Files
GR-raytracing/benchmarks/hip_psf_replay_2026-09-07/summary.log
T
wyj 11e703e33d Doc: record bounded HIP PSF replay results and provenance
Archive fixed catalog subsets and captured events, manifests, reproducible commands, raw timing and regression logs, and device resource evidence. Document distribution-dependent tile reduction gains and independent indexed producer throughput.

Update the GPU plan and renderer design while retaining the production atomic path. Preserve historical measurement metadata and exclude generated Python bytecode.
2026-09-11 16:45:48 -04:00

65 lines
6.1 KiB
Plaintext

v2
v2_disperse_center_65536_atomic repeats 1 replay_ms 85.452 with_create_ms 123.554 range_ms [85.452, 85.452]
v2_large_center_65536_16 repeats 3 replay_ms 74.054 with_create_ms 112.393 range_ms [68.598, 77.621]
v2_large_center_65536_32 repeats 3 replay_ms 75.721 with_create_ms 114.148 range_ms [73.366, 76.261]
v2_large_center_65536_atomic repeats 3 replay_ms 99.683 with_create_ms 137.433 range_ms [98.39, 102.176]
v2_medium_center_16384_16 repeats 3 replay_ms 24.282 with_create_ms 65.63 range_ms [24.053, 26.177]
v2_medium_center_16384_32 repeats 3 replay_ms 26.311 with_create_ms 66.045 range_ms [23.744, 26.852]
v2_medium_center_16384_atomic repeats 3 replay_ms 32.315 with_create_ms 71.402 range_ms [30.973, 32.675]
v2_medium_lensed_16384_16 repeats 3 replay_ms 26.774 with_create_ms 66.009 range_ms [24.103, 28.123]
v2_medium_lensed_16384_32 repeats 3 replay_ms 26.144 with_create_ms 66.499 range_ms [25.076, 27.079]
v2_medium_lensed_16384_atomic repeats 3 replay_ms 26.239 with_create_ms 66.147 range_ms [22.994, 27.756]
v2_medium_ordinary_12765_16 repeats 3 replay_ms 22.537 with_create_ms 62.384 range_ms [21.729, 22.871]
v2_medium_ordinary_12765_32 repeats 3 replay_ms 19.633 with_create_ms 60.091 range_ms [19.503, 23.481]
v2_medium_ordinary_12765_atomic repeats 3 replay_ms 23.768 with_create_ms 63.752 range_ms [23.604, 24.604]
v2_small_center_4096_16 repeats 3 replay_ms 10.689 with_create_ms 48.588 range_ms [10.488, 10.844]
v2_small_center_4096_32 repeats 3 replay_ms 12.403 with_create_ms 51.216 range_ms [10.82, 13.49]
v2_small_center_4096_atomic repeats 3 replay_ms 15.714 with_create_ms 54.666 range_ms [15.309, 17.685]
v2_small_lensed_4096_16 repeats 3 replay_ms 13.589 with_create_ms 52.228 range_ms [13.44, 13.631]
v2_small_lensed_4096_32 repeats 3 replay_ms 11.02 with_create_ms 50.269 range_ms [10.825, 11.047]
v2_small_lensed_4096_atomic repeats 3 replay_ms 10.889 with_create_ms 49.715 range_ms [10.761, 10.926]
v2_small_ordinary_4096_16 repeats 3 replay_ms 10.533 with_create_ms 50.115 range_ms [10.443, 13.578]
v2_small_ordinary_4096_32 repeats 3 replay_ms 11.01 with_create_ms 50.588 range_ms [10.49, 13.236]
v2_small_ordinary_4096_atomic repeats 3 replay_ms 11.874 with_create_ms 50.241 range_ms [10.412, 16.834]
v3
v3_disperse_center_65536_16 repeats 3 replay_ms 97.573 with_create_ms 137.62 range_ms [91.878, 99.797]
v3_disperse_center_65536_32 repeats 3 replay_ms 79.217 with_create_ms 118.072 range_ms [77.01, 79.688]
v3_disperse_center_65536_atomic repeats 3 replay_ms 80.459 with_create_ms 119.9 range_ms [69.512, 83.669]
v3_large_center_65536_16 repeats 3 replay_ms 72.631 with_create_ms 112.42 range_ms [69.754, 73.514]
v3_large_center_65536_32 repeats 3 replay_ms 69.657 with_create_ms 107.631 range_ms [66.273, 71.433]
v3_large_center_65536_atomic repeats 3 replay_ms 99.19 with_create_ms 137.292 range_ms [98.205, 99.258]
v3_medium_center_16384_16 repeats 3 replay_ms 26.757 with_create_ms 66.185 range_ms [24.946, 26.935]
v3_medium_center_16384_32 repeats 3 replay_ms 23.263 with_create_ms 63.395 range_ms [23.183, 26.364]
v3_medium_center_16384_atomic repeats 3 replay_ms 31.452 with_create_ms 70.241 range_ms [31.06, 31.854]
v3_medium_lensed_16384_16 repeats 3 replay_ms 23.613 with_create_ms 64.696 range_ms [23.527, 25.844]
v3_medium_lensed_16384_32 repeats 3 replay_ms 25.976 with_create_ms 65.981 range_ms [25.769, 26.691]
v3_medium_lensed_16384_atomic repeats 3 replay_ms 26.422 with_create_ms 67.353 range_ms [26.263, 28.369]
v3_medium_ordinary_12765_16 repeats 3 replay_ms 21.836 with_create_ms 62.243 range_ms [21.43, 22.152]
v3_medium_ordinary_12765_32 repeats 3 replay_ms 20.118 with_create_ms 60.24 range_ms [19.703, 20.916]
v3_medium_ordinary_12765_atomic repeats 3 replay_ms 23.872 with_create_ms 64.293 range_ms [23.644, 24.875]
v3_small_center_4096_16 repeats 3 replay_ms 12.451 with_create_ms 52.377 range_ms [10.155, 13.731]
v3_small_center_4096_32 repeats 3 replay_ms 13.019 with_create_ms 51.135 range_ms [12.644, 14.29]
v3_small_center_4096_atomic repeats 3 replay_ms 15.941 with_create_ms 54.879 range_ms [15.923, 17.838]
v3_small_lensed_4096_16 repeats 3 replay_ms 12.926 with_create_ms 51.185 range_ms [10.251, 13.921]
v3_small_lensed_4096_32 repeats 3 replay_ms 10.096 with_create_ms 52.262 range_ms [10.09, 13.24]
v3_small_lensed_4096_atomic repeats 3 replay_ms 10.813 with_create_ms 49.061 range_ms [10.756, 11.146]
v3_small_ordinary_4096_16 repeats 3 replay_ms 13.164 with_create_ms 54.801 range_ms [12.834, 15.103]
v3_small_ordinary_4096_32 repeats 3 replay_ms 10.62 with_create_ms 50.334 range_ms [9.839, 10.679]
v3_small_ordinary_4096_atomic repeats 3 replay_ms 12.953 with_create_ms 53.111 range_ms [11.99, 13.82]
final
final_center_16384_16 repeats 3 replay_ms 22.621 with_create_ms 55.894 range_ms [22.427, 25.791]
final_center_16384_32 repeats 3 replay_ms 23.247 with_create_ms 58.128 range_ms [22.836, 25.648]
final_center_16384_atomic repeats 3 replay_ms 31.803 with_create_ms 65.47 range_ms [30.801, 31.883]
final_center_65536_16 repeats 3 replay_ms 69.823 with_create_ms 102.323 range_ms [69.782, 70.063]
final_center_65536_32 repeats 3 replay_ms 73.093 with_create_ms 105.949 range_ms [72.984, 73.511]
final_center_65536_atomic repeats 3 replay_ms 104.107 with_create_ms 137.111 range_ms [103.795, 104.969]
final_dispersed_65536_16 repeats 3 replay_ms 104.389 with_create_ms 138.363 range_ms [103.147, 106.065]
final_dispersed_65536_32 repeats 3 replay_ms 79.899 with_create_ms 112.963 range_ms [77.449, 81.358]
final_dispersed_65536_atomic repeats 3 replay_ms 85.302 with_create_ms 119.046 range_ms [84.854, 85.394]
final_lensed_16384_16 repeats 3 replay_ms 23.814 with_create_ms 57.39 range_ms [22.522, 25.344]
final_lensed_16384_32 repeats 3 replay_ms 25.388 with_create_ms 58.896 range_ms [22.801, 29.41]
final_lensed_16384_atomic repeats 3 replay_ms 26.536 with_create_ms 59.364 range_ms [25.932, 27.395]
final_ordinary_12765_16 repeats 3 replay_ms 18.511 with_create_ms 53.027 range_ms [18.175, 18.706]
final_ordinary_12765_32 repeats 3 replay_ms 19.286 with_create_ms 53.295 range_ms [19.154, 19.637]
final_ordinary_12765_atomic repeats 3 replay_ms 26.423 with_create_ms 61.551 range_ms [24.441, 27.035]