wyj
e1ec480669
Cooperate across 32 lanes per PSF and use two completion-protected staging slots with complete batch timing. Restore coarse OpenMP event production while serializing shared GPU submissions and direct fallback boundaries. Add bounded benchmarks, streaming and renderer regressions, and preserve validation evidence and ownership documentation.
10 lines
742 B
Plaintext
10 lines
742 B
Plaintext
PSF cache ready: 64x64 phases, radius 47 px, relative tail 1e-08, tail abs 1e-06, boundary 1e-07, build 0.709 s
|
|
Frame 0: finding and prefetching catalog tiles...
|
|
Frame 0: catalog prefetch finished (0 candidate tiles).
|
|
Frame 0: splatting 130882 lens triangles...
|
|
Frame 0: catalog splatting finished in 0.3 s; writing image...
|
|
Rendered 411 images from 256 catalog stars to /tmp/gr-hip-baseline-256.png (ok; imported lens map)
|
|
PSF splats: cached 411, cached wing-clipped 0, direct fallbacks 0, discarded below min-Y 0
|
|
HIP PSF: 411 events in 1 batches (1 timed); upload 0.000014 s, kernel 0.122597 s, download 0.021485 s
|
|
OMP_NUM_THREADS=16 timeout --kill-after=5s 60s --catalog --psf-relative-tai 10.69s user 0.17s system 774% cpu 1.403 total
|