-
Notifications
You must be signed in to change notification settings - Fork 4.4k
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
use faster variant of OmniClusterRef and RefCoreWithIndex initialization #47174
use faster variant of OmniClusterRef and RefCoreWithIndex initialization #47174
Conversation
cms-bot internal usage |
-code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-47174/43414 Code check has found code style and quality issues which could be resolved by applying following patch(s)
|
9045912
to
6689fe4
Compare
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-47174/43415 |
A new Pull Request was created by @slava77 for master. It involves the following packages:
@Dr15Jones, @cmsbuild, @jfernan2, @makortel, @mandrenguyen, @smuzaffar can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
@cmsbuild please test |
+1 Size: This PR adds an extra 40KB to repository Comparison SummarySummary:
|
+1 |
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-47174/43422 |
Pull request #47174 was updated. @Dr15Jones, @cmsbuild, @makortel, @smuzaffar can you please check and sign again. |
@cmsbuild please test even though only the comment was updated; perhaps no need to wait to sign and merge |
+1 Size: This PR adds an extra 32KB to repository Comparison SummarySummary:
|
+core |
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. @rappoccio, @sextonkennedy, @mandrenguyen, @antoniovilela (and backports should be raised in the release meeting by the corresponding L2) |
+1 |
type performance-improvements |
this was motivated by inspection of
SiStripRecHitConverterAlgorithm
in the HLT context (runs without matching; expected to be used in HLT in combination with mkFit)The overhead of making the
OmniClusterRef
viaedmNew::DetSet::makeRefTo
appeared too costly due to mostly trivial calls cluttered by unnecessary out-of-line calls that prevented the compiler optimization.SiStripRecHitConverter::produce
is around 6% (relative);SiStripRecHitConverter
No physics changes are expected.
There are probably benefits from using a similar
OmniClusterRef
constructor in other places, likeSiPixelRecHitFromSoAAlpaka
, but it's not a significant contributor to the total cost.@cms-sw/tracking-pog-l2 @cms-sw/trk-dpg-l2