Document the producer-private adaptive path, the dummy support workload diagnostics, the negative result of the old serial binning and the 2026-09-13 full-day comparison (splat 282.944 s vs atomic 481.417 s). Add the corresponding raw logs, hotspot fixtures and runner evidence, and lay out the staged plan toward a sub-240 s full-day render without changing the default atomic path.
9 lines
1.2 KiB
Plaintext
9 lines
1.2 KiB
Plaintext
COMMAND OMP_NUM_THREADS=16 OMP_DYNAMIC=FALSE env HIP_PSF_ACCUMULATION=adaptive build/Release/replay_psf benchmarks/hip_psf_replay_2026-09-07/center_hotspot_65536.events 65536 production 128
|
|
TIMEOUT 45s
|
|
GIT 2d01c2c101ccea8193a64bb46ceafca3009f0ef5
|
|
BINARY_SHA256 d056d76de917e764c4f1d516a51657c8d9e9cf3adf97615ccb50b6c81a565002
|
|
PSF cache ready: 64x64 phases, radius 47 px, relative tail 1e-08, tail abs 1e-06, boundary 1e-07, build 0.690 s
|
|
device=AMD Radeon RX 9070 arch=gfx1201 events=65536 rounds=128 frame=3840x2160 mode=production distribution=captured contributions_per_round=432818979 CPU_reference=0.812561989 create=0.043085814
|
|
complete_replay=4.044305801 replay_wall=4.001219988 select=0.023102934 bin=1.362492204 upload=0.093367771 prepare=0.000000000 accumulation=3.760218307 merge=0.000000000 download=0.004831033 cleanup=0.000000000 events_s=2096512.570 contributions_s=13845984344.765 center_tiles=0 tile_chunks=512 atomic_chunks=0 refs=0 tasks=0 scratch_used_peak=0 scratch_device_reserved=0 RSS_KiB=714012 max_abs=6.9388939039072284e-16 max_rel=3.2951664415507829e-14 flux_R=-3.1661685557721843e-16 flux_G=1.3715371417772548e-15 flux_B=4.6151360141508173e-16 flux_Y=7.34376815540858e-16 validation=scaled-cpu-double-reference
|
|
PROCESS_WALL 5.878062827 EXIT 0
|