10 most recent check-ins occurring on or after 2026-08-11 22:45:56.
More ↑
|
2026-08-13
| ||
| 13:08 | git: gather every interop root into src/git/ — cli.zig joins the zit module (nine importers rerouted), moved roots take parent deps through the module; ~986 duplicate test executions eliminated check-in: 49c08bb191 user: dmestas tags: sub-manifest | |
| 11:41 | submanifest: replay history to pick k (#41) — analyze replays last 200 commits through the real builder+delta encoder (both cost axes, bootstrap split, baseline-charged materialization); enable pins the replayed argmin, snapshot objective as fallback. Advisory only. check-in: b9ceda433f user: dmestas tags: sub-manifest | |
| 03:06 | clone: stop inflating served/received delta containers for their header — prefix-inflate 16B, base frame read header-only; server CPU per 1M-artifact clone 300s+ -> 22s, clone verifies 0 faults check-in: f5eafc6a2f user: dmestas tags: sub-manifest | |
| 01:25 | serve: notice a --force shun/pack generation swap — stat the path beside the held handle, reopen between rounds; shunned bytes stop serving with no restart. Closes #17 check-in: faa8befc3b user: dmestas tags: sub-manifest | |
| 00:16 | serve: scale the write deadline with the reply — 4MB clone batches at the flat 5s cut healthy slow peers mid-write (#22 shape); slow-drain e2e proves it check-in: fa230db9e8 user: dmestas tags: sub-manifest | |
|
2026-08-12
| ||
| 23:21 | timeline: order tied runs by generation, not per-pair ancestry walks — 305k-commit CVS import renders in 5s vs CPU-hours; B1 descendant-first preserved exactly check-in: d823a2811d user: dmestas tags: sub-manifest | |
| 21:47 | clone: byte-budgeted batches (4MB) — 159 rounds vs ~1000 for a 1M-artifact store; warm clones plant the checkpoint (first command 0.06s); print transfer/derive phases. Closes #53 check-in: c32414008b user: dmestas tags: sub-manifest | |
| 19:56 | checkpoint: accept delta manifests; anchor at newest check-in (mainline fold collapsed on real repos); sync: index the pending-delta queue (O(1) hasTarget, O(deps) completeFor) — zit now clones 2GB pkgsrc-2012 from real fossil in 10min, resumes opens in 9s vs 4min derive check-in: c70f48f825 user: dmestas tags: sub-manifest | |
| 09:04 |
checkpoint: retire the tree shape — sidecar only, structurally
The tree shape minted permanent SUBMANIFEST artifacts for benefits already had elsewhere: the rebuild win lives in the prefix record, verify's §6.2 oracle runs buildTree in memory, and the subtree-sharing bytes re-materialize at the first real S-emitting commit at identical cost. Minting early only created unreferenced permanent artifacts and the allow-tree/demotion/gate apparatus needed to keep automation from doing it by accident — all deleted with it. build no longer opens the log for writing or reads the emission settings at all; sidecar layout v3 (old sidecars miss and rebuild in seconds). Proven at scale: SHA-256 of pkgsrc's 593MB store is byte-identical across a full build; cold 4:12, rebuild 3.4s. check-in: 2a8fdcff60 user: dmestas tags: sub-manifest | |
|
2026-08-11
| ||
| 22:45 |
checkpoint: one-pass build, incremental advance, opportunistic refresh
Build paid a bypass derive, a whole-prefix delta scan, and a capture re-derive; a view-free captureRecords (byte-identical to the deriving walk, pinned at every boundary) now does capture and delta detection in one pass, and an existing sidecar's body plus its delta-free attestation reduce a rebuild to the unscanned tail. pkgsrc: cold build 17:58 -> 4:21; rebuild or anchor advance -> 5.1s. Repository.open's healing paths keep an existing checkpoint fresh: witness shape forced (the automatic path can never mint an artifact, whatever §8.1 says — a tree pin advances by demotion and is re-pinned deliberately), gated on 4MB growth AND anchor progress so blob-only growth never churns the sidecar. No checkpoint, no auto-creation: building once is the opt-in. check-in: adae602908 user: dmestas tags: sub-manifest | |