igneum-labs
eb96a047f5
Merge master 0a1639d6b into v607-floor-memory under the master-landing lock
2026-10-08 22:20:07 +00:00
igneum-labs
c298adc10e
Merge remote-tracking branch 'box/master' into v607-floor-memory
...
# Conflicts:
# docs/plans/igneum-2.0-test-harness-map.md
# docs/plans/igneum-2.0-test-registry.json
2026-10-08 22:12:32 +00:00
igneum-labs
6814f12210
Merge master 73b8a9170 into census-chainseed-trace under the master-landing lock
2026-10-08 22:04:15 +00:00
igneum-labs
f9419f2138
Class v6 census: the window trace on the signing program 2a1d6caab4c24564 (per-site ratio 1.0026, 1.021x of the foldrw control on the item share)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:57:42 +00:00
igneum-labs
d4718b2df7
Merge remote-tracking branch 'box/master' into same-work-harness
...
# Conflicts:
# docs/plans/igneum-2.0-test-registry.json
# tools/ci/batches/same-work-20261008-02.json
2026-10-08 21:51:48 +00:00
igneum-labs
81113bb2df
V6-07: the host bisect names 7dd7c35e1 (FixtureEnv field without the guest re-pin) as the 0-bytes fault; amendments
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:49:15 +00:00
igneum-labs
52c2483392
same-work: the batch notes and contexts name no rig short name (the acceptance page rule, 22:4x UK); statuses unchanged
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:43:02 +00:00
igneum-labs
165cdb51ed
Merge master 18f8aad34 into class-v6-family-gate under the master-landing lock
2026-10-08 21:39:56 +00:00
igneum-labs
7e6a7e467d
Merge master e619e6433 into same-work-harness under the master-landing lock
2026-10-08 21:36:05 +00:00
igneum-labs
e619e6433a
Merge hash-registry-1 835da0cb into master (gate: green on c3b04f45, recorded by tools/ci/pre-push.sh; landed on the build mirror)
2026-10-08 21:34:44 +00:00
igneum-labs
835da0cbc7
Merge master 762391a30 into hash-registry-1 under the master-landing lock
2026-10-08 21:33:45 +00:00
igneum-labs
351f4967ba
Merge remote-tracking branch 'box/master' into same-work-harness
...
# Conflicts:
# docs/plans/igneum-2.0-test-registry.json
2026-10-08 21:28:29 +00:00
igneum-labs
c3b04f454f
pairing row: UTC stamps (the identity grep forbids local-time zones in the export list)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:24:09 +00:00
igneum-labs
a36bb38bd9
Class v6 census: the chain-seed draw 2a1d6caab4c24564 on the frozen object (the engine's pair: seed af89be5d, era genesis, day 20730, node1 state) through the (c''') rule and the attempts census, PASS; the block-hash-era draw beside it as the comparison; registry run census-v6-chainseed-20261008-2230
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:23:10 +00:00
igneum-labs
3bd6d19c44
same-work: context 03 (the live chain's class v5 object 0x81fbfa3aa413173f at epoch 2) as read back on build-1
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:21:17 +00:00
igneum-labs
a556c3ad53
V6-07: the aggregate and chain floors at 8,500 MiB from the measured aggregation peak; the rows and their consequences
...
docs/analysis/floor-memory-profile-2026-10-08.md: the 3060 and 4060 rows on the default, unoverridden job path (the served 0317 host on the V6-07 server: 3060 shard 11.4 s at 7,601 MiB, the whole chain 27.4 s at an 8,306 MiB aggregation peak, both VERIFIED; 4060 shard 16.3 s at 7,504 MiB VERIFIED, the chain aborts at the aggregation's 486 MiB allocation on 176 MiB free), the host's profile and refusal rows (exit 78 under the floor and beside the miner on both cards, the miner unharmed), the recursion constant isolated (no change of the device peak), the known-failed first server, the master-host 0-bytes fault with its NOT RUN rows, and the consequences per card tier. memory_profile.rs: aggregate and chain floors 8,500 MiB (an 8 GB card proves shards and is refused the chain before setup), tests re-pinned; suite 7 passed on box 3.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:20:21 +00:00
igneum-labs
bfa1bfbcfe
family gate: 6.13 names the signing object's column (era edc4fa84) after the hash lane's 22:00 ruling
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:15:26 +00:00
igneum-labs
257b7c310f
Class v6 census: the live-dataset census on the frozen object 21edd23c9 (Phase 1 item (j)): hl-v6-all 4de7b836cc40a4ea and sixteen chain seeds on the chain's dataset with the node1 leaves under header-bound init words, uniform inside every window to the Poisson floor, F8 form 0.999x of the window model; F06's live half LIVE on every register and load; the registry run census-v6-live-20261008-2210
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:12:34 +00:00
igneum-labs
c655ee07bb
test registry: the hash lane's batch 3 (the pairing class: the chain-paired packs, the signing object's P01 reference, the export tool's pairing check)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 21:04:30 +00:00
igneum-labs
2fc7d5c4d7
Merge box/master 36816182c into class-v6-floor-k-docs-2 (the steward's 36816182 under it; the repeat batch trimmed to floor lane 2's rows ADV-05 and ADV-06, method model; the registry master's with that batch recorded)
2026-10-08 21:02:08 +00:00
igneum-labs
9074dbfd93
family gate: section 6.13, the chain-seed draw at 21edd23c9 through the acceptance at both era inputs
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:56:29 +00:00
igneum-labs
b3925d7adb
Merge remote-tracking branch 'box/master' into same-work-harness
...
# Conflicts:
# tools/ci/p01-vectors.py
2026-10-08 20:55:02 +00:00
igneum-labs
1458a5e828
same-work: the two job contexts (01 the frozen object 0x4de7b836cc40a4ea, 02 the chain-seed object 0x2a1d6caab4c24564) as read back on build-1 at 21:49 UK
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:50:37 +00:00
igneum-labs
1d75fdf3c0
Merge master ebdb89163 into pool-registry under the master-landing lock
2026-10-08 20:48:39 +00:00
igneum-labs
716d08820a
shadow-k: the placed gated 64-register window core (9.45 pJ per lane-op ASAP7, k 0.77 at the lock at N3; the board 2.0x node-for-node, 2.45x a node ahead; the window's placed residual +0.23 of k)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 86e5b0fb286e5b0fb24
2026-10-08 20:42:29 +00:00
igneum-labs
816303c5b5
shadow-k: the placed gated 32-register base (6.7 pJ per lane-op ASAP7, the board 2.4x node-for-node and 2.8x a node ahead at the lock), the 8 KB scratch row (104 pJ per read at N3, k 0.074 against the L2 hit)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 07029f7ee86e5b0fb24
2026-10-08 20:42:29 +00:00
igneum-labs
17a0bf5176
shadow-k: the mixed-resource lane's FP32 units routed (6.6 to 7.1 pJ per op ASAP7 as an upper bound, k 0.63 to 0.68 at the lock at N3)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 089546fb186e5b0fb24
2026-10-08 20:42:29 +00:00
igneum-labs
3aa976cd1d
shadow-k: the time-multiplexed core built (8.8 pJ, not the lever), the connected-state variant on the gated window core (6.3 pJ per lane-op, k 0.51 at the lock at N3), the lanes' agreement on the window's cost
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 08447ca9786e5b0fb24
2026-10-08 20:42:28 +00:00
igneum-labs
13617b891c
shadow-k: the adversary's gated rows (4.5 and 6.2 pJ per lane-op ASAP7; the window's residual +0.13 of k), the placed ungated base (11.3 pJ), the ICG simulation model
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 9737ffb2686e5b0fb24
2026-10-08 20:42:28 +00:00
igneum-labs
0dc59c302e
shadow-k: the window's measured GPU side (no spill, 83 and 67 percent occupancy, at most 5 percent per load); the time-multiplexed core, the connected-state draw, the simplified FP32 units and the pod mode
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of 26844fde786e5b0fb24
2026-10-08 20:42:28 +00:00
igneum-labs
eb4bcb98e0
shadow-k: the adversary's 64-register core (the live-state analysis: 95 percent of the window necessary across every wait under the fold-reads-all rule; the state's cheaper forms modelled; the gated rows pending), the gated variants' configs, livestate.py
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only replay of f1ddf0b3986e5b0fb24
2026-10-08 20:42:27 +00:00
igneum-labs
6db5646a9c
registry: pool batch 03 at 7ffd29af with its evidence page docs/analysis/pool/pool-pair-2026-10-08.md (UX-05 vote-key PASS in the crate; the devnet-4 live pair NOT RUN, its wrong_hash an unsynced node's unsettled class v5 state, not a generator skew; the guard on pool-2.0; two 2.0 gaps closed)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:41:26 +00:00
igneum-labs
26417c3ab6
Merge master fbade9172 into class-v6-adversary-docs-3 under the master-landing lock
2026-10-08 20:41:17 +00:00
igneum-labs
0bb8cdb9dd
Class v6 family gate, 6.12: the pairing read back at the final class-v6 sha 21edd23c9 (both pack ids equal to the 38128dff0 read); master merged (the steward's evidence rule: a new evidence file lands free)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:31:48 +00:00
igneum-labs
eb49865421
Merge remote-tracking branch 'build/master' into class-v6-family-gate
2026-10-08 20:31:47 +00:00
igneum-labs
81387ea729
Merge box/master dab456cc8 into class-v6-adversary-docs-3 (rule 26: master's map, page and registry taken; the adversary cells re-added with ADV-05; the batch recorded)
2026-10-08 20:23:20 +00:00
igneum-labs
dab456cc81
Merge p22-stage-2 6407fc80 into master (gate: green on 6407fc80, recorded by tools/ci/pre-push.sh; landed on the box mirror)
2026-10-08 20:21:13 +00:00
igneum-labs
30505a90c4
Merge build/master into class-v6-family-gate: the CI steward's family-gate cell and batch taken as master has them; this landing's batch family-gate-20261008b moves the rows with their evidence; 6.12's subset census at 64 of 64 (0 hot sets, 0 over 1.2x)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:13:17 +00:00
igneum-labs
67d07a267e
Merge box/master 19c945915 into class-v6-adversary-docs-3 (rule 26: master's map, page and registry taken; the lane's cells adversary:mf-placed adversary:d2b re-added; batches re-recorded)
2026-10-08 20:12:11 +00:00
igneum-labs
6407fc80f9
Merge remote-tracking branch 'box/master' into p22-stage-2
...
# Conflicts:
# proving/igneum-prove/host/src/bin/pin.rs
# tools/ci/export-exclude.txt
2026-10-08 20:07:50 +00:00
igneum-labs
f3c24e202f
Merge floor-eco-record-02 c2fc307c into master (gate: green on c2fc307c, recorded by tools/ci/pre-push.sh; landed on the box mirror)
2026-10-08 20:07:38 +00:00
igneum-labs
6898c72dbc
Merge worker-select-pass ed93b542 into master (gate: green on ed93b542, recorded by tools/ci/pre-push.sh; landed on the box mirror under the exception declared by main: main's ruling, 7 Oct 2026 19:5x UK: the GitHub account is suspended, lanes land on the box mirror's master, the box gate stamp is the verdict; GitHub gets the fast-forward when it answers)
2026-10-08 20:07:26 +00:00
igneum-labs
987e0d85c7
Class v6 family gate, 6.12: the final tree (class-v6 38128dff0, the packs' draw read back equal on both ids) through the acceptance under the harness's predicate and the crate's own, which agree on a2af73716's fix (r 0.144 and 0.137 at width 4, 0 exhausted in 6,000 eras, the floors at 0 to 0.5 percent, the fold's half of the bit-R class): PASS; the registry batch at manifest 38128dff0, method native
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:05:56 +00:00
igneum-labs
0614fece04
multi-family adversary: section 17 bound restated on the connected-state class's own shape (the prefix form 8 nJ per hash; the bracket moves at most 0.02x)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Documents-only copy of class-v6-adversary bb8cf8c30 over master 156f3f54e
2026-10-08 20:04:15 +00:00
igneum-labs
b31c6b9ff4
V6-10: pin C held off master on p22-stage-2-pin ( 5144e87c1); proving/igneum-prove/elf/ back to master's served pair for the stages' landing (the 2.0.1 kits embed master's elf/ and their start check refuses another top pair)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:02:11 +00:00
igneum-labs
c2fc307c5d
ECO-05 batch 02 on the RX 7600's class v6 cell (15.79 MH/s measured, the knee 5.4 modelled; the metered knee is batch 03) and the B580's class v6 cell (11.0 measured, watts estimated): one more sustainable world (10R at 0.10, by 1.6 percent on the 7600's entry), 0 of 108 cells pass the envelope, RUN FAIL unchanged; the driver's override hook carried; the cubes and the registry rows recorded at manifest efd03e2c
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 20:00:11 +00:00
igneum-labs
156f3f54eb
Merge hash-registry-1 ac89897c into master (gate: green on ac89897c, recorded by tools/ci/pre-push.sh; landed on the build mirror)
2026-10-08 19:59:34 +00:00
igneum-labs
5144e87c18
V6-10: the repin note with every row read back (three deterministic builds, the id read-back on two boxes, the old and new fee mode compressed proofs VERIFIED under the new pair, the node equality replay, the RPC flag class)
...
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 19:58:48 +00:00
igneum-labs
0d208582a5
V6-07: the host's row from the lesser of the engine's grant and its free reading; prover-floor.md names the free-memory rule
...
The window lane's engine half (reviewb-202 af5ade43
2026-10-08 19:54:26 +00:00
igneum-labs
ed93b5426f
F10 rows, I05 epoch tag, I06 pack geometry rule across the three hosts
...
Rows: docs/analysis/review-2026-10-08-b/f10/rows.md with the pod rows (5090: steady
70.4 MH/s select against 63.7 full, 86.2 against 88.7 J per 2^24 batch; 3090 at the
card's clock 30.1 against 27.6, 184.6 against 197.9 J; the 3090 pod's power cap at
210 MHz named per run; ptxas 92 and 87 registers, no spills) beside the kernel timer,
the emulation and loader logs, the mini's geometry and serve logs. serve-bench.sh reads
a kit pack's seeds from program.h, execs the worker in its coproc (the recorded pid
and the CPU seconds are the worker's).
I05: every worker's found line carries a fifth field, the epoch the hash was computed
under (16 hex): CUDA and OpenCL from the pair, Metal from the job; select-check.sh and
serve-check.sh assert it, the swap case with pack B's epoch. The miner reads four or
more fields.
I06 (gate INT-10): one normative geometry rule. packfile.h reads the word count at
64 bits and refuses a count above 2^32 - 1 and a log2 that is not the floor of log2 of
the count; proto-metal/main.swift gains packGeometry with the same rules and refusal
words, sizes the day's dataset from the exact word count before any Metal work, and
--geometry-check <vectors> [pack dir...]. emu/geometry-vectors.txt is the shared
conformance file (ds55, ds85, ds115, the power-of-two sizes, wrong count or log2,
overflowing counts, missing leaves, leaves on a v2 pack): 19 vectors green in C on
build-7 (packfile-test.sh) and in Swift on the mini, with the two checked-in packs.
Record: cells harness:cuda-select-check (ROT-01, POW-01 partial) and
harness:pack-geometry (POW-02 partial); batch f10-worker-20261008.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-08 19:53:24 +00:00
