Skip to content

Update CUDA to version 13.4.1 - #10844

Open
fwyzard wants to merge 1 commit into
cms-sw:IB/CMSSW_20_1_X/masterfrom
fwyzard:IB/CMSSW_20_1_X/master_CUDA_13.4.1
Open

Update CUDA to version 13.4.1#10844
fwyzard wants to merge 1 commit into
cms-sw:IB/CMSSW_20_1_X/masterfrom
fwyzard:IB/CMSSW_20_1_X/master_CUDA_13.4.1

Conversation

@fwyzard

@fwyzard fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor

Update CUDA to version 13.4.1 (general availability).

Major changes since CUDA 13.3:

  • add support for GCC 16 and clang 22 as host compilers;
  • the NVIDIA drivers are now packaged separately.

Patch the cuda-gdb shell script rather than adding a second wrapper.

The NVIDIA drivers are now packaged separately.

Patch the cuda-gdb shell script rather than adding a second wrapper.
@cmsbuild

Copy link
Copy Markdown
Contributor

A new Pull Request was created by @fwyzard for branch IB/CMSSW_20_1_X/master.

@akritkbehera, @cmsbuild, @iarspider, @raoatifshad, @smuzaffar can you please review it and eventually sign? Thanks.
@ftenchini, @mandrenguyen, @sextonkennedy you are the release manager for this.
cms-bot commands are listed here

@cmsbuild

cmsbuild commented Sep 10, 2026

Copy link
Copy Markdown
Contributor

cms-bot internal usage

@fwyzard

fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor Author

@smuzaffar @akritkbehera, if this works for GCC 14 in C++20 mode, how do I test it with GCC 15 and GCC 16 in C++23 mode ?

@fwyzard

fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor Author

enable gpu

@fwyzard

fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor Author

please test

@fwyzard

fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor Author

please test for el9_amd64_gcc16 with cms-sw/cms-bot#2838

@fwyzard

fwyzard commented Sep 10, 2026

Copy link
Copy Markdown
Contributor Author

please test for el9_amd64_gcc15 with cms-sw/cms-bot#2838

@cmsbuild

Copy link
Copy Markdown
Contributor

-1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-e693b7/55790/summary.html
COMMIT: 2077d12
CMSSW: CMSSW_20_1_X_2026-09-08-2300/el9_amd64_gcc16
Additional Tests: GPU,AMD_MI300X,AMD_W7900,NVIDIA_H100,NVIDIA_L40S,NVIDIA_T4
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmsdist/10844/55790/install.sh to create a dev area with all the needed externals and cmssw changes.

Failed External Build

I found compilation error when building:

cwd: /data/cmsbld/jenkins/workspace/ib-run-pr-tests/testBuildDir/BUILD/el9_amd64_gcc16/external/py3-torch-cuda/2.13.0-c725d6eced9d91b59d33ddc3ba0b3142/external_py3-torch-cuda_2.13.0-c725d6eced9d91b59d33ddc3ba0b3142-1-build/cmsdist-pip-src/torch-2.13.0
Building wheel for torch (pyproject.toml): finished with status 'error'
ERROR: Failed building wheel for torch
Failed to build torch
ERROR: Failed to build one or more wheels
error: Bad exit status from /data/cmsbld/jenkins/workspace/ib-run-pr-tests/testBuildDir/tmp/rpm-tmp.xJbBlY (%build)

RPM build warnings:
Macro expanded in comment on line 674: %{pkginstroot}/bin/*

Macro expanded in comment on line 679: %{pkginstroot}/${PYTHON3_LIB_SITE_PACKAGES}


@fwyzard

fwyzard commented Sep 11, 2026

Copy link
Copy Markdown
Contributor Author

The failure with GCC 16 is a general CUDA issue, already reported here: https://forums.developer.nvidia.com/t/nvcc-13-4-rejects-conforming-libstdc-string-with-g-16/382852 .

@cmsbuild

Copy link
Copy Markdown
Contributor

-1

Failed Tests: UnitTests nvidia_l40sUnitTests
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-e693b7/55789/summary.html
COMMIT: 2077d12
CMSSW: CMSSW_20_1_X_2026-09-10-1100/el9_amd64_gcc14
Additional Tests: GPU,AMD_MI300X,AMD_W7900,NVIDIA_H100,NVIDIA_L40S,NVIDIA_T4
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmsdist/10844/55789/install.sh to create a dev area with all the needed externals and cmssw changes.

Failed Unit Tests

I found 10 errors in the following unit tests:

---> test unitTestsGroup_1_u2 had ERRORS
---> test unitTestsGroup_4_u20 had ERRORS
---> test TestFWCoreIntegrationStandalone had ERRORS
and more ...

Comparison Summary

Summary:

  • You potentially added 8 lines to the logs
  • Reco comparison results: 0 differences found in the comparisons
  • DQMHistoTests: Total files compared: 45
  • DQMHistoTests: Total histograms compared: 3592269
  • DQMHistoTests: Total failures: 105
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 3592146
  • DQMHistoTests: Total skipped: 18
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 44 files compared)
  • Checked 195 log files, 163 edm output root files, 45 DQM output files
  • TriggerResults: no differences found

AMD_MI300X Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 62 differences found in the comparisons
  • DQMHistoTests: Total files compared: 7
  • DQMHistoTests: Total histograms compared: 173739
  • DQMHistoTests: Total failures: 8057
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 165682
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
  • Checked 25 log files, 20 edm output root files, 7 DQM output files
  • TriggerResults: no differences found

AMD_W7900 Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 49 differences found in the comparisons
  • DQMHistoTests: Total files compared: 7
  • DQMHistoTests: Total histograms compared: 173739
  • DQMHistoTests: Total failures: 7175
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 166564
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
  • Checked 25 log files, 20 edm output root files, 7 DQM output files
  • TriggerResults: no differences found

NVIDIA_H100 Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 66 differences found in the comparisons
  • DQMHistoTests: Total files compared: 7
  • DQMHistoTests: Total histograms compared: 173739
  • DQMHistoTests: Total failures: 8849
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 164890
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
  • Checked 25 log files, 20 edm output root files, 7 DQM output files
  • TriggerResults: no differences found

NVIDIA_L40S Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 95 differences found in the comparisons
  • DQMHistoTests: Total files compared: 7
  • DQMHistoTests: Total histograms compared: 173739
  • DQMHistoTests: Total failures: 9167
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 164572
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
  • Checked 25 log files, 20 edm output root files, 7 DQM output files
  • TriggerResults: no differences found

NVIDIA_T4 Comparison Summary

Summary:

  • No significant changes to the logs found
  • Reco comparison results: 85 differences found in the comparisons
  • DQMHistoTests: Total files compared: 7
  • DQMHistoTests: Total histograms compared: 173739
  • DQMHistoTests: Total failures: 9030
  • DQMHistoTests: Total nulls: 0
  • DQMHistoTests: Total successes: 164709
  • DQMHistoTests: Total skipped: 0
  • DQMHistoTests: Total Missing objects: 0
  • DQMHistoSizes: Histogram memory added: 0.0 KiB( 6 files compared)
  • Checked 25 log files, 20 edm output root files, 7 DQM output files
  • TriggerResults: no differences found

@cmsbuild

Copy link
Copy Markdown
Contributor

-1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-e693b7/55801/summary.html
COMMIT: 2077d12
CMSSW: CMSSW_20_1_X_2026-09-08-2300/el9_amd64_gcc15
Additional Tests: GPU,AMD_MI300X,AMD_W7900,NVIDIA_H100,NVIDIA_L40S,NVIDIA_T4
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmsdist/10844/55801/install.sh to create a dev area with all the needed externals and cmssw changes.

Failed External Build

I found compilation error when building:

cwd: /data/cmsbld/jenkins/workspace/ib-run-pr-tests/testBuildDir/BUILD/el9_amd64_gcc15/external/py3-torch-cuda/2.13.0-69c963f7f69016829584583110c7f92b/external_py3-torch-cuda_2.13.0-69c963f7f69016829584583110c7f92b-1-build/cmsdist-pip-src/torch-2.13.0
Building wheel for torch (pyproject.toml): finished with status 'error'
ERROR: Failed building wheel for torch
Failed to build torch
ERROR: Failed to build one or more wheels
error: Bad exit status from /data/cmsbld/jenkins/workspace/ib-run-pr-tests/testBuildDir/tmp/rpm-tmp.JwEzSB (%build)

RPM build warnings:
Macro expanded in comment on line 674: %{pkginstroot}/bin/*

Macro expanded in comment on line 679: %{pkginstroot}/${PYTHON3_LIB_SITE_PACKAGES}


@smuzaffar

smuzaffar commented Sep 11, 2026

Copy link
Copy Markdown
Contributor

torch still fails with Error: Internal Compiler Error (codegen): "unsupported float variant!" for gcc15 with c++23

@fwyzard

fwyzard commented Sep 11, 2026

Copy link
Copy Markdown
Contributor Author

torch still fails with Error: Internal Compiler Error (codegen): "unsupported float variant!" for gcc15 with c++23

Is that because of gcc15 or because of c++23 ?

@smuzaffar

Copy link
Copy Markdown
Contributor

I think this is due to c++23, last time we tested cuda update with CPP23/GCC14 and we have seen the same ICE for that.

@fwyzard

fwyzard commented Sep 11, 2026

Copy link
Copy Markdown
Contributor Author

Would it be useful to have CUDA with gcc15 or gcc16 in c++20 mode ?

@fwyzard

fwyzard commented Sep 11, 2026

Copy link
Copy Markdown
Contributor Author

What I mean is that, given that CUDA still does not really support c++23, we could at least update the compiler to gcc 15 or gcc 16 and remain in c++20 mode 🤷🏻‍♂️

@smuzaffar

smuzaffar commented Sep 11, 2026

Copy link
Copy Markdown
Contributor

I have no objections moving default gcc15/gcc16 to c++20. May be we can add extra CPP23 IBs for gcc15 and gcc16 just to make sure we build the sw stack with c++23 and newer compilers

@fwyzard

fwyzard commented Sep 11, 2026

Copy link
Copy Markdown
Contributor Author

I have no objections moving default gcc15/gcc16 to c++20.

OK, let's bring this up at the next Core Software meeting.

May be we can add extra CPP23 IBs for gcc15 and gcc16 just to make sure we build the sw stack with c++23 and newer compilers.

Yes, absolutely.

@makortel

Copy link
Copy Markdown
Contributor

I have no objections moving default gcc15/gcc16 to c++20.

OK, let's bring this up at the next Core Software meeting.

FWIW I'm fine with the suggestion (in case you don't want to wait for the next core meeting that might take a while).

May be we can add extra CPP23 IBs for gcc15 and gcc16 just to make sure we build the sw stack with c++23 and newer compilers.

Yes, absolutely.

This sounds good to me as well.

@smuzaffar

Copy link
Copy Markdown
Contributor

ok, I will change GCC15/16 OBs to use default c++20 and also open CPP23 IBs for GCC15/16

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.

4 participants