Closed efaulhaber closed 6 days ago
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 89.81%. Comparing base (
768c62a
) to head (ebd89fe
).
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
The
Iterators.flatten
stuff does not work inside AMD GPU kernels.This version is also faster than the original code with
Iterators.flatten
on the CPU:This is a speedup of ~20% for the ultra cheap count neighbors benchmark.
For an actual WCSPH simulation, the new code is faster for small problems, but the difference disappears as the problem becomes larger: