| .. |
|
amd-prove
|
docs/analysis/amd-proving.md: no zkVM proves on AMD (SP1, RISC Zero, Jolt, OpenVM, ICICLE cited), the SP1 CPU prover measured on PC 1 beside the miners (282 s a shard at any size, 30 GB RSS: no CPU tier), the tier consequences and the public line; PC 1 job scripts with the bash -n gate; bench-log entry
|
2026-10-05 21:02:59 +00:00 |
|
ca3-pc1-amd
|
Merge commit '711a12c' into ca3-pc1-amd
|
2026-10-06 18:10:29 +00:00 |
|
ca3-reserve
|
Counter ASIC 3.0 items 6 and 7: PC 2 family playbook posts both card key forms from settings.json, confirms the card quiet by the process list, restores cards and prover in a finally block
|
2026-10-06 08:34:28 +00:00 |
|
ca3-shadow
|
Counter ASIC 3.0 item 8: the PC 2 playbook switches the prover back on in its finally block (the clear file's constraint)
|
2026-10-06 08:26:22 +00:00 |
|
ca3-v4
|
Counter ASIC 3.0 gates (hash): follow-up 1 on the program-id fix (ceaccac merged): crate suite 97 of 97, the seven v4 packs re-exported byte-identical (generator 4, id c120d7963abdcd96), fingerprints unchanged on Metal and Apple OpenCL, Mac G2 16 x 1,024 of 1,024 through the v4 class token, the harness 4 of 4 on the class and on the era with the id assert_ne holding; the job scripts take the class token from the pack
|
2026-10-06 16:38:48 +00:00 |
|
ci
|
Ember Tune tiers from the team's table (tiers-table-26, 8 October 2026, main's order: the denominator lane's measured table wires the three tiers before a card's own search runs; the founder's definition unchanged: Ember tunes the card once, three tabs, a tap flips the card to the measured point at once, a class flip re-runs the sweep and re-keys the tabs). app/igneum-app/tiers/class-v5-tiers.json, .mjs and .test.mjs are the lane's files at 7ab204b6 (30 card classes, 5 measured; the lane fills every rentable class by 15:15 BST and the file is taken again then). src/tiertable.rs embeds the table, matches a card by the lane's rule (the longest match string in the name wins, "5070 Ti" over "5070") and gives the three rows in the apply shape with table: true and the class's label (measured or estimated); a class with no lever gives its stock row only; an AMD row's core offset becomes the absolute clock the knob sends once the card's stock clock is known (clock_from_offset). hotplug.rs: apply_pref restores the card's own measured rows when the pref kept them (CardPref.tiers, new, written at the search's end), else the table seeds the tabs; settle_new seeds a card with no pref. state.rs: tiers_table on the card. engine.rs: the search's end clears the table flag and keeps the measured rows in the pref; /api/tune/tier applies a table row as it applies a measured one (an AMD table row through its offset). The window (ui/app.js View.tierSet kind "table"): the buttons carry the table's rate, watts, MH/W and the saving with the word table, the panel says "From the team's table, measured on this card class" or "estimated for this card class" and that Ember measures this card 2 min into steady mining, the lock words read the row in force, the strip's count reads "0 of 2 cards measured, 2 from the team's table", the sentence names the table before the first tune; a row's note is the button's title. Tests known-failed first on build-2 (the UI read a table row as a measured one with no word: expected 'table', actual 'ready'), then green; tiertable.rs tests: the shipped table parses and covers the brief's cards, the longest match, the 5090's three rows with the knee at 1,300 MHz in the apply shape, a no-lever card's one row, an AMD offset converting only once the stock clock is known; the lane's table test joins the gate's app test line. The H100 tier's --sm-sparse auto is the worker's own self-tune from the tuning file the manifest carries (class-v6-floor-sm, proto-cuda/nvrtc/worker.cpp reads sm_sparse per card); the app passes that file through unchanged, so the tier applies the lock and cap and the worker takes the SM shape on its own.
|
2026-10-08 12:30:52 +00:00 |
|
community
|
Discord structure: the build sheet (channel map, moderation, onboarding, bots, events) and the hourly network feed (7 October 2026, 11:4x UK)
|
2026-10-07 12:14:12 +00:00 |
|
dev-fee
|
Dev fee: the test-network measurement in the bench log; run.mjs waits for the nodes and edits the override as text
|
2026-10-04 23:15:27 +00:00 |
|
evm-smoke
|
Execution layer devnet v3: implementation notes, bench entry, viem smoke test
|
2026-10-03 22:05:33 +00:00 |
|
exec-attacks
|
Kill by exact command line or pid file, never by a name: tools/ci/kill-by-name-check.sh in the gate; the 36 pgrep/pkill literals in the tree fixed
|
2026-10-06 22:10:31 +00:00 |
|
exec-sync
|
0.3.16 app side: the forged-record harness; ledger P21 round 4 and X31 (the export-disk class); bench-log verify costs
|
2026-10-06 21:05:19 +00:00 |
|
finality-attacks
|
Merge remote-tracking branch 'origin/master' into ca2-v3
|
2026-10-05 22:34:07 +00:00 |
|
fleet
|
Merge remote-tracking branch 'origin/miner-reliability-21' into release-0.3.21
|
2026-10-07 13:16:12 +00:00 |
|
harness
|
Harness s8 steady state and the bench-log paragraph: RSS per 1,000 blocks before and after the M30 fix
|
2026-10-05 01:49:46 +00:00 |
|
keys
|
Key custody: inventory, encrypted backup and restore, no-secrets CI check
|
2026-10-05 17:01:01 +00:00 |
|
ladder
|
Latency ladder (Horizon finding 4, 0.3.17): docs/design/latency-ladder.md (where N lives, the ladder as a genesis list, the 90 percent seven-window step rule on the class signal's machinery, the verifier bound and its measured table, the X9 chip anchor, the hostile review, consequences per tier); igneum-pow: v4_class_at, v4_rung_reps, v4_counted_ops, generate_from_seed_bytes_program_class_shadow and Epoch::chain_program_shadow (rung 0 is V4_CLASS byte for byte; a rung above carries its shadow size in the id through program_id_class), era draws 8 and 9 consumed and not used, --shadow-reps on the CLI, the known-failed test first (a changed N hashed another program under one id); tools/ladder/verify-bench-remote.sh (the bench per rung on core 40 alone and with core 88 loaded under the box's measure hold; the first run's sibling finished early and is recorded as the script's failed case) with both runs' raw lines under docs/design/latency-ladder-bench; infra/fast-time/latency-ladder.mjs (step, no-step and the known-failed case) and the 60x file's three ladder fields plus the two 0.3.16-lane fields it lacked; ledger M34; the litepaper's hash-class paragraph and the corrected RandomX precedent (Bitmain's Antminer X9, approximate)
|
2026-10-06 22:18:48 +00:00 |
|
launch
|
Income per tier: the three tariffs are USD 0.05, 0.10 and 0.25 a kWh (cheap industrial, US retail, UK retail; the coordinator's answer, 7 October 2026), the table regenerated, the cost per mined IGN shown to three significant figures
|
2026-10-07 08:54:44 +00:00 |
|
lock
|
Kill by exact command line or pid file, never by a name: tools/ci/kill-by-name-check.sh in the gate; the 36 pgrep/pkill literals in the tree fixed
|
2026-10-06 22:10:31 +00:00 |
|
observer
|
Merge site-ui-4: the /live Proven tile on a 600 s summary read, the observer's claims flag
|
2026-10-07 10:14:45 +00:00 |
|
prove-fixtures
|
Proving v1: prover on by default, the aggregated segment record, host modes chain, aggregate and verify-segment, the app's aggregator step, spec 7.8, the fast-time harness and the coverage tool
|
2026-10-05 20:21:04 +00:00 |
|
prover-floor
|
Kill by exact command line or pid file, never by a name: tools/ci/kill-by-name-check.sh in the gate; the 36 pgrep/pkill literals in the tree fixed
|
2026-10-06 22:10:31 +00:00 |
|
proving-v0
|
Packaging for proving v0: the Mac DMG carries igneum-prove-host and igneum-prove-export, the Windows payload carries the Linux host and exporter under wsl2/bin with the WSL2 scripts and the fixtures; the prover probes the shipped WSL2 binaries first and runs helpers by absolute path; push-inputs takes IGNEUM_NODE_SRC; the test script's --network-only mode
|
2026-10-04 14:36:59 +00:00 |
|
proving-v1
|
Merge remote-tracking branch 'origin/master' into release-0.3.21
|
2026-10-07 13:11:18 +00:00 |
|
reliability
|
reliability injector: catch-up after a kept datadir reads the readiness line before the first worker start (the record exists from the first second there)
|
2026-10-07 15:48:32 +00:00 |
|
repo
|
rotation phase 2, executed: old folder stripped to the 0.3.5 manifest, local files renamed, relay on the new key, DL_TOKEN rotated; 7 October swap and the owner's rewrite checklist
|
2026-10-05 15:37:02 +00:00 |
|
repro
|
Repro 0.3.14: the box reproduces itself on all four artefacts (A vs B MATCH), the shipped bytes DIFFER by toolchain; two non-determinisms found and fixed in the check; night battery dry run recorded
|
2026-10-06 19:55:09 +00:00 |
|
scene
|
The shard sentence from the block's facts (7 October 2026, the site's animation said "no shard plan yet" on every planless block): IgneumDag.shardWords(block, nowMs) in scene/live-dag.js (2.0.6) is the one source for the tooltip, the /live inspector and the app's inspector: no shards reads shard plan loading (a chain block under 10 s), excluded, nothing to prove, not yet ordered (pending), off the selected chain, no shards planned, or no shard plan after N s; a planned shard with no prover reads awaiting a prover with the block's age, assigned to <prover> when one holds it, proving, verified, paid as before; the app's inspector prints these per shard and as the count line; the copies synced; tools/scene/shard-words.test.mjs on the gate's scene line, known-failed first on 2.0.5
|
2026-10-07 18:13:23 +00:00 |
|
site-redesign
|
site-redesign: the package's look 1:1, header and footer included
|
2026-10-07 10:57:32 +00:00 |
|
txgen
|
EVM transaction relay: design doc 1.4 "Relay", rows 463 and 9 closed, bench-log entry, 3-node relay harness
|
2026-10-05 17:53:07 +00:00 |
|
ui-mock
|
Merge branch 'earnings-26' into app-ia-26
|
2026-10-08 13:45:10 +00:00 |
|
ui-ota
|
ui-ota publish: the dl token follows packaged-config.sh's .next rule (IGNEUM_DL_TOKEN_FILE, else dl-token.next while a rotation runs, else dl-token), so a bundle staged during the rotation lands in the new folder and --verify reads the new URL; tokenFile and readToken exported, token.test.mjs on the gate, known-failed first on the direct read (7 October 2026, the dl token rotates with 0.3.22)
|
2026-10-07 19:20:28 +00:00 |
|
upstream
|
Provenance: credit every borrowed component, upstream merge procedure, prover customer brief
|
2026-10-03 20:02:16 +00:00 |
|
windows
|
Per-monitor DPI awareness for the Windows host (dpi-23, 7 October 2026, PC 2's 5120 by 2160 at 200 percent rendered stretched and blurry): host.manifest declares PerMonitorV2 (and true/pm), embedded by host.rc as RT_MANIFEST with the link at /MANIFEST:NO; SetProcessDpiAwarenessContext at start as the runtime path for a Windows that ignores the manifest, SetProcessDPIAware as the last fallback; WM_DPICHANGED takes the suggested rectangle so WebView2 re-renders at the new monitor's scale; the opening-size rule now runs in logical pixels (opening_size.h igneum_opening_size_scaled: the primary monitor's physical work area and its DPI from GetDpiForMonitor, looked up at run time, LOGPIXELSX as the fallback), so the 1440 by 900 and 1920 caps mean CSS pixels as on the Mac (PC 2 opens 1625 by 1016 logical, 3250 by 2032 physical, sharp); the remembered frame is logical on both hosts (the unaware host read virtualised coordinates), kept as it is, with a WindowUnits marker from here; six scaled cases in opening_size_test.c (the test could not compile before the function existed); tools/windows/dpi-check.ps1 is the read-only PC-job check (awareness, rectangle, monitor DPI, logical size; -Expect PerMonitorV2 fails on the 0.3.22 host, which reads Unaware); host.cpp touched
|
2026-10-07 19:28:34 +00:00 |
|
workers
|
Worker dashboard: only a build-* slot held without a line shows as a "held, no holder line" card (the measure lock is not a build)
|
2026-10-07 11:16:17 +00:00 |
|
build-job.mjs
|
Merge commit '6478040' into release-0.3.17
|
2026-10-06 23:10:56 +00:00 |
|
build-remote.sh
|
Build boxes: the class router is a preference with spill-over (a held or overloaded box hands the job to the other one); build-2 gets a third slot and every run there is bounded on its own 32-core band; the spill decision in the first route line, the slot label and the JSONL row
|
2026-10-07 13:08:19 +00:00 |
|
console.mjs
|
Merge commit '6478040' into release-0.3.17
|
2026-10-06 23:10:56 +00:00 |
|
cross-remote.sh
|
Build tools: whole-body blocks (the edited-while-running class, with its check), the igneum-pow pairing line, move-hand.sh restart with its readbacks
|
2026-10-07 03:12:51 +00:00 |
|
heat-gate.mjs
|
ember-heat: the gate reader, the PC 1 four-hour runbook, the plan and the light and dark captures
|
2026-10-07 10:50:23 +00:00 |
|
jobs.mjs
|
jobs: one signed object (igneum-jobs.signed.json) so a file and a signature from two deployments can never pair
|
2026-10-05 15:52:34 +00:00 |
|
ledger-page.mjs
|
site-redesign: the package's look 1:1, header and footer included
|
2026-10-07 10:57:32 +00:00 |
|
logs.mjs
|
rotation phase 2, executed: old folder stripped to the 0.3.5 manifest, local files renamed, relay on the new key, DL_TOKEN rotated; 7 October swap and the owner's rewrite checklist
|
2026-10-05 15:37:02 +00:00 |
|
relay.mjs
|
relay: three auth tiers, signed run tasks, machine secrets, retention; clients on headers; TZ=UTC and curl -K checks (X23 X24 X25 X26 X27 X28 X29 G13 G14)
|
2026-10-05 18:46:13 +00:00 |
|
ship-app.mjs
|
Merge commit '6478040' into release-0.3.17
|
2026-10-06 23:10:56 +00:00 |
|
site-serve.mjs
|
Explorer pages: /explorer, /block/<hash|number>, /address/<addr>, the search router, a local preview server, CI
|
2026-10-05 19:35:40 +00:00 |
|
tuning.mjs
|
Ember Tune: every card tuned for MH per watt out of the box, the fleet prior per card model in the signed manifest, the console and /miners priors table
|
2026-10-05 21:25:09 +00:00 |
|
workers-remote.sh
|
Build tools: whole-body blocks (the edited-while-running class, with its check), the igneum-pow pairing line, move-hand.sh restart with its readbacks
|
2026-10-07 03:12:51 +00:00 |
