Add grid_constant qualifier to finding kernels

[stephenswat]

This commit adds the __grid_constant__ qualifier to the CUDA track finding kernel, allowing the compiler to make some additional optimisations. This should also help us better understand performance issues such as the ones in #1259.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[krasznaa]

Looks reasonable.

Does it affect the build or the runtime performance?

[stephenswat]

Physics performance summary

Here is a summary of the physics performance effects of this PR. Command used:

traccc_seeding_example_cuda --input-directory=/data/Acts/odd-simulations-20240506/geant4_ttbar_mu200 --digitization-file=geometries/odd/odd-digi-geometric-config.json --detector-file=geometries/odd/odd-detray_geometry_detray.json --grid-file=geometries/odd/odd-detray_surface_grids_detray.json --material-file=geometries/odd/odd-detray_material_detray.json --input-events=10 --use-acts-geom-source=on --check-performance --truth-finding-min-track-candidates=5 --truth-finding-min-pt=1.0 --truth-finding-min-z=-150 --truth-finding-max-z=150 --truth-finding-max-r=10 --seed-matching-ratio=0.99 --track-matching-ratio=0.5 --track-candidates-range=5:100 --seedfinder-vertex-range=-150:150

Seeding performance

Total number of seeds went from 298341 to 298340 (-0.0%)

Seeding plots

Track finding performance

Total number of found tracks went from 50224 to 50222 (-0.0%)

Finding plots

Track fitting performance

Fitting plots

Seeding to track finding relative performance

Seeding to track finding plots

Note

This is an automated message produced on the explicit request of a human being.

[stephenswat]

Performance summary

Here is a summary of the performance effects of this PR:

Graphical

Tabular

Kernel	Reciprocal Throughput			Parallelism
	3ab5c4f	e7e96b0	Delta	3ab5c4f	e7e96b0
propagate_to_next_surface	7.82 ms	7.82 ms	0.0%	3.46	3.45
find_tracks	1.74 ms	1.73 ms	-0.3%	1.83	1.83
ccl_kernel	825.74 μs	826.94 μs	0.1%	1.37	1.37
count_doublets	813.39 μs	815.23 μs	0.2%	1.61	1.61
count_triplets	566.56 μs	568.98 μs	0.4%	1.02	1.02
find_doublets	527.29 μs	532.53 μs	1.0%	3.08	3.08
Thrust::sort	379.35 μs	379.62 μs	0.1%	7.32	7.32
find_triplets	171.94 μs	170.65 μs	-0.8%	1.31	1.31
estimate_track_params	146.32 μs	146.24 μs	-0.1%	2.68	2.68
build_tracks	124.95 μs	125.25 μs	0.2%	3.71	3.71
select_seeds	58.68 μs	58.82 μs	0.2%	1.34	1.34
populate_grid	23.98 μs	24.05 μs	0.3%	1.22	1.22
count_grid_capacities	22.13 μs	22.16 μs	0.1%	1.22	1.22
remove_duplicates	23.49 μs	20.00 μs	-14.8%	26.11	25.65
fill_sorted_measurements	19.77 μs	19.74 μs	-0.1%	1.13	1.13
update_triplet_weights	14.78 μs	14.85 μs	0.5%	1.27	1.27
apply_interaction	13.86 μs	13.91 μs	0.4%	6.72	6.70
fill_finding_propagation_sort_keys	8.82 μs	8.82 μs	-0.0%	7.65	7.66
form_spacepoints	8.30 μs	8.30 μs	-0.0%	1.48	1.48
reduce_triplet_counts	5.63 μs	5.67 μs	0.6%	3.09	3.08
unknown	5.07 μs	5.08 μs	0.2%	4.26	4.27
fill_finding_duplicate_removal_sort_keys	1.57 μs	1.57 μs	-0.2%	38.02	38.04
Total	13.32 ms	13.32 ms	0.0%	2.99	2.98

Important

All metrics in this report are given as reciprocal throughput, not as wallclock runtime.

Note

This is an automated message produced upon the explicit request of a human being.