Skip to content

Please consider the following formatting changes to #14747 #6

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.

Sign up for GitHub

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

Closed
alibuild wants to merge 2 commits into from
Closed

Please consider the following formatting changes to #14747 #6

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d8b4c8e
GPU TPC: improvement for dynamic buffer size for track-model decoding
cima22 Oct 17, 2025
214bf4e
Please consider the following formatting changes
alibuild Oct 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions GPU/GPUTracking/DataCompression/GPUTPCDecompression.cxx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -117,8 +117,10 @@ void GPUTPCDecompression::RegisterMemoryAllocation()

void GPUTPCDecompression::SetMaxData(const GPUTrackingInOutPointers& io)
{
mMaxNativeClustersPerBuffer = *std::max_element(mInputGPU.nSliceRowClusters, mInputGPU.nSliceRowClusters + mInputGPU.nSliceRows);
float clsRatio = mInputGPU.nUnattachedClusters > 0 ? float(mInputGPU.nAttachedClusters) / float(mInputGPU.nUnattachedClusters) : 1.0f;
mMaxNativeClustersPerBuffer *= clsRatio;
mMaxNativeClustersPerBuffer = std::min(mMaxNativeClustersPerBuffer, mRec->GetProcessingSettings().tpcMaxAttachedClustersPerSectorRow);
uint32_t maxAttachedClsMargin1 = *std::max_element(mInputGPU.nSliceRowClusters, mInputGPU.nSliceRowClusters + mInputGPU.nSliceRows);
float clsRatio1 = std::max(mInputGPU.nUnattachedClusters > 0 ? float(mInputGPU.nAttachedClusters) / float(mInputGPU.nUnattachedClusters) : 1.5f, 1.5f); // minimum raio = 1.5
maxAttachedClsMargin1 *= clsRatio1;
uint32_t maxAttachedClsMargin2 = mInputGPU.nAttachedClusters / mInputGPU.nSliceRows * 3; // mean #attached cls per SectorRow multiplied by tuned number 3
mMaxNativeClustersPerBuffer = std::max({maxAttachedClsMargin1, maxAttachedClsMargin2, 1000u}); // take biggest margin, 1000 clusters minimum
mMaxNativeClustersPerBuffer = std::min(mMaxNativeClustersPerBuffer, mRec->GetProcessingSettings().tpcMaxAttachedClustersPerSectorRow); // upperbound given by configurable param
}