jan/rrt
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Rule out chairman and queue neighbors in 319 lane

Browse source
This commit is contained in:
Jan Petykiewicz 2026-04-19 13:58:13 -07:00
commit 43271e4d3d
3 changed files with 25 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

13
docs/control-loop-atlas/post-load-generation-paintterrain-and-save-load-restore.md
View file

@ -798,6 +798,19 @@ The same brush strip is tighter now too:
work, not the missing later republisher for `[region+0x2a4]` or
`[region+0x310/+0x338/+0x360]`.
The
remaining `319` neighbors are ruled down in the same negative way now too. The post-`319` call
`0x004377a0` stays on chairman-slot/profile materialization: it normalizes the `16` slot bundles
rooted at `[world+0x69da]`, republishes selector bytes into `[0x006cec7c+0x87]`, populates the
live chairman-profile collection `0x006ceb9c`, and clears selected company/chairman bytes
`[world+0x21/+0x25]` before the optional `0x0047d440` news follow-on. Its neighboring queue
strip is explicit too: `0x004348e0` only gates whether the transient list at `[world+0x66a6]`
should run, `0x00437c00` is a typed dispatcher over queue-node kind byte `[node+0x08]`, and the
later `0x0044d410` calls are world-rect/grid refresh work. Those helpers therefore stay outside
the missing region restore seam, leaving the placed-structure replay strip
`0x004133b0 -> 0x0040ee10 -> 0x00480710` as the meaningful `319` bridge and pushing the
unresolved region handoff farther into the later post-load continuation above it rather than the
chairman/queue-side neighbors.
The
region-side setup branches are narrower in the same negative way now too: `0x00421c20` is just
the collection dispatcher, and its worker `0x004235c0` stays inside live region demand and
placement by routing through `0x00422900` cached category accumulation, `0x004234e0` projected

11
docs/rehost-queue.md
View file

@ -1596,6 +1596,17 @@ Working rule:
`[region+0x27a/+0x27e/+0x282/+0x286]`. So the next non-hook region work should start from the
post-load `319` placed-structure replay seam and only then revisit the narrower region-side
`320/321` branches if the exact field bridge is still missing.
- The `319` adjacency is tighter in another negative way now too. Direct disassembly of
`0x004377a0` shows the post-`319` call staying on chairman-slot/profile materialization: it
normalizes the `16` slot bundles at `[world+0x69da..]`, republishes selector bytes into
`[0x006cec7c+0x87]`, populates the live chairman-profile collection `0x006ceb9c`, and clears the
selected company/chairman bytes `[world+0x21/+0x25]` before the optional `0x0047d440` news
follow-on. The same trace rules down the companion queue strip too: `0x004348e0` is only the
gate for the transient list at `[world+0x66a6]`, `0x00437c00` is its typed dispatcher over
queue-node kind byte `[node+0x08]`, and the later `0x0044d410` calls are world-rect/grid refresh
work. None of those helpers republish `[region+0x2a4]` or `[region+0x310/+0x338/+0x360]`, so the
remaining region search stays centered on `0x004133b0 -> 0x0040ee10 -> 0x00480710` and the later
post-load continuation above it rather than the chairman/queue-side `319` neighbors.
- The `319` plateau is split more cleanly now too: its neighboring world helpers are no longer
plausible region-body owners. `0x00437220` and `0x004377a0` are the chairman-slot selector and
profile materialization family over `[world+0x69d8]`, `[world+0x69db]`, and scenario selector