Update the compile-time warp size constant for ROCm 7.0 [15.1.x]

[fwyzard]

PR description:

ROCm 7.0 no longer defines the warp size as a compile-time constant.
Work around it using preprocessor macros.

PR validation:

Code compiles with ROCm 7.0.

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:

Backport of #48947 and #49016 to CMSSW 15.1.x.

[cmsbuild]

A new Pull Request was created by @fwyzard for CMSSW_15_1_X.

It involves the following packages:

• HeterogeneousCore/AlpakaInterface (heterogeneous)

@cmsbuild, @fwyzard, @makortel can you please review it and eventually sign? Thanks.
@makortel, @missirol, @rovere this is something you requested to watch as well.
@ftenchini, @mandrenguyen, @sextonkennedy you are the release manager for this.

cms-bot commands are listed here

• Backported from Update the compile-time warp size constant for ROCm 7.0 #48947

[cmsbuild]

cms-bot internal usage

[fwyzard]

please test with cms-sw/cmsdist#10085 for el9_amd64_gcc12

[cmsbuild]

Pull request #48948 was updated. @cmsbuild, @fwyzard, @makortel can you please check and sign again.

[fwyzard]

please test with cms-sw/cmsdist#10085 for el9_amd64_gcc12

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-bf01be/48206/summary.html
COMMIT: 8d51780
CMSSW: CMSSW_15_1_X_2025-09-14-0000/el9_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/48948/48206/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially added 239 lines to the logs
• ROOTFileChecks: Some differences in event products or their sizes found
• Reco comparison results: 0 differences found in the comparisons
• DQMHistoTests: Total files compared: 50
• DQMHistoTests: Total histograms compared: 4113751
• DQMHistoTests: Total failures: 1258243
• DQMHistoTests: Total nulls: 430
• DQMHistoTests: Total successes: 2855058
• DQMHistoTests: Total skipped: 20
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: -6.904 KiB( 49 files compared)
• DQMHistoSizes: changed ( 10224.0 ): 0.773 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 13034.0 ): 1.182 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 141.042,... ): -0.012 KiB JetMET/SUSYDQM
• DQMHistoSizes: changed ( 145.104 ): -0.004 KiB JetMET/SUSYDQM
• DQMHistoSizes: changed ( 145.5 ): -0.023 KiB JetMET/SUSYDQM
• DQMHistoSizes: changed ( 145.604 ): 0.055 KiB JetMET/SUSYDQM
• DQMHistoSizes: changed ( 17034.0 ): -3.607 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 250202.181 ): 0.117 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 25202.0 ): 0.129 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 7.3 ): -5.502 KiB SiStrip/MechanicalView
• Checked 215 log files, 184 edm output root files, 50 DQM output files

[cmsbuild]

Pull request #48948 was updated. @cmsbuild, @fwyzard, @makortel can you please check and sign again.

[fwyzard]

test parameters:

• enable_tests = gpu
• gpu = nvidia_t4,amd_mi300x,amd_w7900

[fwyzard]

please test

[fwyzard]

+heterogeneous

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_15_1_X IBs after it passes the integration tests and once validation in the development release cycle CMSSW_16_0_X is complete. This pull request will now be reviewed by the release team before it's merged. @mandrenguyen, @ftenchini, @sextonkennedy (and backports should be raised in the release meeting by the corresponding L2)

[cmsbuild]

+1

Size: This PR adds an extra 20KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-bf01be/48323/summary.html
COMMIT: 6970d6c
CMSSW: CMSSW_15_1_X_2025-09-28-2300/el8_amd64_gcc12
Additional Tests: GPU,AMD_MI300X,AMD_W7900,NVIDIA_T4
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/48948/48323/install.sh to create a dev area with all the needed externals and cmssw changes.

AMD_W7900 Comparison Summary

Summary:

• You potentially removed 2 lines from the logs
• Reco comparison results: 236 differences found in the comparisons
• DQMHistoTests: Total files compared: 11
• DQMHistoTests: Total histograms compared: 146188
• DQMHistoTests: Total failures: 27729
• DQMHistoTests: Total nulls: 12
• DQMHistoTests: Total successes: 118447
• DQMHistoTests: Total skipped: 0
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 10 files compared)
• Checked 42 log files, 45 edm output root files, 11 DQM output files
• TriggerResults: no differences found

NVIDIA_T4 Comparison Summary

Summary:

• You potentially removed 5 lines from the logs
• Reco comparison results: 236 differences found in the comparisons
• DQMHistoTests: Total files compared: 11
• DQMHistoTests: Total histograms compared: 146188
• DQMHistoTests: Total failures: 27511
• DQMHistoTests: Total nulls: 10
• DQMHistoTests: Total successes: 118667
• DQMHistoTests: Total skipped: 0
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 10 files compared)
• Checked 42 log files, 45 edm output root files, 11 DQM output files
• TriggerResults: no differences found

[fwyzard]

backport #48947

[fwyzard]

@ftenchini please do not put the release branch before the name in the title of the PR.

IMHO a better way to filter which release is targeted by each PR is to use the milestones e.g. for CMSSW_15_1_X.

When one looks at the changes that entered a release, IMHO it's not very helpful that they all begin with the release branch:

If you suggest to standardise on a common syntax, e.g. [15_1_X] vs [15.1.x] vs (15.1.x) or some other variant, I'm happy to follow that.

Or even nothing, and we just use the milestone.

[mandrenguyen]

+1