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

Browse Source

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.

...

This commit is contained in:

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

1 parent ce256dd30f

commit 9921944772

13 files changed

+1358 -4060

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

python/tests/ClusterFinderCUDA.ipynb

+567 -1036

File diff suppressed because it is too large. Load diff