[14.1.X] Generalize warp size in pixel clustering kernel

[AuroraPerego]

PR description:

In the FindClus kernel there was an assumption on the warp size being equal to 32, which is true for NVIDIA GPUs, but not always for AMD GPUs. Now the warp size is taken from the __AMDGCN_WAVEFRONT_SIZE macro for HIP and uses 32 for CUDA.
Note that alpaka::warp::getSize(acc) cannot be used because it's evaluated at runtime.

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 #46426

FYI @fwyzard

[cmsbuild]

A new Pull Request was created by @AuroraPerego for CMSSW_14_1_X.

It involves the following packages:

• RecoLocalTracker/SiPixelClusterizer (reconstruction)

@cmsbuild, @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

• Backported from Generalize warp size in pixel clustering kernel #46426

[cmsbuild]

cms-bot internal usage

[jfernan2]

please test

[cmsbuild]

+1

Size: This PR adds an extra 12KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-70b131/42307/summary.html
COMMIT: cf3b297
CMSSW: CMSSW_14_1_X_2024-10-19-1100/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/46444/42307/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially removed 2 lines from the logs
• Reco comparison results: 8 differences found in the comparisons
• DQMHistoTests: Total files compared: 44
• DQMHistoTests: Total histograms compared: 3331633
• DQMHistoTests: Total failures: 6
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3331607
• DQMHistoTests: Total skipped: 20
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 43 files compared)
• Checked 191 log files, 161 edm output root files, 44 DQM output files
• TriggerResults: no differences found

[jfernan2]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_14_1_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_14_2_X is complete. This pull request will now be reviewed by the release team before it's merged. @rappoccio, @sextonkennedy, @mandrenguyen, @antoniovilela (and backports should be raised in the release meeting by the corresponding L2)

[mandrenguyen]

backport of #46426

[mandrenguyen]

what's the motivation to backport this PR to 14_1_X?

[fwyzard]

We need it in 14.0.x to test and measure the performance of the HLT on AMD GPUs.

Usually when making a backport for a given release (e.g. 14.0), we were asked to make it also for the more recent release (so 14.1).

It might also be useful if we later decide to test the Heavy Ions HLT on AMD GPUs.

[mandrenguyen]

+1