Improve the Phase 1 pixel raw to cluster (part 2)

[fwyzard]

PR description:

Implement a few small optimisations:

• use only 64 groups, instead of one per module;
• restrict the domain of atomic operations to per-block.

Together with #48580 (minus the changes reverted in #48609) these changes give 7% speed up to the pixel local reconstruction:

machine type	CMSSW_15_0_10_patch2	with #48581	relative difference
2022 HLT node (2× Milan 7763, 2× T4)	2495 ± 3 ev/s	2670 ± 6 ev/s	+7%
2024 HLT node (2× Bergamo 9754, 3× L4)	11598 ± 23 ev/s	12447 ± 25 ev/s	+7%

The performance of the full HLT menu is not affected, as it is limited by the CPU part.

PR validation:

None.

No changes expected.

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

To be backported to 15.0.x for the 2025 data taking.

[cmsbuild]

cms-bot internal usage

[fwyzard]

enable gpu

[fwyzard]

please test

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48644/45632

• There are other open Pull requests which might conflict with changes you have proposed:
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/PixelClustering.h modified in PR(s): Pixelimage morphing #48343, To correct L1T Zero Suppression we move ZDC to position 54 and moves ZS up to 52 and 53 #48633
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/SiPixelRawToClusterKernel.dev.cc modified in PR(s): Pixelimage morphing #48343, Updated SoA View accessors from raw pointers to span #48377, To correct L1T Zero Suppression we move ZDC to position 54 and moves ZS up to 52 and 53 #48633

[cmsbuild]

A new Pull Request was created by @fwyzard for master.

It involves the following packages:

• RecoLocalTracker/SiPixelClusterizer (reconstruction)

@jfernan2, @mandrenguyen can you please review it and eventually sign? Thanks.
@GiacomoSguazzoni, @VinInn, @VourMa, @dkotlins, @felicepantaleo, @ferencek, @gpetruc, @missirol, @mmusich, @mroguljic, @mtosi, @rovere, @threus, @tsusa, @tvami this is something you requested to watch as well.
@antoniovilela, @mandrenguyen, @rappoccio, @sextonkennedy you are the release manager for this.

cms-bot commands are listed here

[fwyzard]

please test

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-48644/45633

• There are other open Pull requests which might conflict with changes you have proposed:
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/PixelClustering.h modified in PR(s): Pixelimage morphing #48343, To correct L1T Zero Suppression we move ZDC to position 54 and moves ZS up to 52 and 53 #48633
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/SiPixelRawToClusterKernel.dev.cc modified in PR(s): Pixelimage morphing #48343, Updated SoA View accessors from raw pointers to span #48377, To correct L1T Zero Suppression we move ZDC to position 54 and moves ZS up to 52 and 53 #48633

[cmsbuild]

Pull request #48644 was updated. @jfernan2, @mandrenguyen can you please check and sign again.

[cmsbuild]

+1

Size: This PR adds an extra 32KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-69fd75/47434/summary.html
COMMIT: 1d744f7
CMSSW: CMSSW_15_1_X_2025-07-28-2300/el8_amd64_gcc12
Additional Tests: CUDA,ROCM
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/48644/47434/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially added 5 lines to the logs
• Reco comparison results: 4 differences found in the comparisons
• DQMHistoTests: Total files compared: 50
• DQMHistoTests: Total histograms compared: 4076162
• DQMHistoTests: Total failures: 28
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 4076114
• DQMHistoTests: Total skipped: 20
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 49 files compared)
• Checked 215 log files, 184 edm output root files, 50 DQM output files
• TriggerResults: no differences found

CUDA Comparison Summary

Summary:

• You potentially added 1 lines to the logs
• Reco comparison results: 61 differences found in the comparisons
• DQMHistoTests: Total files compared: 7
• DQMHistoTests: Total histograms compared: 53445
• DQMHistoTests: Total failures: 7359
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 46086
• DQMHistoTests: Total skipped: 0
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
• Checked 24 log files, 30 edm output root files, 7 DQM output files
• TriggerResults: no differences found

ROCM Comparison Summary

Summary:

• You potentially added 3 lines to the logs
• Reco comparison results: 64 differences found in the comparisons
• DQMHistoTests: Total files compared: 7
• DQMHistoTests: Total histograms compared: 53445
• DQMHistoTests: Total failures: 8852
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 44593
• DQMHistoTests: Total skipped: 0
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
• Checked 24 log files, 30 edm output root files, 7 DQM output files
• TriggerResults: no differences found

[mandrenguyen]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will be automatically merged.