Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix implicit host-to-device copy for ES data products with labels in Alpaka ESProducers #41019

Merged
merged 1 commit into from Mar 13, 2023

Conversation

makortel
Copy link
Contributor

@makortel makortel commented Mar 9, 2023

PR description:

It was noticed during the Patatrack hackathon that the implicit host-to-device copy of EventSetup data products did not work correctly in Alpaka ESProducers when the data product had a label set (either via explicit argument to setWhatProduced() or via the appendToDataLabel configuration parameter). The test code that I added in #40813 turned out to have a flaw, because of which the problem was not caught earlier.

Requiring the concrete ESProducer classes to explicitly call the base ESProducer class' constructor with the ParameterSet argument is a (minor) breaking change, but this was (by far) easiest way to fix the problem.

PR validation:

The unit tests succeeds on machines with and without a GPU.

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 13_0_X (using the same commit).

…Alpaka ESProducers

Taking the "appendToDataLabel" and module label from the module's
ParameterSet was, by far, the easiest way to fix, supported by the
hindsight that having all modules to pass the ParameterSet to the base
class' constructor would have made some features much easier and
cleaner to develop.
@makortel
Copy link
Contributor Author

makortel commented Mar 9, 2023

@fwyzard The branch of this PR is based on abba246 (i.e. #40849 / #40935), which, I think, was the latest commit/PR in our joint branch.

@cmsbuild
Copy link
Contributor

cmsbuild commented Mar 9, 2023

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41019/34541

  • This PR adds an extra 20KB to repository

@cmsbuild
Copy link
Contributor

cmsbuild commented Mar 9, 2023

A new Pull Request was created by @makortel (Matti Kortelainen) for master.

It involves the following packages:

  • HeterogeneousCore/AlpakaCore (heterogeneous)
  • HeterogeneousCore/AlpakaTest (heterogeneous)

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

cms-bot commands are listed here

@makortel
Copy link
Contributor Author

makortel commented Mar 9, 2023

enable gpu

@makortel
Copy link
Contributor Author

makortel commented Mar 9, 2023

@cmsbuild, please test

@cmsbuild
Copy link
Contributor

cmsbuild commented Mar 9, 2023

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-d47084/31177/summary.html
COMMIT: aea87d6
CMSSW: CMSSW_13_1_X_2023-03-09-1100/el8_amd64_gcc11
Additional Tests: GPU
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/41019/31177/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

  • You potentially removed 10 lines from the logs
  • Reco comparison results: 19 differences found in the comparisons
  • DQMHistoTests: Total files compared: 49
  • DQMHistoTests: Total histograms compared: 3530381
  • DQMHistoTests: Total failures: 122
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 3530237
  • DQMHistoTests: Total skipped: 22
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 48 files compared)
  • Checked 213 log files, 164 edm output root files, 49 DQM output files
  • TriggerResults: no differences found

GPU Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 0 differences found in the comparisons
  • DQMHistoTests: Total files compared: 4
  • DQMHistoTests: Total histograms compared: 19862
  • DQMHistoTests: Total failures: 10
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 19852
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 3 files compared)
  • Checked 12 log files, 9 edm output root files, 4 DQM output files
  • TriggerResults: no differences found

@makortel
Copy link
Contributor Author

Hmh, many phase 2 workflows show differences in some JetMET histograms.

@makortel
Copy link
Contributor Author

@cmsbuild, please test

Let's see if they persist

@missirol
Copy link
Contributor

Hmh, many phase 2 workflows show differences in some JetMET histograms.

(Just for the record, I saw something similar yesterday, #41016 (comment).)

@cmsbuild
Copy link
Contributor

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-d47084/31214/summary.html
COMMIT: aea87d6
CMSSW: CMSSW_13_1_X_2023-03-10-1100/el8_amd64_gcc11
Additional Tests: GPU
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/41019/31214/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

  • You potentially added 2 lines to the logs
  • Reco comparison results: 3 differences found in the comparisons
  • DQMHistoTests: Total files compared: 49
  • DQMHistoTests: Total histograms compared: 3530381
  • DQMHistoTests: Total failures: 7
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 3530352
  • DQMHistoTests: Total skipped: 22
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 48 files compared)
  • Checked 213 log files, 164 edm output root files, 49 DQM output files
  • TriggerResults: no differences found

GPU Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 8 differences found in the comparisons
  • DQMHistoTests: Total files compared: 4
  • DQMHistoTests: Total histograms compared: 19862
  • DQMHistoTests: Total failures: 455
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 19407
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 3 files compared)
  • Checked 12 log files, 9 edm output root files, 4 DQM output files
  • TriggerResults: no differences found

@makortel
Copy link
Contributor Author

The JetMET difference didn't repeat, but the latest incarnation of #39803 came visible.

@makortel
Copy link
Contributor Author

@fwyzard Do you have any comments?

@fwyzard
Copy link
Contributor

fwyzard commented Mar 13, 2023

I trust you on this :-)

@makortel
Copy link
Contributor Author

+heterogeneous

@makortel
Copy link
Contributor Author

type bug-fix

@cmsbuild
Copy link
Contributor

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 now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)

@perrotta
Copy link
Contributor

+1

@cmsbuild cmsbuild merged commit fdaea76 into cms-sw:master Mar 13, 2023
13 checks passed
@makortel makortel deleted the fixAlpakaESDataLabel_v2 branch March 17, 2023 07:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

5 participants