From bb9f0cc0ba65352f43091e57907fda9eed010ea8 Mon Sep 17 00:00:00 2001
From: Felix Weiglhofer <felix.weiglhofer@cern.ch>
Date: Mon, 22 Jun 2026 12:41:23 +0200
Subject: [PATCH] GPU/TPC: Increace assumed cacheline size to 128 byte in
 cluster finder

---
 GPU/GPUTracking/TPCClusterFinder/CfArray2D.h           | 10 ++++++++--
 .../TPCClusterFinder/GPUTPCCFCheckPadBaseline.h        |  1 -
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/GPU/GPUTracking/TPCClusterFinder/CfArray2D.h b/GPU/GPUTracking/TPCClusterFinder/CfArray2D.h
index e61ec532bf7e0..358cc682c7982 100644
--- a/GPU/GPUTracking/TPCClusterFinder/CfArray2D.h
+++ b/GPU/GPUTracking/TPCClusterFinder/CfArray2D.h
@@ -86,10 +86,13 @@ class LinearLayout
 template <tpccf::SizeT S>
 struct GridSize;
 
+// GridSize for 1 byte and 2 byte elements are adjusted for 128 byte cachelines,
+// as these are prevelant on modern GPUs.
+
 template <>
 struct GridSize<1> {
   enum {
-    Width = 8,
+    Width = 16,
     Height = 8,
   };
 };
@@ -98,10 +101,13 @@ template <>
 struct GridSize<2> {
   enum {
     Width = 8,
-    Height = 4,
+    Height = 8,
   };
 };
 
+// GridSize for 4 bytes is only used for MC indexing on CPU.
+// So assume 64 byte cachelines here instead.
+
 template <>
 struct GridSize<4> {
   enum {
diff --git a/GPU/GPUTracking/TPCClusterFinder/GPUTPCCFCheckPadBaseline.h b/GPU/GPUTracking/TPCClusterFinder/GPUTPCCFCheckPadBaseline.h
index c2c5a1e339256..b96dfaeaada9a 100644
--- a/GPU/GPUTracking/TPCClusterFinder/GPUTPCCFCheckPadBaseline.h
+++ b/GPU/GPUTracking/TPCClusterFinder/GPUTPCCFCheckPadBaseline.h
@@ -53,7 +53,6 @@ class GPUTPCCFCheckPadBaseline : public GPUKernelTemplate
     TimebinsPerCacheline = TPCMapMemoryLayout<uint16_t>::Height,
     EntriesPerCacheline = PadsPerCacheline * TimebinsPerCacheline,
     NumOfCachedPads = GPUCA_WARP_SIZE / TimebinsPerCacheline,
-    NumCLsPerWarp = GPUCA_WARP_SIZE / EntriesPerCacheline,
     NumOfCachedTBs = TimebinsPerCacheline * 8,
     // Threads index shared memory as [iThread / MaxNPadsPerRow][iThread % MaxNPadsPerRow].
     // Rounding up to a multiple of PadsPerCacheline ensures iThread / MaxNPadsPerRow < NumOfCachedTBs