ANSFR: purge stale ServerOptimised zip when engine rebuilds from ONNX

A ServerOptimised zip is a per-GPU cache of a previous engine build. When
the engine profile changes (batch gate, TRT/CUDA upgrade), the cached
engine inside it goes stale: every re-extract (e.g. after Temp cleanup)
restores the old engine and costs a full ONNX rebuild, forever, because
the server only regenerates the zip when the file is missing.

Engine now records rebuiltFromOnnx() + engineFilePath() in
buildLoadNetwork (set only on the generate-from-ONNX branch; cache hits
and alternative-precision loads don't raise it). After a rebuild,
ANSFaceRecognizer::Initialize deletes the stale same-GPU engine variants
from the model folder and the supplied optimised zip, so the server's
next optimize pass recreates it with the fresh engine.

Safety: only zips whose parent folder is ServerOptimised/-ized qualify —
base model zips can never match. Engine purge is scoped to the
"<model>.engine.<GPU>." prefix, leaving other GPUs' engines intact on
multi-GPU machines. All deletes are best-effort (error_code); multi-GPU
pool paths never set the flag, so the purge is a no-op there.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
2026-08-21 11:21:48 +10:00
co-authored by Claude Opus 4.8
parent da9d922498
commit 799f755d5a
3 changed files with 73 additions and 0 deletions
+10
View File
@@ -180,6 +180,14 @@ public:
[[nodiscard]] const std::vector<nvinfer1::Dims> &getOutputDims() const override { return m_outputDims; };
[[nodiscard]] const ANSCENTER::Options& getOptions() const { return m_options; }
/// True when the last buildLoadNetwork() had to build the engine from ONNX
/// because no cached engine matched the current profile (GPU / precision /
/// batch / TRT version). Callers use this to invalidate stale caches that
/// carried the old engine (e.g. ServerOptimised model zips).
[[nodiscard]] bool rebuiltFromOnnx() const { return m_rebuiltFromOnnx; }
/// Full path of the engine file resolved by the last buildLoadNetwork().
[[nodiscard]] const std::string& engineFilePath() const { return m_engineFilePath; }
/// Return the CUDA device index that preprocessing should target.
/// In pool mode this returns the device of the first alive slot;
/// in single-engine mode it returns m_options.deviceIndex.
@@ -393,6 +401,8 @@ private:
bool m_usingCachedEngine = false;
bool m_skipEngineCache = false; // true = bypass TRTEngineCache (for OptimizeModelStr)
bool m_skipOnnxRebuild = false; // true = don't delete/rebuild corrupt engine files (elastic growth)
bool m_rebuiltFromOnnx = false; // last buildLoadNetwork built from ONNX (no matching cached engine)
std::string m_engineFilePath; // engine file resolved by last buildLoadNetwork
bool m_skipOnnxBuild = false; // true = skip ONNX→TRT build if no cached engine (demand growth)
bool m_lastLoadFailedVRAM = false; // set by loadNetwork when failure is due to insufficient VRAM
bool m_forceNoPool = false; // true = force single-GPU path, skip multi-GPU pool