swung0x48
aba329cc53
Two things land together: the plan and contract draft P6 needs before it starts, and the
working notes that until now existed on exactly one machine.
1. P6's plan and contract draft (neither started; not one line of code)
docs/Disaggregated/P6-SPAWN-PLAN.md the package plan
docs/Disaggregated/P6-CONTRACT-DRAFT.md the contract draft (moves to
MG_Remote/CONTRACT-P6.md when c6 lands)
The three findings that mattered, all written into the documents:
a. P6 is SMALLER than its roadmap cell suggests. The transport primitives are already in
the tree and already tested: SocketDoorbell including hangup detection and the two
lost-wakeup fences (Doorbell.cpp:211-360, four cases in FdPassingTest), SCM_RIGHTS fd
passing, ShmSegment::Adopt, and the SurfaceOp / WindowKind schema already defined in
protocol.fbs. inproc uses a real second shm mapping rather than an alias, and
SessionRings.h:22-28 says that was done so today is not the first time it runs. P6 is
assembly plus a control plane plus a process, not "write a socket transport".
b. The evidence for "P6 is only a transport swap" has EXPIRED. It was bought by P5c, whose
audit found 59 direct memory accesses in a tree that already looked clean. P5e changed
the tree again. So the first package, a6, is a read-only audit that writes no code, and
every contract row that depends on it is marked PENDING a6 rather than decided in
advance.
c. P6 does not get a game running on the phone. ANativeWindow* is meaningless in the child,
and protocol.fbs:186-196's own comment says Android transfers the window out of band -
that band is P12. P6 lands pbuffer / surfaceless / offscreen, which is what HeadlessGL,
the split-equivalent lane and trace replay need, and which is exactly what P6's own exit
gate measures. A real window reaching the server is a named refusal. This gets its own
section because "P6 is done" would otherwise read as "the game runs now".
The contract adds rule G: a role may not name a process-local handle. It extends P5c's
rule E (a role may not name the other role's memory) from memory to identity - EGLDisplay /
EGLSurface / EGLContext / ANativeWindow* / fd number / pid / mapped address either cross as
a token this contract defines or they do not cross, and a descriptor crosses only through
SCM_RIGHTS, where the kernel renumbers it.
What P5e left for P6 gets its own section. The latch semantics are already designed
(ARCHITECTURE.md:504) and are copied unchanged. What is new is that after run-ahead the
client no longer waits while the peer process can DIE: the client holds records that were
published and will never be applied, and Android's freezer suspends the child without
closing anything. So slow and dead must be distinguishable WITHOUT a timeout threshold.
That is P6's only new design; everything else is carriage. It shares a predicate with E1's
outstanding re-specification (ID-122), and the two should be designed together.
Two gates carried over from P5e's lessons: G2 requires integration-spawn's test-name set to
match integration-split's name for name, because a differing set means the lane is quietly
running less; and the arm-proof gate requires every entry to record the child pid and
transport=spawn in its OWN private log - ID-124 established that this kind of false green
really happens, and a spawn lane that actually ran monolith would pass every other gate.
S2's shape must be settled before it is written, and the document says why: E1 is the
cautionary case.
2. The working notes move into the repository (docs/Disaggregated/notes/, 415 .md files)
ROADMAP / CURRENT_STAGE_PROGRESS / MEASUREMENTS cite ruling numbers, audit line numbers and
the provenance of gate figures from ~/w7/notes/, a directory that exists on one machine and
is under no version control. A record cited hundreds of times that lives only locally is not
a record, so it moves in whole, with its directory structure intact.
Only .md moves. Scripts, run logs, census and gate result JSON, patches, APKs and nm dumps
stay where they are: they are evidence DATA rather than documents - large, regenerated per
run, and mostly carrying absolute paths and device serials that mean something only on the
machine that produced them. The mapping and this split are written in
docs/Disaggregated/notes/README.md.
Two exceptions are recorded there too: P6's BRIEF has been promoted to a first-class
document and keeps no copy under notes/, so the two cannot drift apart; and recovered/ holds
material salvaged from failed workflow runs, which may duplicate the like-named phase
directories and is kept because a few of those were the only complete version at the time.
CI's citation lint takes docs/Disaggregated/*.md only (non-recursive), and notes/ is
deliberately outside it: those notes' file:line citations were written against their own
code heads, and resolving them against today's HEAD would fail wholesale - which is not an
error, it is their correct state as a historical record. To check one, use the revision it
declares (--rev).
3. Inconsistencies fixed in passing
- README.md's status line still said P5e was in progress while the stage table already said
closed.
- CURRENT_STAGE_PROGRESS's next-step item 1 still listed the exit gate as pending, though
(a) and (b) were run in the previous commit; rewritten to closed plus the one item that
actually remains (E1).
- ROADMAP's P6 row still had the pre-phase plan in three of its cells; it now carries the
package order, the scope boundary and the two new gates.
Checked with check_doc_citations.py: 125 citations across docs/Disaggregated/*.md, 0 problems.
3.3 KiB
P5b sync migration — 2026-09-16
Commit: 08fd9e9ca1888d795654fe133968526411be334c, parent 9606466a.
Worktree: /home/swung/w7/p5b-d1-codex. Integrator-approved contract is CONTRACT-P5B §9.
Result
Five existing opcodes now serve FenceSync, ClientWaitSync, GetSyncStatus, WaitSync and
DeleteSync. The client allocates Fence-kind handles and retains local opaque proxies; the
wire carries only {slot, gen}. The apply thread owns native fences, validates generations,
preserves timeout/failed wait results, and deletes native orphan fences before backend detach.
Frontend orphan cleanup after session shutdown releases only proxies. New slots legitimately
start at generation zero; recycled generations must increase, and stale/destroyed handles
are protocol corruption. No host pointer crosses the wire.
MGPFenceWait is 24 bytes (was 16), with Flags/Pad0 appended; ClientWaitSync retains its flush flag. FenceStatus and FenceWait use their existing reply slots with a 4-byte answer. All five opcodes have matching verb stamps. Field declarations and generated stamp tables agree. The class counts after this commit are A=2/B=53/C=16 (before integrator's named-blit migration).
The frontend FenceSync guard previously returned nullptr for every split session through MGL_BACKEND_SLOT_PTR_LOCAL. This explains why the dynamic d1 census had no FenceSync first blocker despite many trace calls. The guard now reads the class-B emitter pointer; in a pull build this is the exact expression the macro previously produced. A backend with no native fence slot, or a fence creation that returns nullptr, retains GL_Sync.cpp's old monolith always-signaled fallback on the server. Real native wait answers are never replaced.
Validation
- Split build successful, clang Release, maximum -j8.
- Related unit selection: 40/40 (RemoteEmitTable/FieldOwnership/FenceWireRoundTrip at that point). Final FenceWireRoundTrip selection after adding the decline control: 4/4.
- Real GL
DirectGLES.Split.SyncWireScenarioplus private-log gate: 2/2. - The GL scenario asserts all five sync records crossed, waits for GPU completion, validates rendered pixels and leaves one orphan to exercise session teardown.
- Wire controls preserve timeout (rather than silently returning signaled), flush flags and 64-bit timeout; reject stale reused handles; verify native deletion and null-native fallback; missing consumers produce DECLINED with no manufactured answer.
- Representative Minecraft retraces: 4/4. Sodium DirectGLES/DirectVulkan SSIM 0.999986; vanilla in-world both 0.999976. Same values as standalone d1.
- gen_pipe.py --check and gen_pipe_field_ownership.py --check pass; git diff --check passes.
- No full-stage gate or review repeated. G1 and remaining quick lanes belong to the integrator.
Logs: /home/swung/w7/p5b-sync-{unit,unit-final,integration}.log;
trace summary /home/swung/w7/retrace-out/p5b-sync-targets/summary.json.
Integration notes
The four A/B trace targets' immediate missing verb was named framebuffer blit, owned by the integrator; sync completes real fence semantics rather than removing that particular blocker. Cherry-picking over the integrator's blit change may need B/C count arithmetic combined and both new ServerVerbSink declarations retained. r1 was deliberately not merged here. No backend implementation changes, no adb, no push.