Repository navigation
HermitCrab engine: share frozen syntactic feature structures on internal Word clones (about 15% of a heavy Sena profile, never isolated) #595
Description
Internal clones of a frozen Word deep-copy its SyntacticFeatureStruct, then freeze and hash it again when building analysis keys. The change adds Word.CloneForEngine, which shares the frozen structure until a rule must mutate it. The three in-place analysis mutation sites call EnsureOwnSyntacticFeatureStruct first. A per-Morpher setting (HC_SHARE_SYNTACTIC_FS=false opts out) controls sharing. Public Word.Clone() stays a deep, mutable clone. This is sharing, not a cache. CloneForEngine and EnsureOwnSyntacticFeatureStruct are on the archived branch only.
Measured. The heaviest Sena profile attributed roughly 15% of CPU to freezing and hashing a newly cloned syntactic feature structure while building analysis keys. The five-language measurement is branch-wide, so no standalone speedup is assigned. The wall-clock figure is a profile share and is not attributed.
What a revival must do first. Time it with the toggle alone, in clones and freezes avoided. Do not use the old morph-count signature for parity. Sharing is unavailable for mutable source words.
Record: see the archived write-up below
Ledger: #490 (comment); disposition: #490 (comment)
From #490 (closed).
Activity
johnml1135 commentedon Nov 7, 2026
Archived from
docs/hc-optimization-ledger@ 2eba2f5 (docs/optimizations/syntactic-feature-structure-sharing.md), which is being deleted:Copy-on-write syntactic feature structures in engine clones
Status: retained on
perf/hc-optimization-archive, not merged (Word.CloneForEngine,EnsureOwnSyntacticFeatureStruct). Portable: partly; immutable sharing is general, while the ownershipplumbing is specific to this engine.
Invariant
HermitCrab freezes a
Wordbefore it becomes an input to the analysis or synthesis graph. Internal clones maytherefore share that frozen word's syntactic feature structure until a rule needs to mutate it. The public
Word.Clone()contract remains a deep, mutable clone and never exposes copy-on-write sharing to callers.Implementation
Internal engine sites use
CloneForEngine(). A per-Morphersetting is copied onto new words and controlswhether those clones share a frozen syntactic feature structure. The three in-place analysis mutation sites
call
EnsureOwnSyntacticFeatureStruct()before modification; assignments of a fresh structure need no guard.The opt-out is internal and captured per
Morpher, preventing one parser or test from changing anotherexisting parser's behavior.
Evidence
The heaviest Sena profile attributed roughly 15% of CPU to freezing and hashing a newly cloned syntactic
feature structure while constructing analysis keys. Sharing removes that repeated deep clone/freeze on the
engine path. The final five-language measurement is branch-wide rather than an isolated toggle result, so no
standalone speedup is assigned here.
Correctness coverage and limits
WordSyntacticFsSharingTestscovers copy-on-write, nested/shared structures, public-clone mutability, andisolation between morphers. The broader HermitCrab suite exercises integration behavior and passed both with
sharing enabled and with
HC_SHARE_SYNTACTIC_FS=false; the full conformance-inclusive run passed 603 testswith one expected host skip. Sharing is intentionally unavailable for mutable source words.