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
Update openloops-2.0.b to openloops-2.0.0 #4796
Update openloops-2.0.b to openloops-2.0.0 #4796
Conversation
A new Pull Request was created by @SiewYan (SiewYan) for branch IB/CMSSW_10_6_X/gcc700. @cmsbuild, @smuzaffar, @gudrutis, @mrodozov can you please review it and eventually sign? Thanks. |
@SiewYan , for OpenLoops 2.0.b we have some optimization done by @davidlange6 and Philipp Maierhoefer cms-externals/openloops@5141c0e Looks like these are missing in 2.0.0. Do we not need them? |
Hi @smuzaffar , yes, please include the optimization. Thanks. |
@mrodozov , can you please get Openloop 2.0.0 tar file and create cms/2.0.0 branch under cms-externals/openloops and include our patches on top of it? |
@SiewYan , I've added the the tar as initial commit on this branch https://github.com/cms-externals/openloops/tree/cms/v2.0.0 and this commit cms-externals/openloops@5141c0e but couldn't add this one cms-externals/openloops@a0fd889 |
Hi @mrodozov thanks for creating the commit. I am adding previous committer @davidlange6 , @davidlange6 could you comment on the patch cms-externals/openloops@a0fd889 is necessary for new openloop 2.0.0? cc @GurpreetSinghChahal |
I think we need a conversation with open loops before committing to some long term maintenance of patches. At least mine came from the authors directly, so its presumably not forgotten, but its hard to tell that..
Its impossible to tell from the commit history what is there or not, as the gitlab is maintained just by dumping in snapshots
https://gitlab.com/openloops/OpenLoops/commits/master
since the code has diverged quite a bit (at least in the spot checks I did) some more systematic checks are needed. [well, if anything my checks suggest that the fixes are not in, or else were reimplemented]
… On Mar 25, 2019, at 4:44 PM, SiewYan ***@***.***> wrote:
Hi @mrodozov thanks for creating the commit. I am adding previous committer @davidlange6 , @davidlange6 could you comment on the patch ***@***.*** is necessary for new openloop 2.0.0? cc @GurpreetSinghChahal
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.
|
Hi @davidlange6, all: thanks! Let me add @jonaslindert (OpenLoops author). Jonas, a few changes were included via [1] and [2] while integrating OpenLoops 2.0.b in CMS software last year. Please suggest us if we still need those updates in OpenLoops 2.0. @davidlange6: Could you please remind me why these updates were suggested while integrating OpenLoops 2.0.b in CMS software whereas standalone OpenLoops 2.0.b was working without these updates (afaik). Thanks! [1]: optimize the factorial function |
Hi @GurpreetSinghChahal, @davidlange6, right. There have been structural changes in the code so [2] can not directly be applied anymore. cheers, |
@davidlange6: Could you please remind me why these updates were suggested while integrating OpenLoops 2.0.b in CMS software whereas standalone OpenLoops 2.0.b was working without these updates (afaik). Thanks!
Right "working" but with a massive startup time in at least some examples. There are no physics changed, but probably a 50% increase in event throughput in a relval workflow that started to fail because it couldn't run ten events in 2 hours.
…
[1]: optimize the factorial function
[2]: Avoid Collier initialisation in EW renorm with every integral call
|
process libraries seems missing. That should included in the cms version |
…ist into IB/CMSSW_10_6_X/gcc700
Hi, here you find a patch equivalent to [2] linked above. Future versions of OpenLoops will always include this: cheers, |
…ist into IB/CMSSW_10_6_X/gcc700
Pull request #4796 was updated. |
@jonaslindert thanks for the patch. thanks @davidlange6 for the clarification. @mrodozov could you proceed with the patching: https://gitlab.com/openloops/OpenLoops/commit/8ff8f52fe6e0f9f4f3f666b0d89279da64554fc3 Thanks. |
Pull request #4796 was updated. |
please test |
The tests are being triggered in jenkins. |
Comparison is ready Comparison Summary:
|
Propagating changes in cms-sw#4796 1.) Include proclib 2.) Process libraries generation 3.) changes openloops source by imitating gitlab tag hash
Propagating changes in cms-sw#4796 1.) Include proclib 2.) Process libraries generation 3.) changes openloops source by imitating gitlab tag hash
Propagating changes in cms-sw#4796 1.) Include proclib 2.) Process libraries generation 3.) changes openloops source by imitating gitlab tag hash
please test workflow 534.0,536.0 |
The tests are being triggered in jenkins. |
Comparison job queued. |
Comparison is ready @slava77 comparisons for the following workflows were not done due to missing matrix map:
Comparison Summary:
|
+1 the needed patches have been applied |
merge |
No description provided.