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 build/Release/replay_psf benchmarks/hip_psf_replay_2026-09-07/center_65536.events 65536 adaptive 256
|
|
TIMEOUT 45s
|
|
GIT 2d01c2c101ccea8193a64bb46ceafca3009f0ef5
|
|
BINARY_SHA256 4b2e99b04076bce7856d0ca2437ec69809dc9846cc1f28eb31e9441da7fb5554
|
|
PSF cache ready: 64x64 phases, radius 47 px, relative tail 1e-08, tail abs 1e-06, boundary 1e-07, build 0.707 s
|
|
device=AMD Radeon RX 9070 arch=gfx1201 events=65536 rounds=256 frame=3840x2160 mode=adaptive distribution=captured contributions_per_round=432818979 CPU_reference=0.812426805 create=0.038770914
|
|
complete_replay=12.178713083 replay_wall=12.139942169 select=0.039789200 bin=1.752534628 upload=0.268706799 prepare=0.242101908 accumulation=9.592133045 merge=0.114989281 download=0.004442974 cleanup=0.000149965 events_s=1381984.837 contributions_s=9127033480.045 center_tiles=8960 tile_chunks=1024 atomic_chunks=0 refs=761286400 tasks=3055360 scratch_used_peak=35382764 scratch_device_reserved=184549376 RSS_KiB=713688 max_abs=0.01589053438331841 max_rel=255.00000000000605 flux_R=255.00000000000003 flux_G=255.0000000000001 flux_B=255.00000000000009 flux_Y=255.00000000000007 validation=skipped-for-sustained-timing
|
|
PROCESS_WALL 14.035563654 EXIT 0
|