kferjaoui 9921944772 CUDA: zero-copy collection, pipelined slots, optional kernel timing ...

Host-side pipeline work ported from the benchmark branch. The kernel is
unchanged, so cluster results are identical.

  - collect_view()/BatchView: read clusters in place from the pinned D2H
    buffer instead of copying one ClusterVector per frame. The view is
    released back to the finder, so keep nothing that borrows it.
  - reserve_output_slots(), chunk_size_for(): pre-pin the output slots and
    size a batch to fit one.
  - find_cluster_views_batched_iter(): submits chunk i+1 before collecting
    chunk i.
  - time_kernels, default off. avg_kernel_time_ms() returns NaN unless
    enabled, and always for the Graph variant.
  - n_streams defaults to 4 everywhere.

ClusterFinderCUDA.ipynb cut to CPU MT vs batched CUDA plus a cluster-set
diff, and it now reports peak clusters/frame.

Removed src/ClusterFinderCUDA{,_old}.test.cu and src/Makefile (early C++
benchmarks, never in the CMake build), the perf and Frozen-vs-CUDA
notebooks, and helper.py.

2026-09-02 12:20:38 +02:00

..

aare

CUDA: zero-copy collection, pipelined slots, optional kernel timing

2026-09-02 12:20:38 +02:00

examples

Feature/minuit2 wrapper (#279)

2026-03-30 09:12:23 +02:00

src

CUDA: zero-copy collection, pipelined slots, optional kernel timing

2026-09-02 12:20:38 +02:00

tests

CUDA: zero-copy collection, pipelined slots, optional kernel timing

2026-09-02 12:20:38 +02:00

CMakeLists.txt

Merge 'origin/main' into feature/cuda_clusterfinder

2026-07-21 15:21:49 +02:00