From aaabe04578113f9af166f7f987a95c1e329df159 Mon Sep 17 00:00:00 2001
From: Felix Weiglhofer <weiglhofer@fias.uni-frankfurt.de>
Date: Wed, 31 May 2023 14:36:18 +0000
Subject: [PATCH] algo: Add comments / potential issues to GPU Hitfinder.
---
algo/detectors/sts/StsHitfinder.cxx | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/algo/detectors/sts/StsHitfinder.cxx b/algo/detectors/sts/StsHitfinder.cxx
index 94a5c4173b..0e11237f0d 100644
--- a/algo/detectors/sts/StsHitfinder.cxx
+++ b/algo/detectors/sts/StsHitfinder.cxx
@@ -193,8 +193,13 @@ XPU_D void sts::Hitfinder::FindClustersParallel(FindClusters::context& ctx) cons
if (myDigi.GetChannel() >= otherDigi.GetChannel()) continue;
if (float(myDigi.GetTimeU32()) + deltaT < float(otherDigi.GetTimeU32())) { break; }
+
+ // This check is not necessary? We already found the first possible digi via binary search...
if (float(myDigi.GetTimeU32()) - deltaT > float(otherDigi.GetTimeU32())) { continue; }
+ // How to handle noise? Digis in same channel within a small time window of the previous digi are discarded by old cluster finder
+
+ // How to handle if multiple digis try to connect to the same digi?
digiConnector[iLocal].SetNext(nextIter);
digiConnector[nextIter].SetHasPrevious(true);
break;
--
GitLab