Slava Krutelyov

> pow(x,2) : no the compiler will not substitute with x*x: we need to introduce our own inline function "square" depends on the optimization flags according to [some old posts](https://stackoverflow.com/questions/6321170/is-there-any-advantage-to-using-powx-2-instead-of-xx-with-x-double)

> #39811 provides another example (again in `12_5_X`): > > https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-b4c6cc/28422/summary.html there is one more pixelPair step track candidate relative to the baseline https://tinyurl.com/26w5bw6a This iteration is not using mkFit....

> there is one more pixelPair step track candidate relative to the baseline https://tinyurl.com/26w5bw6a and apparently 2 "existing" track candidates are different (in addition to having one more), based on...

> This iteration is not using mkFit. So, it's not obvious why the difference would be localized in the mkfit wf. uhm, I'm wrong, pixelPair in this setup is using...

> > IIUC, this has apparently stopped in recent PR tests - without either an explicit fix or removing the workflow ? > > For a strange coincidence I was...

> * This can suggest that covariance packing schema is a kind of global this is by design.

PackedCandidate interface does not support a switch between variants during runtime; there is a `call_once`.

wouldn't it be more practical to use PF isolation? CaloTower-based isolation in phase-2 isn't supposed to work in the endcaps; and from this issue it seems like the barrel is...

> > wouldn't it be more practical to use PF isolation? > > I was told to use tracker isolation instead. track iso should work OK

it would be nice to get a stack trace with line numbers