rustytorch

embedded3 PERF-3 P3-1: the device predictor's interior fast path (bit 11, default ON; RTX_E3_PREDICT_FAST=0 = uniform path), the face update templated on the component + unrolled (392-byte local frame -> 64), exact-zero divisions/sqrt skipped on the fa…

Performance Benchmarks #961:Commit b126915759 pushed by osobh

main

2026-09-20 01:30:05 +00:00

6s

embedded3: GuessBasis re-export behind the cuda feature (the host-only build broke); A2 build: no convective exchange with prescribed faces (RTX_E3_EXCHANGE_CONVECTION=off, host + device bit 10, load route consistent)

Documentation #960:Commit 77ecc54b24 pushed by osobh

main

2026-09-20 00:22:17 +00:00

5s

embedded3 A3-i: the advancing-wall friction closure (RTX_E3_WALL_ADVANCING=1, host + device, per-face normal-velocity table; f(0) = 1 exactly) — REFUTED as built on the slab ladder (flag shear 34.2 / 33.6 / 31.1 vs 28.9 / 32.3 / 30.1: spread 3.4 -> 3.…

Documentation #957:Commit d7250bc4cf pushed by osobh

main

2026-09-20 00:21:29 +00:00

1m35s

embedded3 A3-i: the advancing-wall friction closure (RTX_E3_WALL_ADVANCING=1, host + device, per-face normal-velocity table; f(0) = 1 exactly) — REFUTED as built on the slab ladder (flag shear 34.2 / 33.6 / 31.1 vs 28.9 / 32.3 / 30.1: spread 3.4 -> 3.…

Performance Benchmarks #955:Commit d7250bc4cf pushed by osobh

main

2026-09-20 00:19:59 +00:00

6s

embedded3 PERF-3 P2: the Poisson initial guess by projection onto the last K corrections of the same corrector index (GuessBasis, DeviceCg::project_guess: Gram matrix rebuilt per solve, Cholesky with pivot dropping; e3_cg_axpy) behind RTX_E3_POISSON_PR…

Documentation #954:Commit d4cd7d9545 pushed by osobh

main

2026-09-19 23:58:04 +00:00

6s

embedded3 PERF-3 P1-2: the flag test's centreline polyline cached per thread and time (the 41-point rebuild with four transcendental evaluations each ran per surface-velocity call, ~10^6 calls per step): table faces 1,620 -> 54 ms, impose 622 -> 12 ms,…

Documentation #951:Commit 845e0ae01a pushed by osobh

main

2026-09-19 23:44:39 +00:00

1m37s

embedded3 PERF-3 P1-2: the flag test's centreline polyline cached per thread and time (the 41-point rebuild with four transcendental evaluations each ran per surface-velocity call, ~10^6 calls per step): table faces 1,620 -> 54 ms, impose 622 -> 12 ms,…

Performance Benchmarks #949:Commit 845e0ae01a pushed by osobh

main

2026-09-19 23:43:05 +00:00

5s

embedded3 PERF-3 P1-2: the predictor's device tables derived from the projection's (shared buffers move over; 282 -> 48 ms), the operator key lazy after a refresh (240 -> 0 ms), parallel active / owner maps — slab flag CSV byte-identical; RTX_E3_TABLE…

Documentation #948:Commit d9a2752cb9 pushed by osobh

main

2026-09-19 23:36:37 +00:00

1m27s

embedded3 PERF-3 P1-2: the predictor's device tables derived from the projection's (shared buffers move over; 282 -> 48 ms), the operator key lazy after a refresh (240 -> 0 ms), parallel active / owner maps — slab flag CSV byte-identical; RTX_E3_TABLE…

Performance Benchmarks #946:Commit d9a2752cb9 pushed by osobh

main

2026-09-19 23:35:16 +00:00

6s

embedded3: export_fine_from is crate-private (it takes a crate-private level)

Documentation #945:Commit f6f1fe048b pushed by osobh

main

2026-09-19 21:43:52 +00:00

1m35s

embedded3: export_fine_from is crate-private (it takes a crate-private level)

Performance Benchmarks #943:Commit f6f1fe048b pushed by osobh

main

2026-09-19 21:42:21 +00:00

5s

embedded3 PERF-3 P1-1: the level constructor's masking pass, cell lists and the diagonal sum as parallel per-entry maps (level 0.90 -> 0.60 s per step at 11.6 M cells), digit-identical (slab flag CSV byte-identical; host Poisson / MMS / identity and de…

Documentation #942:Commit 73e43d597d pushed by osobh

main

2026-09-19 21:41:56 +00:00

20s

embedded3 PERF-3 P1-1: the level constructor's masking pass, cell lists and the diagonal sum as parallel per-entry maps (level 0.90 -> 0.60 s per step at 11.6 M cells), digit-identical (slab flag CSV byte-identical; host Poisson / MMS / identity and de…

Performance Benchmarks #940:Commit 73e43d597d pushed by osobh

main

2026-09-19 21:41:37 +00:00

5s

embedded3 PERF-3 P1-1: the moving path's host work, digit-identical — impose_from's solid faces by z plane in parallel (1.95 -> 0.62 s per step at 11.6 M cells), the f32 fine export derived from the f64 level the refresh already builds (export_fine 1.…

Documentation #939:Commit 40268ad19b pushed by osobh

main

2026-09-19 20:59:31 +00:00

7s

embedded3 S2-8: RTX_E3_PROBE_D (diagnostic: the reconstructed route's probes at D h and 2 D h)

Documentation #936:Commit 103e2576ae pushed by osobh

main

2026-09-19 15:09:08 +00:00

6s

embedded3 S2-8 instrument: the load routes in PARTS per body — exchange split into diffusive / convective (cut_wall_exchange_parts), a diagnostic x window on every load route (exchange::set_load_window), the flag test as a 2D periodic slab (RTX_E3_FLA…

Documentation #933:Commit 5e1b3e0731 pushed by osobh

main

2026-09-19 13:53:52 +00:00

6s

embedded3 oblique instrument: RTX_E3_OBLIQUE_DRIVE=pressure (no body force; the sides pin the flow rate)

Documentation #930:Commit 3f0bd0650a pushed by osobh

main

2026-09-19 00:12:49 +00:00

15s

embedded3 oblique instrument: RTX_E3_OBLIQUE_DRIVE=pressure (no body force; the sides pin the flow rate)

Performance Benchmarks #928:Commit 3f0bd0650a pushed by osobh

main

2026-09-19 00:12:38 +00:00

5s

embedded3 S2-6: the three closures default ON (oblique distance, transverse centroid correction, fine floor) — RTX_E3_WALL_DISTANCE=face RTX_E3_DIFFUSION_TRANSVERSE=0 RTX_E3_DISTANCE_FLOOR=coarse reproduce every earlier record digit for digit (checked…

Documentation #927:Commit dbc5dbaefb pushed by osobh

main

2026-09-19 00:06:14 +00:00

16s

embedded3 S2-6: the three closures default ON (oblique distance, transverse centroid correction, fine floor) — RTX_E3_WALL_DISTANCE=face RTX_E3_DIFFUSION_TRANSVERSE=0 RTX_E3_DISTANCE_FLOOR=coarse reproduce every earlier record digit for digit (checked…

Performance Benchmarks #925:Commit dbc5dbaefb pushed by osobh

main

2026-09-19 00:06:03 +00:00

5s

embedded3 S2-6: the transverse gradient from the face at least 0.2 h off the wall (full neighbours over their own distance along the cut partner's normal) — the first form blew up at DFG ny 61 with the fine floor (coefficient ~ 1/d_f); host = device 7…

Documentation #924:Commit 9c3fac0755 pushed by osobh

main

2026-09-18 21:56:00 +00:00

1m2s

embedded3 S2-6: the transverse gradient from the face at least 0.2 h off the wall (full neighbours over their own distance along the cut partner's normal) — the first form blew up at DFG ny 61 with the fine floor (coefficient ~ 1/d_f); host = device 7…

Performance Benchmarks #922:Commit 9c3fac0755 pushed by osobh

main

2026-09-18 21:55:00 +00:00

6s

embedded3 S2-6: the oblique-wall instrument (tests/embedded3_wall_position_oblique.rs: z-flow / in-plane Poiseuille + Couette linear exactness; gate oblique_wall_position_is_second_order) and the closures it found, host + device, default OFF — the tra…

Documentation #921:Commit e9b2e7887b pushed by osobh

main

2026-09-18 21:23:03 +00:00

18s

embedded3 S2-6: the oblique-wall instrument (tests/embedded3_wall_position_oblique.rs: z-flow / in-plane Poiseuille + Couette linear exactness; gate oblique_wall_position_is_second_order) and the closures it found, host + device, default OFF — the tra…

Performance Benchmarks #919:Commit e9b2e7887b pushed by osobh

main

2026-09-18 21:22:49 +00:00

6s

embedded3 S2-3d: face-momentum unsteady term for the box (Mask::box_momentum_rates, Solver::previous_apertures); falsifier prints the residual against it

Documentation #918:Commit 9ae0e42dcd pushed by osobh

main

2026-09-18 17:58:44 +00:00

6s

embedded3 S2-3d: face-momentum unsteady term for the box (Mask::box_momentum_rates, Solver::previous_apertures); falsifier prints the residual against it

Performance Benchmarks #916:Commit 9ae0e42dcd pushed by osobh

main

2026-09-18 17:58:40 +00:00

7s

embedded3 S2-2b: the moving path's host work 5.0 s → 0.6 s per step at 1.45 M cells, bit-identical — Problem::diagonals (the per-cell link diagonal was cells × links: 1.8 s twice per step), counting CSR + rayon in the device tables, rayon in the cut …

Documentation #915:Commit cdd5ad0b66 pushed by osobh

main

2026-09-18 17:50:25 +00:00

1m34s

embedded3 S2-5: centroid diffusion default ON (RTX_E3_DIFFUSION_CENTROID=0 restores the records); flat_wall_position_is_second_order gate; closure.rs split (cutwall.rs under 700)

Documentation #912:Commit 99e4a7214b pushed by osobh

main

2026-09-18 17:39:00 +00:00

1m0s

embedded3 S2-5: centroid diffusion default ON (RTX_E3_DIFFUSION_CENTROID=0 restores the records); flat_wall_position_is_second_order gate; closure.rs split (cutwall.rs under 700)

Performance Benchmarks #910:Commit 99e4a7214b pushed by osobh

main

2026-09-18 17:38:04 +00:00

6s

embedded3 S2-5: the exchange route mirrors the predictor's centroid spacing toward solid neighbours (operator = box back to the box's own error)

Documentation #909:Commit 37086300ba pushed by osobh

main

2026-09-18 16:01:40 +00:00

6s