wyj
7ddc58515a
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
1.3 KiB
Plaintext
9 lines
1.3 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-parallel 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 0.689 s
|
|
device=AMD Radeon RX 9070 arch=gfx1201 events=65536 rounds=128 frame=3840x2160 mode=production-parallel distribution=captured contributions_per_round=432818979 CPU_reference=0.821504116 create=0.043420076
|
|
complete_replay=4.114833117 replay_wall=4.071413040 select=0.024810469 bin=2.526540803 upload=0.096509804 prepare=0.000000000 accumulation=3.760781384 merge=0.000000000 download=0.004825422 cleanup=0.000000000 events_s=2060367.719 contributions_s=13607273142.154 center_tiles=0 tile_chunks=512 atomic_chunks=0 refs=0 tasks=0 scratch_used_peak=0 scratch_device_reserved=0 RSS_KiB=769216 max_abs=7.6327832942979512e-16 max_rel=3.262701747545357e-14 flux_R=-3.8227195245727636e-16 flux_G=1.6427472194509663e-15 flux_B=3.4955858077692879e-16 flux_Y=8.7225458050005395e-16 validation=scaled-cpu-double-reference
|
|
PROCESS_WALL 5.977715039 EXIT 0
|