The lone js opt-in-JIT residual was async/await_in_loop, which failed to PARSE
under JIT ("Unexpected token: op '<'" on `i < 5`) while passing on CEK. The js
exclusion was "js-*", but the recursive-descent parser is the jp-* namespace
(75 functions in lib/js/parser.sx) — only the lexer/transpile/runtime are js-*.
So the parser was left JIT-eligible and a jp-* function miscompiled this
construct (the long-standing parser-miscompile class).
Fix: extend the js exclusion to "js-* jp-*" so the parser is interpret-only too,
matching how every other guest's front-end is handled. js conformance under
SX_SERVING_JIT=1 is now 148/148, == CEK.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The 6 common-lisp opt-in-JIT failures were all condition-system continuation
escape: cl-restart-case/cl-handler-case/cl-handler-bind wrap their body in
call/cc (restarts + non-local handler exit). When an SX function that drives
the condition system (the parse-recover / interactive-debugger fixtures, e.g.
parse-numbers, make-policy-debugger) is JIT-compiled, the call/cc form runs in
a NESTED cek-run where invoking the captured continuation
runs-to-completion-and-returns instead of escaping — so a restart fails to
abort and the body falls through. Observed as result accumulation
(got (1 3 0 3) vs (1 3)) and no-abort (restart returns the 999 sentinel).
These callers are arbitrary user/fixture code, not a fixed namespace, so they
can't be prefix-excluded. New data-driven mechanism:
- jit-exclude-callers-of! registers call/cc-establishing form names in
Sx_types.jit_excluded_caller_names.
- jit_compile_lambda skips any function whose constant pool (recursively,
incl. nested closures) references a registered name — code_refs_escaping_caller.
Guarded by Hashtbl.length > 0 so it's a no-op for every guest that doesn't
register (zero effect outside CL).
- lib/common-lisp/runtime.sx registers the establish side (cl-restart-case,
cl-handler-case, cl-handler-bind) and the invoke side (cl-invoke-restart,
cl-invoke-debugger, cl-signal, cl-error-with-debugger).
Result: CL conformance under SX_SERVING_JIT=1 = 487/0, EXACTLY matching the CEK
baseline (was 484/6 with a +3 double-execution over-count). parse-recover
3/4 -> 6/0, interactive-debugger 7/2 -> 7/0.
Note: the geometry/mop-trace suites report 0/0 on BOTH CEK and JIT — they error
"Undefined symbol: refl-class-chain-depth-with" (the CLOS suites don't preload
lib/guest/reflective/class-chain.sx). Pre-existing conformance-harness gap, not
a JIT issue; left as-is.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Enabling the epoch serving-mode JIT globally regressed continuation-based guest
interpreters (the epoch mode is the shared command channel every loop's
conformance runner uses). Two-part fix:
1. SAFE DEFAULT GATE. register_jit_hook in the persistent server branch is now
opt-in via SX_SERVING_JIT=1 (default OFF). Default behaviour is unchanged
(no JIT in epoch serving) → zero regression for sibling loops. The
content/Smalltalk page server opts in.
2. GENERAL FIXES + per-guest interpret-only declarations:
- callable? (sx_server/run_tests/integration_tests/mcp_tree) now accepts
VmClosure. A JIT-compiled higher-order function returns its inner closure
as a VmClosure; callable? previously rejected it, so scheme-apply's
(callable? proc) guard failed with "not a procedure: <vm:anon>".
- jit-exclude! gains a trailing-"*" namespace-prefix form
(Sx_types.jit_excluded_prefixes), the robust way to mark a whole guest
interpreter interpret-only (a name-list misses functions in extra files —
it left erlang's vm/dispatcher JIT'd and 13 tests short).
- Per-guest exclusions in each guest's runtime.sx:
scheme "scheme-*" "scm-*" erlang "er-*" "erlang-*"
prolog "pl-*" common-lisp "cl-*" "clos-*"
js "js-*" haskell "hk-*"
Verified under opt-in JIT (== CEK, no hang): smalltalk 847/847, scheme/flow
166/166, erlang 530/530, prolog 590/590, apl 152/152, js 147/148. Residual
(documented, protected by the default gate): common-lisp 6 fails in advanced
suites (parser-recovery/debugger/CLOS/MOP). lua (0/16) and tcl (3/4) fail
identically on CEK — pre-existing, not JIT. run_tests --jit/no-jit unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
register_jit_hook is now installed in the persistent (epoch) serving-mode
branch of sx_server.ml, not just --http/cli/site. Smalltalk-on-SX conformance
under JIT is 847/847 — identical to the no-JIT baseline; Datalog 356/356.
run_tests --jit/no-jit are byte-identical before/after (no regression).
Five distinct root causes fixed (not one "miscompile"):
1. Serving mode never loaded lib/compiler.sx, so JIT used the native
Sx_compiler.compile stub (arity-0 bytecode, params as GLOBAL_GET →
"VM undefined: <param>"). Server-mode branch now loads compiler.sx
before registering the hook, matching http/cli/site.
2. compile-cond / compile-case-clauses / compile-guard-clauses only treated
keyword :else and true as the catch-all, not the bare symbol `else` that
the CEK's is-else-clause? accepts → GLOBAL_GET "else". (lib/compiler.sx)
3. OP_DIV produced a float for non-divisible Integer/Integer (1/2 → 0.5)
instead of the exact Rational the "/" primitive returns. Now delegates to
the primitive, matching CEK. (sx_vm.ml)
4. OP_EQ / _fast_eq lacked Rational/ListRef cases that the "=" primitive's
safe_eq has → (= 1/2 1/2) false under JIT. OP_EQ now delegates non-scalars
to the "=" primitive; _fast_eq gained rational + ListRef. (sx_vm.ml,
sx_runtime.ml)
5. Continuation-based control flow (Smalltalk ^expr non-local return, block
escape, exceptions via call/cc) can't run in the stack VM. New data-driven
exclusion set Sx_types.jit_excluded + `jit-exclude!` primitive, consulted in
jit_compile_lambda (covers both the CEK hook and vm_call's tiered path).
lib/smalltalk/eval.sx self-declares its continuation dispatch core
interpret-only; pure helpers still JIT. The SUnit suite-runner test helper
pharo-test-class miscompiles mid-loop and is excluded in tests/tokenize.sx.
Also adds SX_JIT_DENY / SX_JIT_ONLY env-var bisection filters to the serving
hook. Known residual documented in plans/jit-bytecode-correctness.md: the hook
re-runs a failed VM execution via CEK (correct result, possible duplicate side
effects); adopting run_tests' propagate-don't-rerun semantics is deferred to
avoid changing shared VM/CEK behavior under this loop.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Assert mau/confluent? actually discriminates: the Peano-arithmetic variant of the
optimisation laws is flagged non-confluent with named non-joinable pairs, so the green
'opt module is confluent' is real evidence rather than a rubber stamp. maude-optimize
40/40, total 198/198.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
artdag/opt-improvement compares the original output cone (dce to id) vs the
maude-reduced DAG under an injected cost-fn, returning before/after total-work and
critical-path. opt-cheaper? asserts optimisation never increases cost: the 5-node
chain drops to 2 (work 5->2, path 5->2) and stays cheaper under radius-weighted cost
(5->3); over dedup and untouched DAGs are never pessimised. Consumes cost.sx. Phase 7
base + (later) cost box done. maude-optimize 38/38, total 196/196.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
artdag/opt-reduce: encode a DAG cone -> opt-term, mau/creduce against the
optimisation module, decode the normal form back to build-entries and rebuild.
Result-preserving: a 5-node blur;blur;id;bright0 chain collapses to 2 nodes and an
over(I,I) dedup 3->2, both executing identically to the original; non-optimisable
DAGs round-trip their radius faithfully (unary 1+1+1 -> 3). Completes Phase 7's
bridge-back + equivalence boxes. maude-optimize 33/33, total 191/191.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/artdag/optimize-rules.sx — the effect-pipeline optimisation passes (identity
elim, no-op/zero-radius elim, adjacent fusion, idempotent over dedup) as a maude
module. Radius algebra is _+_ [assoc comm id: 0] (NOT Peano successor rules, which
are non-confluent here); mau/confluent? certifies 0 non-joinable critical pairs, so
the optimised pipeline's normal form / content id is rewrite-order stable. Consumes
lib/maude/confluence.sx. maude-optimize 25/25, total 183/183.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds lib/maude/confluence.sx — the CID-stability oracle the artdag optimiser
needs. 274 tests.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Substrate fix: two-line change to lib/erlang/runtime.sx that lets
http-listen handler routes call gen_server:call without deadlocking.
1. er-sched-step-alive!: pass :pending-args (when set) to the
initial-fun call instead of always passing an empty list.
Default behavior (no field) stays (list) — drop-in safe.
2. er-bif-http-listen sx-handler: instead of er-apply-fun handler
inline (which blows up on receive's er-suspend-marker because
the connection thread has no scheduler step on its stack),
create a real er-process with :initial-fun = handler and
:pending-args = (list req-pl), then er-sched-run-all! to drain.
Any receive (e.g. gen_server:call) suspends + resumes inside
the SX scheduler frame the process owns. Read :exit-result
for the response proplist; marshal back to SX dict.
Investigation arc (see plans/fed-sx-milestone-2.md Blockers #4 +
Progress log):
- loops/fed-prims bf8d0bf2 diagnosed it as Erlang-substrate, not
OCaml mutex (Pattern A wrong, Pattern B right but sketchy).
- First Pattern B attempt failed: tried er-spawn-fun on a raw SX
lambda, hit (er-fun? fv) gate. Connection-thread bisect
pinpointed the exact line.
- Real fix: use the existing er-fun (user's handler) directly,
but feed it via :pending-args so step-alive's hardcoded
(list) doesn't drop the request arg.
Acceptance:
- new next/tests/smoke_kernel_route.sh: 6/6 over real HTTP
(welcome /, /actors/alice, /actors/alice/outbox with
gen_server-backed tip, /actors/alice/inbox, unknown-actor,
via http_server:start(P, [{kernel, nx_kernel}])).
- next/tests/http_server_tcp.sh: 5/5 (bumped wait_bound from
30s to 180s — cold boot is slow under sibling-loop CPU load
and the per-handler scheduler ramp adds a small margin).
- Erlang conformance: 761/761.
Step 12's two-instance smoke test is now unblocked — its full
Follow / Accept / Note flow can layer on top of this kernel-route
surface. m2 plan updated.
Pre-existing httpc_request.sh flakiness ("Undefined symbol:
http-request" on the live-call epochs) reproduces WITHOUT this
change — see git stash A/B in the investigation. Unrelated.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
A tz event now exports DTSTART;TZID=<name>:<local> (EXDATE/RDATE likewise;
UNTIL stays UTC per RFC), and the VCALENDAR emits a VTIMEZONE per distinct zone
with DAYLIGHT/STANDARD sub-components generated from the zone's transition rules
(offsets + FREQ=YEARLY;BYMONTH;BYDAY) — London/Paris blocks match real-world
definitions. Clients recur at fixed wall-clock time, DST-correct (prior caveat
gone). Importer tolerates ;TZID= params. 376/376 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/tests/effects.sx — proves artdag's effect-pipeline optimisations
(fusion, no-op/dead-op elim, identity elim, CSE/idempotent dedup) are
equational rewriting: the optimised pipeline is the normal form, confluence
gives a stable content id. The 'second consumer' spike for a maude-driven
optimiser in lib/artdag. Surfaced faithfulness note: id: affects matching/canon
not auto-reduction.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Bug: tz events store wall-clock LOCAL times but export stamped them with a Z
(UTC) suffix, so a London 18:00 event falsely read as 18:00 UTC. ev-ical-conv
now converts a tz event's DTSTART/UNTIL/EXDATE/RDATE local->UTC before
formatting (London summer 18:00 -> 170000Z; Paris -> 160000Z); non-tz events
unchanged. Caveat: UTC RRULE drifts from wall-clock-stable tz recurrence across
a DST boundary (VTIMEZONE deferred). 366/366 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ical.sx parses VEVENT/VCALENDAR text back into events (ev/ical-lines->event,
ev/parse-vcalendar): DTSTART/DURATION/RRULE (ordinal BYDAY, BYMONTHDAY, UNTIL/
COUNT/INTERVAL) + EXDATE/RDATE. Round-trip is occurrence-exact — export->import
expands to the identical occurrence set. Completes bidirectional interop.
360/360 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
run.sx now handles 'search START =>* GOAL .' (reports the witness path) and
mau/run-pretty prints Maude-style 'result SORT: TERM' using least-sort
inference. searchpath.sx exposes mau/search-path-terms (term-level entry).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/sorts.sx — mau/term-sort computes the least sort of a term (smallest
result sort among op declarations whose arg sorts the actuals satisfy modulo
subsorting); overloaded f(1)=NzNat vs f(s 0)=Nat. mau/has-sort? for
membership-style checks. Answers the plan's order-sorted substrate question.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Infix ops parse left (default / gather (E e)) or right (gather (e E)) per the
gather attribute, so _:_ [gather (e E)] reads a : b : c as right-nested. Full
insertion sort now runs over bare cons lists with no parentheses.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Parser reads trailing eq attributes (eq L = R [owise] .) via mau/split-attrs.
mau/crewrite-top is two-pass: ordinary equations first, owise last — an owise
catch-all fires only when no ordinary equation applies, regardless of
declaration order. Verified a catch-all declared first still defers.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/searchpath.sx — mau/search-path returns the shortest sequence of
states from start to goal (the solution moves), mau/search-length its step
count. BFS over all one-step successors, threading the path.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/run.sx — mau/run-program / mau/run parse a module plus trailing
reduce/red/rewrite/rew commands (with optional 'in MOD :' qualifier) and
execute them, rendering results in mixfix surface syntax. An idiomatic
.maude file now runs end-to-end.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/pretty.sx — mau/term->maude renders internal prefix terms back
in Maude mixfix syntax driven by op forms; mau/red->maude / mau/rew->maude
reduce-then-render. Output now reads as idiomatic Maude.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/meta.sx — up-term/down-term encode terms as data (mt-var/mt-app),
reflective meta-reduce/meta-rewrite/meta-apply, the meta-circular law
down(metaReduce(up t)) =AC= reduce t, and meta-prove-equal? as a generic
equational theorem helper. Verified round-trips, reflection agreement,
single-rule meta-apply, and proving commutativity/associativity instances.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/strategy.sx — first-class set-valued strategies: idle/fail/all/
rule/seq/alt/star/plus/bang/name combinators, named-strategy env. Same
rule set computes different things under different strategies; verified
with single-rule vs all vs seq-order vs alt vs star vs bang.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/rewrite.sx: rl/crl transitions interleaved with eq normalisation.
mau/rewrite = default strategy (top-down, leftmost-outermost, first rule);
mau/rew bounded; mau/search = BFS reachability over all successors.
lib/maude/fire.sx: short-circuiting matcher (mau/fire-eq) — finds the first
productive match instead of enumerating the whole solution set. Fixes the
exponential blowup of AC rewriting on many identical elements (8 coins:
60s+ to <1s). Eager match-multiset kept only for match-all / search.
Verified on AC coin-change, traffic light, branching search, crl clock.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ev/book-series! / ev/cancel-series! apply a booking/cancel to every occurrence
of one event in a window (RSVP the whole weekly class), returning per-
occurrence (occ-key status) results; capacity still enforced per occurrence
(some :booked, some :full), idempotent re-book (:already). ev/series-count,
ev/series-booked. 341/341 green.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
lib/maude/conditional.sx — condition-aware reducer. ceq fires only when
its guard holds: equational guards (l=r reduce to same normal form) and
boolean guards (term reduces to true), evaluated by recursing through the
same reducer. Verified on gcd, insertion sort, max, even.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The chisel. lib/maude/matching.sx: multi-valued matcher mau/mm returning
ALL substitutions, dispatching on op theory (free/comm/assoc/AC). Identity
lets variables grab empty blocks. AC-canonical form (mau/canon) powers
ac-equal? and deterministic printout. AC rewriting extends f-AC equations
with rest vars so a rule fires on any sub-multiset/subword; mau/first-change
only commits rewrites that change the canonical form (idempotency/identity
terminate). Verified on multiset rewriting, set theory, group equations.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
