Color code: (critical, news from this week: blue, news from last week: purple, no news: black)

Sync reconstruction

Async reconstruction

Need to investigate short GPU stall problem.
Limiting factor for pp workflow is now the TPC time series, which is to slow and creates backpressure (costs ~20% performance on EPNs). Enabled multi-threading as recommended by Matthias - need to check if it works.
Test with GPU GRID jobs at NERSC pending.
New builds with GENERIC_NVIDIA / GENERIC_AMD Grid site architecture support working, also virtual sm_75 architecture working generically, we can run on the RTX Pro 6000 in alibicompute with that software.
- Problem that /opt/rocm and nvidia cuda runtime provided by the apptainer nvidia runtime not present in LD_LIBRARY_PATH, Maksim is checking how to fix it. I think this should come from system level, not from O2.
- Will tune existing 16-core settings, add a SITEARCH for 16core CPU, and 16coreCPU + generic NVIDIA / AMD GPU, like for 8 core.
- Will retune EPN async workflow for TPC + ITS on GPU on 2025 data.

GPU ROCm / compiler topics:

Problem with building ONNXRuntime with MigraphX support.
Need to find a way to build ONNXRuntime with support for CUDA and for ROCm.
Try to find a better solution for the problem with __device__ inline functions leaking symbols in the host code.
Miscompilation / internal compiler error fixed in new clang for ROCm 7.x, SDMA engine synchronization bug still not fixed.
- Serialization bug pending.
- Miscompilation on MI 100 leading to memory error pending.
- New miscompilation on MI 50 with ROCm 7.0 when RTC disabled.
- New miscompilation on MI 50 on ROCm 6.3 and 7.0 when RTC enabled, with latest software. Have a workaround for Pb-Pb data taking, but not compatible to latest tracking developments.
- Waiting for ROCm 7.2, which could fix the MI100 serialization issue for good. Not clear yet with regards to miscompilation problems.

TPC / GPU Processing

WIP: Use alignas() or find a better solution to fix alignment of monte carlo labels: https://its.cern.ch/jira/browse/O2-5314
Waiting for TPC to fix bogus TPC transformations for good, then we can revert the workaround.
Final solution: merging transformation maps on the fly into a single flat object:
- Matthias checked the latest version and can reproduce the issues I reported, he is checking with Sergey. Could be related to extrapolation.
Need to check the problem with ONNX external memory allocator.
Next high priority topic: Improvements for cluster sharing and cluster attachment at lower TPC pad rows. PR: https://github.com/AliceO2Group/AliceO2/pull/14542

Other topics:

Ordered connectors and cables for the GPU power supply in the CI server, to build adapter cables manually.
ONNXRuntime was setting /opt/rocm to LD_LIBRARY_PATH. Don't understand why, but we should do this on system level, not on alidist level. In any case, O2 had it only accidentally. Will remove it and see if something breaks, but if yes should be fixed on system level.

EPN GPU Topics: