErrorTool.cxx
• 8cd40478 - Update references

Compare with previous version

I updated the references.

This merge request affects 2 packages:

• MuonSpectrometer/MuonAlignment/MuonAlignError/MuonAlignErrorTool
• Tools/WorkflowTestRunner

Affected files list will not be printed in this case

Adding @jojungge ,@rosati ,@pagessin ,@stavrop ,@apsallid ,@kluit ,@pscholer as watchers

added Tools label

CI Result SUCCESS (hash 8cd40478)

	Athena	AthSimulation	AthGeneration
externals
cmake
make
tests

Full details available on this CI monitor view. Check the JIRA CI status board for known problems
Athena: number of compilation errors 0, warnings 0
AthSimulation: number of compilation errors 0, warnings 0
AthGeneration: number of compilation errors 0, warnings 0
For experts only: Jenkins output [CI-MERGE-REQUEST-EL9 1359]

All threads resolved and passes CI. Approving for L1.

added review-approved label and removed Run2-DataReco-output-changed Run3-DataReco-output-changed review-pending-level-1 labels

added Run2-DataReco-output-changed Run3-DataReco-output-changed labels

@llewitt, make sure you do not remove the labels by accident.

mentioned in commit b1dc010f

merged