2026 RelVal Workflows#49419
Conversation
|
enable gpu |
|
cms-bot internal usage |
|
please test |
AdrianoDee
force-pushed
the
relval_2026_wfs
branch
from
f92c857 to
009f087
Compare
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-49419/46876
|
|
A new Pull Request was created by @AdrianoDee for master. It involves the following packages:
@AdrianoDee, @DickyChant, @Martin-Grunewald, @antoniovagnerini, @arunhep, @atpathak, @cmsbuild, @davidlange6, @fabiocos, @ftenchini, @mandrenguyen, @miquork, @mmusich, @perrotta can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
please test |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-49419/46877
|
AdrianoDee
force-pushed
the
relval_2026_wfs
branch
from
009f087 to
63ca699
Compare
|
please abort |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-49419/46905
|
|
Pull request #49419 was updated. @AdrianoDee, @Alejandro1400, @DickyChant, @JanChyczynski, @Martin-Grunewald, @antoniovagnerini, @arunhep, @atpathak, @cmsbuild, @davidlange6, @fabiocos, @ftenchini, @mandrenguyen, @miquork, @mmusich, @perrotta can you please check and sign again. |
|
please test |
|
|
||
| from Configuration.Eras.Era_Run3_2025_FastSim_cff import Run3_2025_FastSim | ||
|
|
||
| Run3_2026_FastSim = cms.ModifierChain(Run3_2025_FastSim) |
There was a problem hiding this comment.
Just to point out that all other Run*_FastSim eras are defined starting from Run*. Is the deviation from that pattern now intentional? (e.g. betting that Run3_2026 will always be the same as Run3_2025)
There was a problem hiding this comment.
For me, this is just a placeholder here. I would ask @cms-sw/simulation-l2 for suggestions.
There was a problem hiding this comment.
This is fine for now. I actually hope we will be able to eliminate the Run3_202X_FastSim Eras entirely in the next couple of months, if the GEM developments converge. We will propagate the appropriate updates at that time.
Just to let you know that, as for now, those are the only GTs available that can be used for this purpose. They can be replaced by the dedicated GTs with 2026 conditions as soon as they will become available |
|
I suspect the AMD_w7900 tests are stuck. For this specific PR, maybe we can skip them. |
|
I have restarted the comparison job. results should be available in few minutes |
|
+1 Size: This PR adds an extra 100KB to repository Comparison SummarySummary:
AMD_MI300X Comparison SummarySummary:
AMD_W7900 Comparison SummarySummary:
NVIDIA_H100 Comparison SummarySummary:
NVIDIA_L40S Comparison SummarySummary:
NVIDIA_T4 Comparison SummarySummary:
|
|
+pdmv |
|
+alca |
| @@ -14,6 +14,7 @@ | |||
| 'relval2023' : 'Fake2', | |||
| 'relval2024' : 'Fake2', | |||
| 'relval2025' : '2025v13', | |||
There was a problem hiding this comment.
eventually, now that we have coverage I guess we can demote this to fake (in another PR).
There was a problem hiding this comment.
Yes, I'd wait to do that after pre3 is out just to use 2025 for one last round of validations.
|
+hlt |
|
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. @sextonkennedy, @mandrenguyen, @ftenchini (and backports should be raised in the release meeting by the corresponding L2) |
|
Out of curiosity @cms-sw/dqm-l2 is it expected that in this PR the DQM bin-by-bin comparison is empty: https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_16_0_X_2025-11-21-1100+fd870d/71938/dqm-histo-comparison-summary.html ? |
I'd say that is unusual. I was taking a look at signed and merged PRs from last week that target 16_0_X and the cmsdt pages are still available and most of them also are displaying empty DQM bin-by-bin comparison. On this specific PR, I cannot find the |
|
shall we try to test again? |
|
https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_16_0_X_2025-11-21-1100+fd870d/71938/dqmBinByBinLog.log shows some root/python errors [a] . Could this be the reason the dqm bin-by-bin comparison were not properly done ? Could it be due to ROOT 6.36 update? [a] |
That is right, at the end of the log we can see that there were no files to be uploaded as well.
I don't think testing again will solve the issue. |
so just to be clear, the issue is not coming specifically from this PR, but it's also present in all other recent PRs? |
I don't think this is related to thir PR. Two other cases that this happened, before this PR:
Looks like this is not happening on 15_1_X. |
|
this might be related to ROOT 6.36 change root-project/root@ff65e3b which is in 16.0.X IBs . |
|
Do we need to updates in https://github.com/cms-sw/cmssw/blob/master/DQMServices/FileIO/scripts/compareHistograms.py for ROOT change root-project/root#18735 ? |
|
I am opening a new issue to track dqm bin-by-bin failures |
thanks! |
|
+1 |
9 participants
PR description:
The 2026 version of #46445.
This PR proposes the addition of 2026 wfs to the matrix. For the moment, they are a copy of what we run currently for 2025 with the idea that once 2026 parameters (conditions, HLT menu) come, we will need to fill the dummy ones (keeping them in sync with each other). For the moment:
@relval2026is pointing toGRun(@cms-sw/hlt-l2, let me know if you prefer something different there);auto:phase1_2026_*symbolic GTs are copies of2025ones (@cms-sw/alca-l2 let me know what you think about this, I'll updated them following your suggestion);The era is
Era_2026that, for now, is just a copy ofEra_2025(the same for FastSim, just fyi @cms-sw/simulation-l2). TTbar 2026 noPU/PU wfs are added to the limited matrix to be tested with PRs. The standard PU input for 2026 wfs is the same as 2025 for now. I took the chance to update also the GPU workflows to 2026 (@cms-sw/heterogeneous-l2) so that we avoid running them with the fake menu when 2025 wfs will be switching to that (as happened last year).PR validation:
New wfs run.