[Perf] (MG_Pipe, MG_Remote/Client): a texture respecify for a CONFIRMED object stops waiting - the row pays its round trip once, at first use - and the frame drops from 130.9 s to 31.3 s

- The predicate is one function again: MGPipeResourceRespecifyWantsItsReply (PipeRoute.h) is read
  by the wire emitter, so the two halves of this row cannot disagree about whether an answer is
  owed. The buffer half never owes one (both call sites discard the Bool, PipeFill.cpp:1105/:1142);
  the texture half owes one exactly until MGPipeHandleIsPublished, and that latch is set only by an
  ACCEPTED create - which EmitResourceRespecify already sends, blocking, when it is clear
  (TextureEmit.h:869-879). So the object pays once, at first use, and every record after it is
  fire-and-forget.
- MEASURED on the device, same frame, same 101,421,396 B staged, stock power policy: waits
  25,154 -> 4,543 and wall 130.9 s -> 31.3 s. The op split is the proof of what changed and what
  did not: waits by_op 3=20611 -> absent (only 2=4536, 5=6, 9=1 remain), while the RECORDS sent
  are unchanged - frame-sent still shows 3=20633 and 103,094 records total. The round trips went
  away; the work did not.
- It also lands on an independent prediction: an audit run beside this work (mgl-device/scratch/
  scheme-review.md) computed 4,543 remaining waits and a 24-37 s credible range (midpoint ~30 s)
  from the archived logs, a host-inproc round trip of 35.4 us and a measured 43.1 MiB/s link.
  Measured: 31.3 s, at the midpoint.
- Cumulative on that frame since this work started: 43,232 waits / 221.5 s -> 4,543 / 31.3 s (7.1x).
  What is left is the create row (4,536 waits, ~23 s) - the confirmation itself, not waste - plus
  hard floors of ~13.4 s client CPU and ~2.2 s of transfer. Going below that means batching the
  create row, which keeps real answers.
- TWO DEBTS, STATED: (1) the invalidation guard for the one silent-loss case is NOT wired; the
  audit confirmed independently that no production path can drop a texture record today, and if
  one appears the notice it needs is protocol-surface (a tenth reverse callback). (2) No SSIM or
  visual check has been taken since these rows stopped asking - frame-sent agreement is strong
  evidence but it is not a picture.
- Two stale controls in older cases were fixed rather than left: the item-B gate now names the
  CREATE row (its old total-based control only moved because a respecify was still waiting), and
  the by-op measurement asserts records-sent for the rows that deliberately stopped waiting.
  RemoteClientTest 169/169; red-once for the predicate is forcing it back to "always wait".
This commit is contained in:
BZLZHH committed 2026-09-25 23:29:08 +08:00
1 parent f2b55b9d2d
commit 82ed1df90e
5 files changed
+119 -29

No files matched your search

+34
View File
@@ -88,6 +88,8 @@
#include "MGPipe.h"
#include "PipeApply.h"
// P12: MGPipeHandleIsPublished, the latch the respecify predicate below reads.
#include "PipeMutation.h"
namespace MobileGL::MG_Pipe {
@@ -271,6 +273,38 @@ namespace MobileGL::MG_Pipe {
MGP_ResourceCreate(&desc, &reply);
return MGPipeTakeReplyBool(reply, "resource_create");
}
// P12: WHICH RESPECIFIES STILL WAIT - one predicate, read by the wire emitter, so the two
// halves of this row cannot disagree about whether an answer is owed.
//
// THE BUFFER HALF NEVER DOES. Both call sites discard its Bool outright
// (PipeFill.cpp:1105 and :1142: `MGPipeRouteResourceRespecify(desc, nullptr);`), so the wait
// bought a value that was thrown away - on the device run this row was the second largest
// source of round trips in one frame, at ~5 ms each.
//
// THE TEXTURE HALF WAITS EXACTLY UNTIL THE OBJECT IS CONFIRMED, and "confirmed" is
// MGPipeHandleIsPublished: the latch is set only when the object's create was ACCEPTED, and
// the applier keeps texture records across a make-current (PipeFill.cpp:3756-3759), so for a
// published object the acceptance this row asks for is a fact the client already holds. The
// confirmation itself is paid once, at first use: EmitResourceRespecify sends a create when
// the latch is clear (TextureEmit.h:869-879), and that create is blocking.
//
// AND WHEN THE LATCH IS CLEAR THIS RETURNS TRUE, which is the whole fallback. It also covers
// the one place the two ends' consumer gates are worded differently - the client asks
// "any bit of the family" (PipeFill.cpp:1577-1595) while the applier asks "bit 10"
// (PipeApply.cpp:1408-1419) - because a create refused by the narrower server gate leaves the
// latch clear, so every later row for that object keeps taking the real answer. The cost of
// that case is throughput, never silence.
inline Bool MGPipeResourceRespecifyWantsItsReply(const MGPResourceDesc& desc) {
#if MOBILEGL_BUILD_DISAGGREGATED
if (MGPipeInstalledArm() != MGPipeRouteArm::kClientWire) return true;
if (static_cast<MGPipeResourceTarget>(desc.Target) == MGPipeResourceTarget::Buffer)
return false;
return !MGPipeHandleIsPublished(MGPipeKind::Texture, desc.Resource);
#else
(void)desc;
return true;
#endif
}
inline Bool MGPipeRouteResourceRespecify(const MGPResourceDesc& desc, const void* initialBytes,
const MGPRespecifiedLevel* level = nullptr) {
return gMGPipeRouteEscapes.ResourceRespecify(&desc, initialBytes, level);