vwemirror/BT411
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

#149: measure the filed suspect -- the #141 segment sweep is EXONERATED; ship [segperf] telemetry

Browse Source 
A/B on identical 2-node beam-heavy runs (scratchpad/night14/segperf.sh),
BT_BEAM_SEGFRESH=1 (swept behaviour) vs =0 (pre-sweep compose at the beam
site, the only per-frame swept call):

                    fresh(swept)   legacy(pre-sweep)
  rstat blocks>50ms       0              0
  maxDraw worst         730ms          646ms      (mission-load spike, BOTH)
  segperf dirty-passes   38/s            4/s      <- the sweep DOES multiply
  segperf accessor ms   0.97/s          0.57/s    <- ...by 0.4 ms/s.  Noise.

So the invalidation-storm hypothesis I filed on #149 is wrong by three orders
of magnitude, and Oracle's sustained 50-104ms stall window does NOT reproduce
on this rig at all.  Refusing to guess a third time: the build now carries the
telemetry to answer it on the machine that actually regresses --

  [segperf] calls= dirty= ms=   printed beside every [rstat] window under
  BT_PERF_LOG (JMOVER counters; two integer increments when unset), and
  BT_BEAM_SEGFRESH=0 remains as a one-env A/B for the beam site.

Default stays FRESH (the swept accessor): its measured cost is trivial and it
is the correctness-cautious side while the peer-beam-staleness question is
unmeasured.

Next for #149: Oracle runs one session with BT_PERF_LOG=1.  If [segperf] ms is
large inside his stalled windows, segment work is implicated on HIS
configuration and BT_BEAM_SEGFRESH=0 gives the immediate A/B; if it is small
(as here), the stall is elsewhere in the 817->857 delta and we hunt with his
numbers instead of my theories.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018SgmXGNMXavXiafKXf9MDC
This commit is contained in:
Joe DiPrima
2026-08-09 16:46:10 -05:00
co-authored by Claude Opus 5
parent 84f8b1c415
commit a77c5a1407
4 changed files with 122 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
engine/MUNGA_L4/L4VIDEO.cpp
+14
View File
@@ -9034,6 +9034,20 @@ void DPLRenderer::ExecuteImplementation(RendererComplexity, RendererOrigin::Inte
DEBUG_STREAM << "[rstat] frames=" << sFrames << " avg=" << (sAcc / sFrames)
<< "ms maxDraw=" << sMaxD << " maxPresent=" << sMaxP
<< " batches=" << gNumBatches << " culled=" << gBTNumCulled << "\n" << std::flush;
// #149: segment-refresh telemetry on the same cadence (BT_PERF_LOG).
// calls = GetSegmentToWorld entries; dirty = the mark-every-segment
// invalidation passes (the expensive arm the #141 sweep may have
// multiplied); ms = time inside the accessor for the whole window.
{
static const int sSegPerf = getenv("BT_PERF_LOG") ? 1 : 0;
extern int gBTSegWCalls, gBTSegWDirty;
extern double gBTSegWMs;
if (sSegPerf)
DEBUG_STREAM << "[segperf] calls=" << gBTSegWCalls
<< " dirty=" << gBTSegWDirty
<< " ms=" << gBTSegWMs << "\n" << std::flush;
gBTSegWCalls = 0; gBTSegWDirty = 0; gBTSegWMs = 0.0;
}
sAcc = 0.0; sFrames = 0; sMaxD = 0.0; sMaxP = 0.0;
}
}