- The round-2 report guessed the apply thread's ~48% mutex time was the backend
pendingMutex; the symbolized call graph at 56a77348 says it is the idle poll in
ServerLoop::ApplyThreadMain taking m_controlMutex twice per iteration
(ControlIsPending in the ready lambda, PumpControlRequest), with DrainRing at
29% of the thread. Every package below is cut from that graph, not from self%.
- T (transport): the control mailbox gets an atomic shadow bit m_controlPosted so
the idle poll takes no lock (handshake, C2 critical section and the
NOT_INITIALIZED answer unchanged); Doorbell::Wait spins on a once-calibrated
iteration budget and reads no clock in steady state (the park phase alone
computes deadlines); client/server wait and park counts ride the stats line
as wait[srv= srvpark= cli= clipark=] and the wire ledger as cliwait=/clipark=.
- D (role guards): OnApplyThread() is inline - one relaxed atomic key against the
caller's thread pointer (aarch64 __builtin_thread_pointer, else
std::this_thread::get_id) - and m_applyThreadId, the function-static guard and
the pthread_equal call are gone; the three scope depth counters stop being
thread_local (their only reader is the apply-thread guard, so they count only
when OnApplyThread() is true; a GL-thread scope no longer exempts the apply
thread, pinned by a new case); the guard chains are ordered cheapest-first.
- B (persistent maps): the shadow allocator rounds the SIZE to a page multiple, so
TrackWriteMap aligns OUTWARD when the shadow is page-granular - a sub-page map
is one protected page, edges and their per-draw XXH3 disappear; the inward arm
stays as the fallback. Review found a hole the edge hashes had hidden since
round 1: a map that faulted but was not consumed by the walk that advanced the
epoch was never pushed. The walk now drains every marked member before the
binding walk, registration bumps the epoch so a fresh map ships once on the
draw path, and a non-4 KB kernel page refuses the mprotect arm by name.
- C (client per verb): an image-unit high-water mark (push build only, fed by
glBindImageTexture) replaces the 192-unit sweep per draw; the residual fill's
7-term supplied conjunction is a memo keyed on its four inputs; the server's
63-field stamp per verb is a constexpr per-class mask table.
- generate_mipmap no longer skips its wait under MOBILEGL_IPC_BATCH_WAITS: its sink
is the backend GenerateMipmap, which reads MGB_CTX->GetActiveTextureUnit(), and
a batched publish let the next validate move the serial under it -
Fatal{UnmigratedPipeInput, "GetActiveTextureUnit@ReadPixels"} on the gate tree.
The rule is written at the site: a record may skip its wait only if its apply
reads nothing the residual fill writes. The texture-death scenario gets the
framebuffer case's fence before it reads the object_death tally.
- Redmi 2f7cbe2e, FCL + Minecraft 26.3-rc-3, VD12, CPU pinned: inproc 64.3 fps
(client 15.0 ms/frame, apply 12.4) -> 103-114 fps (client 8.7-9.2 ms/frame on a
mid core, apply 7.0-7.4); monolith 115 (120 Hz vsync-capped runs) / 206 (one
uncapped run, 5.0 ms/frame on the big core). Gates: unit 2203/2203,
integration-split 111/111 (the two formerly flaky cases 6/6 on repeat),
generators, include closure, dirty surface and doc citations green.
VD12 Minecraft on the Redmi (FCL, DirectGLES): the split's in-world frame rate
13 -> 60-66 fps against monolith's 110-114, and persistent-map wire bytes
25 MB/frame -> ~0.23 MB/frame. The residual split-specific costs are named at
the end, not hidden: the R-1 serialization (WaitForApplied ~12%, P11's
versioned-inputs design) and a split-only mutex complex still under
investigation.
Persistent-map push, three arms picked per buffer at map time:
- mprotect dirty tracker: the mapped range's page-aligned interior is
protected PROT_READ; the app's next write faults, the SIGSEGV handler marks
the page in the slot's bitmap and unprotects, and the push ships one 64 KB
block per block containing a faulted page. Ownership is three tests -
SEGV_ACCERR only (a userspace raise and an unmapped page chain verbatim),
inside a live tracked range, and not a refault on a still-set bit (a write
we unprotected succeeds, so that is a non-write access and it crashes
honestly). The alignment is INWARD: an outward-aligned edge page holds
foreign bytes, and a signal-blocking worker thread faulting on one is a
process kill - that was the ctest-only SmallRing segfault. Edge spans
(at most two per map) are hashed per push and ship their containing block
only on change (unconditional edges measured 112 MB/frame). Re-arm is one
mprotect per contiguous run. The first push ships everything once.
- hash-suppress arm (MOBILEGL_IPC_PERSISTENT_HASH_SUPPRESS, default 1):
per-block XXH3 against the last push's hashes; untracked buffers only.
- PushDrawConsumers: only the buffers the upcoming draw/dispatch can read
(VAO attribs/element, indexed binding points, indirect/parameter slots,
TexBuffer backing - the last also retires the per-frame-cadence cloud
twitch), plus the fault-epoch skip: an unmoved epoch proves no interior
byte changed, tracked edges are hashed in place, and any hash-arm member
vetoes. The membership map caches the tracker slot pointer (a 64-slot
rescan per buffer per draw was measured and removed).
MOBILEGL_IPC_BATCH_WAITS (default 1; MOBILEGL_PIPE_VERIFY forces 0): a
value-class record (kCtxState/kCtxCso/kCtxObject with no reply slot) is
published without waiting for its own apply; the barrier is taken at the next
pull-reading verb, the only place BARRIER-PULLED fields are read, so the R-1
fence posts are untouched and only the round trips between them change. Two
consequences are pinned rather than argued: kCtxObject verbs the tests read
synchronously get explicit fences (GenerateMipmapFieldsCross,
FramebufferDeathCrosses), and a frontend object whose last SharedPtr drops on
the apply thread - impossible before batching - now enqueues its
(kind, lifetimeId) on the deferred destroy queue and the GL thread replays
the DestroyAndFree helper at the verb hooks (CopyImageLayeredScenario aborted
Fatal{RoleViolation, MGPipeSlots} without it).
Doorbell::Wait's spin reads the clock once per 64 yields: on the workload
device a steady_clock read is a real syscall (no vDSO) and the per-iteration
reads were ~20% of the whole process at ~1000 waits/frame.
Gates: gen_pipe --check/--self-test green; unit 2188/2188;
integration-split 110/111, the one red being GenerateMipmapDepthPixels whose
DirectGLES and DirectVulkan MONOLITH siblings flake identically with every
code path of this package provably inert (PushIsArmed false, no ClientSession,
deferral's monolith early-out).
Layer 1 arms the texture object surfaces at MipmapStorage (the one point every concrete
texture class and view forwards through): AllocateStorage, TruncateMipmapLevels,
UpdateMipmapSubData, MapMipmapData, MarkStorageDirty(Region), IsStorageDirty,
GetStorageDirtyRegion(s) take Fatal{RoleViolation, texture-legacy-arm} from the apply
thread with an active transport. What the lane then exposed is fixed, not exempted: the
completeness gate answers from the staged store's defined-ness, the two logging reads go
through the descriptor macro, and the remint replay loop keeps its named
Fatal{UnmigratedEmulation} ahead of the guard. Magma's texture sync read path is the one
named exemption (MGPipeTextureLegacyArmScope, P7). Layer 2 wires InBarrierWait as the
client half of the gPipeInputs single-writer rule (RefusePipeInputsTouchWhileApplierOwnsIt:
applier owns it AND this thread not in a barrier wait => Fatal{RoleViolation,
gPipeInputs}; the barrier-off negative control is deferred to its own Fatal), and stale
reads of RECORD-SUPPLIED / APPLIER-DERIVED fields under a server stamp now report
RoleViolation with the field named. The RemoteGuards suite pins all nine texture surfaces
plus HasDefinedContent by name, and a negative control for the fill window. CI gains the
integration-split-strict job: unit lane as the hard gate under MOBILEGL_IPC_STRICT_ERRORS=1,
the scenario lane expected-red with the marker asserted (rewritable to plain green once P7
retires the object-class rows). rsp is per-frame in the stats line and measured on the
locally available traces (bsl in-world 948.5/frame avg over 123 frames, complementary
1591.9, iterationrp 286 until its named remint-pull Fatal); value-class pulls are zero by
rv's pinned list, what remains is the object-class fifteen.
Evidence: unit 2187/2187 (also under STRICT_ERRORS=1); integration-split 111/111; red-once
per layer (the texture guard and the gPipeInputs guard each turned their named death test
red and restored green). CONTRACT-P5C sections 5.4, 6.
set_context_values (opcode 79, MGPContextValues 96B, kCtxState) carries the active texture
unit, the max touched unit, the fifteen touched-binding-point counts and the five value-class
XFB rows, emitted at validate under the subsystem bit and hash-suppressed as a whole; the
emit gate and the residual-skip gate read the same answer so they can never disagree.
MGPipeApplySetContextValues writes gPipeInputs (the PackState shape). MGPAttribValue grows
24->56B to carry the frontend's converted three views, and MGPipeApplySetVertexAttribDefaults
finally writes them (the 'ignores MGPAttribValue::ValueClass' warning dies), so
GetCurrentVertexAttribute leaves the refusal set. The three texture shutters answer the
applier's own serials under a server-stamped verb (APPLIER_DERIVED), storage answer kept
verbatim elsewhere. FieldOwnership.def: nine value-class rows become derived RECORD_SUPPLIED,
the three shutters APPLIER_DERIVED, the nine object-class rows keep their phases;
FieldOwnershipTest pins that the remaining pulls are exactly the object-class list.
gen_pipe.py gains SCAN_EXEMPT_ACCESSORS with per-name reasons and retiring phases (it repairs
the --check gate the hd merge's two G6 probes had left red).
Evidence: unit 2175/2175; integration-split 111/111; both generators' --check and
--self-test green; rsp on the Xfb capture scenario 36 -> 22; red-once - gating the emission
off turns seven named scenarios red, restoring returns 111/111. CONTRACT-P5C section 5.3.
A split-only verb-handle workspace in MGPipeApplierState (written by the sinks before
dispatch, read by the backend within the same verb) carries the record's handles to the
backend entries whose shared function-table signatures cannot grow them without moving
G1. OnBlit resolves ReadFbo/DrawFbo through the FBO twin table by handle (the named arm
of BlitFramebuffer, a loud refusal when the backend does not consume it) and the split
path's ScopedBlitBindings rebinding is deleted; CopyTex resolves MGPCopyFromFramebuffer::Dst;
the mip descriptor check resolves MGPMipPlan::Res; the indirect families resolve their
buffer handles from the records and read the bytes from the server staged shadow with
RequireStagedCoverage. EnsureBufferResourceForHandle answers HasDefinedContent from
Desc.HasDefinedContent OR the staged coverage set (the streaming idiom defines content
through subdata with no new descriptor). The minting GetOrCreate(StatePtr), HandleOf and
the death switch's lifetime-id arms take Fatal{RoleViolation, MGPipeSlots} from the apply
thread, as do the three MGPipeSlots() entry points; the frontend BufferObject legacy
accessors take Fatal{RoleViolation, buffer-legacy-arm}; Magma's four caps reads go to the
server's own backend and the client-mirror fallback arm is a named refusal.
Evidence: unit 2156/2156 (incl. 9 new RemoteF1 + 10 new RemoteGuards death tests pinning
the four Fatal names); monolith smoke green; six production mutations each turned their
named test red and were restored; c1f_redcheck M5 RED then RESTORED GREEN. The mip
descriptor check overlaps tx and is coordinated at the merge (tx's sync rekey wins the
sync body, this handle resolution wins the identity check). CONTRACT-P5C section 3.
swung0x48