`writable.write(chunk, cb)` with `cb` passed as a function VALUE lost the
completion's error: `lowerWriteCb` checked only the callback's syntax and
lowered every other callback through `lowerExprExpecting(cbArg,
funcOf([], VOID))`, which adapts a one-parameter function down to zero
parameters. The native binary therefore printed `undefined` where Node
prints `null`, or the write's own error (`ERR_STREAM_WRITE_AFTER_END` /
`ERR_STREAM_DESTROYED` for late and destroyed writes).
Lower such callbacks through a new `writable.writeEcb` libCall
`(recv, mode, chunk, encoding, cb)`: a zero-parameter callback keeps the
existing synchronous path, and a callback that declares the error (a dyn
value, or `null | Error`) is wrapped by `scr_stream_ecb_wrap` with the
compiler-built adapter that delivers the error the write completion
carries. The stream unit publishes that error only in its
`SCR_STREAM_WRITE_ERRORS` build, selected by the new `streamWriteErrors`
runtime feature (rule: any `writable.writeEcb`). Every other build
compiles `SCR_WCB_CALL` to the same plain call, so their objects are
unchanged.
The destruction path also matched Node: `afterWrite` calls `cb(null)`
even for a stream destroyed after the write was queued (Node 24
`internal/streams/writable`), so the destroyed `SCR_ST_AFTER_WRITE` tick
passes NULL rather than the stream's stored error.
Corpus: `tests/corpus/writable-error-callbacks.ts` covers success, an
erroring write, a write after end, and a write on a destroyed stream.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reading `process.stdout.write` as a value, `write.bind(stream)`, and
`process.stdout.write = fn` now reach the runtime stream object's write
slot in TypeScript, and a program that replaces an output write method
calls its typed writes through that slot (SC8020 refuses the combination
with Node islands until H2 shares the stream object with them).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The socket registry, the http client registry, an agent's entry list and
fetch's list of transfers in flight were singly linked: each append walked
to the tail and each removal walked to the element. A client request on a
pooled connection registers and unregisters a socket and a client and adds
and removes an agent entry, so every request walked lists as long as the
open connections. These lists now link both ways with a last pointer, in
the same order as before. A client finds its agent entry through a
back-pointer, and an agent counts its queued entries, so a finished
request scans for a queued one only when there is one.
scr_str_case_conv scans ASCII text for the first letter to convert eight
bytes at a time, then converts the rest eight bytes at a time; the tail
and big-endian targets take the byte loop. With a refreshed profile, the
byte-at-a-time scan compiled worse for short strings with nothing to
convert, which made validate-errors slower.
The runtime profile is collected again on this change with
scripts/refresh-runtime-profile.mjs (326 raw profiles). Changed
functions hold 0.0% of its counts, against 3.7% for the previous one.
tests/corpus/string-case-ascii-words.ts covers ASCII strings of every
length from 0 to 40, with letters at block boundaries and next to the
bytes just outside A-Z and a-z.
An uncork() flushed a head writeHead stored even when nothing was
written while corked. Node keeps the stored head until write(), end() or
flushHeaders(), so the flush now sends it only when a write() or
flushHeaders() ran while corked. The http-held-head fixture covers the
three cases.
#2468 shipped child.off()/removeListener() and child.signalCode in the gated
scr_child_lifecycle.c unit, but it also restructured the always-linked
scr_child.c terminal dispatch loops to detach their listener arrays before
emitting. scr_child.c is compiled once per program and linked into all of
them, so its object moved and every child-process executable grew: the
zero-cost gate reported 55 runtime-only corpus programs.
Restore scr_child.c byte for byte and move the emit-snapshot discipline into
the gated unit instead. The base loops already bracket each terminal emit
with a state flag that only the emitter writes -- `settled` for 'exit' and
'error' (set after the stream drain, immediately before the terminal branch)
and `close_done` for 'close' (set at the top of the close emit) -- and run no
user code in between. A removal inside that window is therefore a no-op: the
entry stays, the in-flight emit still delivers it, which is exactly Node's
answer (emit copies the listener array), and the emitter's own drop releases
it. Outside the window no emit of that event is in flight, so removal in
place is Node-exact, and the base already releases late registrations once
the window has closed. 'spawn' detaches its array before invoking its
listeners, so removal during it finds an empty registry.
scr_child_private.h repeats the structure layout for the gated unit only;
scr_child.c keeps its own definitions untouched.
Validation: zero-cost gate against 68980c241 reports no runtime-only program
and no scr_child symbol in any IR, LLVM or executable difference; scr_child.o
is byte-identical to the base in all six shipped flavors, and the
scr_child_off_*/scr_child_signal_code entries live in scr_child_lifecycle.o.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The BigInt reciprocal divided a two-limb value by one limb, which compiles
to a __udivti3 call. zig provides that builtin from its compiler-rt, a
single object that also defines memset, memcpy, memmove, memcmp, bcmp,
strlen and the _chk variants. Linking it for __udivti3 made those
definitions take precedence over libc's in every glibc executable built
with SCRIPTC_CC=zigcc, and zig's bcmp and memcmp compare one byte per
iteration and its memset stores one byte at a time.
The reciprocal now divides in 32-bit halves with one-limb operations, so
the runtime references no compiler-rt builtin and glibc executables import
the four functions from libc.
The sweep and the dispatch loop's pending check walked every registered
socket on each loop pass, so a server with many keep-alive connections
paid for all of them on every request. A socket now joins a work queue
when it may have deferred work (bytes buffered for a consumer, write
callbacks waiting for the drain, 'finish', an error or 'close') and
leaves it when a sweep visit leaves it none. The queue is ordered by
registration, so the sweep visits sockets in registry order as before.
With SCR_RC_AUDIT, the dispatch loop checks that no registered socket
outside the queue has deferred work.
res.writeHead() builds the head and keeps it until the first write(),
end() or flushHeaders(), which sends it with that call's data in one
socket write, as Node's _storeHeader does. A response with an Expect
header sends the head at once, as in Node. The http-held-head server
fixture compares the wire timing with Node's.
A client response cut off mid-body emitted 'error' ("aborted") whether or
not it had an 'error' listener, so a response without one ended the
process with "Unhandled 'error' event". Node's IncomingMessage reports its
destroy error only to 'error' listeners; 'aborted' and 'close' still fire.
fetch leaves a redirect response by destroying its request, so following
a redirect whose chunked body had not ended crashed the process before
fetch settled. It now follows the redirect.
Fixes#2395
Co-authored-by: linguofeng <460328+linguofeng@users.noreply.github.com>
readline's emit and EventEmitter's emit copy the listener list before
running it, so that listeners added or removed during the emit don't
change who runs. The copy was a malloc and a free per emitted event: per
line for a readline 'line' listener. A list of at most four listeners is
now copied into a stack buffer; longer lists still allocate.
A new corpus program pins the snapshot semantics for lists below and above
four listeners: a listener added during an emit runs from the next emit, a
removed one still runs in that emit, and a once listener runs once.
- child_process: the exit/close/error dispatch loops read the live
listener registry in their loop condition, so a listener that called
off/removeListener on a sibling cut the in-flight emit short. Node
copies the listener array before it emits; detach the registry
(scr_child_detach_exit/_close/_err), deliver that snapshot, release
each closure, and free the array in both platform arms.
- process: a first-class alias of stdout/stderr (`const out =
process.stdout`) keeps the stream's type but read `writableLength`
through the keyed dynamic get, which answered undefined. Fold that
read to the same synchronous-write 0 the direct lowering answers,
leaving the runtime untouched.
- fs: a `const` write-options record was inlined at the write site even
when it was mutable, so `OPTS.mode = 0o644` before the call diverged
from Node, which reads the record when the call runs. Accept only a
record whose properties are all readonly (`as const`, `Readonly<>`);
a mutable one keeps the SC2020 fence.
- Refresh the node-types-fenced snapshot: `path.win32` sub-namespaces
lower now, so the PlatformPath.sep fence is gone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Resolve tests/harness/npm.test.ts by taking main's npm-suite split and
porting the island path normalisation and the two island KNOWN entries
into tests/harness/npm-suite.ts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adding the six new entry points (scr_child_off_spawn/_exit/_close/_error,
scr_child_has_signal_code, scr_child_signal_code) to scr_child.c changed
the executable of every program that links the unit, although none of
them calls the new functions and IR and LLVM stayed identical: the extra
functions grow the unit's `.eh_frame` contribution, and the linker's
frame bookkeeping for the retained sections changes with it (the first
visible difference is the phdr sizes at byte offset 208). Appending them
after the unit's own functions does not help — the unit's frame section
still differs, so all 89 corpus programs that link scr_child.o got a
different binary.
Move the entries into scr_child_lifecycle.c, a unit of their own
selected only by programs that use them (feature childLifecycle, the
scr_dyn_handle_props.c pattern: the matrix's optional source list, the
RUNTIME_FEATURE_RULES rule for child.offSpawn/offExit/offClose/offError/
signalCode, and the link-feature plumbing through the executable cache,
CcOptions and RUNTIME_UNITS). scr_child.c keeps its original content and
layout; the new unit reads its registries through scr_child_private.h,
which holds the per-platform ScrChild layout both units compile against.
A program that does not call the new entries links exactly the units it
did before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Three shapes from pi's process/child helpers:
1. A cleanup arrow such as `() => process.off(signal, handler)` sits in
@types/node's `Process`-returning slot. The lambda's ABI was decided
from its inferred return type before the contextual void slot was
consulted, so the arrow promised a `Process` value the lowering cannot
produce and the registration fenced. `lambdaSignature` now takes the
contextual void answer first; the adoption block further down already
computes it, so every program that compiles today keeps its IR.
2. `child.off`/`removeListener` were unimplemented. They join the
`child.on` family, are limited to the removable literals `spawn`,
`exit`, `close` and `error`, and require the listener value that was
registered — removal is by closure identity, scanning the registry
from the end as Node does. New C runtime entry points
`scr_child_off_spawn/_exit/_close/_error` release the entry and shift
the tail.
3. `child.signalCode` reads `string | null` from the recorded exit
signal, with `scr_child_has_signal_code`/`scr_child_signal_code`
behind the same typed-union wrap as `spawnRes.signal`; the IR
validator pins the `string | null` result.
The process listener registration paths (stdin events, IPC message and
disconnect, uncaughtException and uncaughtExceptionMonitor,
unhandledRejection, warning, signals, exit) now accept any discarded
expression, matching `child.on` — `resultIsDiscarded` rather than a bare
expression statement — so `handlers.push(() => process.off(…))` compiles.
Evidence: corpus 8412 (removal through `off` and `removeListener` for
all four events, `signalCode` while running, after a normal exit, after
a fatal signal and after a failed spawn, idempotent double removal) and
the `node-types` fixture plus project-config test that pins pi's exact
`() => process.off(...)` cleanup closure. Ambient declarations gain the
`signalCode` property and the four `off`/`removeListener` overloads.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Node's writeFileSync accepts an open file descriptor where it accepts a
path: it writes the data at the descriptor's current offset and leaves the
descriptor open, since closeSync owns its lifetime. The static fs surface
declared only the path forms, so `fs.writeFileSync(fd, text)` failed type
mapping — the compiler was missing a documented overload, not a
conversion.
Declare the (fd, data) and (fd, data, "utf8" | "utf-8") forms in the
node-fallback typings, route the call by the first argument's static f64
type to a new fs.writeFileSyncFd libfn, and implement it as
scr_fs_write_file_fd: a write loop over the whole string through the
descriptor-validating scr_fs_write_bytes, resuming short writes, with the
descriptor operation's errno shape on failure.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A `file:` module namespace handed to typed code now boxes as Node's
namespace object instead of the "converting typed values to 'unknown'"
fence: `const u: unknown = ns`, `take(ns)`, a `Promise<typeof
import("./m")>` widened to `Promise<unknown>` (or `Promise<unknown> |
undefined`), and `await import("./m")` in an `unknown` slot all see one
non-extensible null-prototype object per module whose own properties are
the module's exports in Node's sorted order. They are enumerable
read-only accessors whose reads call the compiled namespace getter, so
they stay live, and the object carries Symbol.toStringTag "Module",
Node's strict-mode write TypeErrors, the not-extensible add error and
the ToPrimitive throw. `node:` builtins have no compiled export table
and keep their SC1101 fence.
The object is built by a new gated runtime unit,
packages/runtime/src/scr_module_namespace_dyn.c, selected by the
`moduleNamespaceDyn` runtime feature (`module.namespaceDyn` in the IR).
It adds no field to ScrDynHandleOps and no branch to a shared unit: the
core reaches it through the ordinary SCR_DYN_OBJ paths, so key listing,
for-in, `in`, JSON, identity, property reads and assignment all reuse
the existing dispatch. The compiler reuses the getter lifted for
module.namespaceValue (existing programs' IR is unchanged) and adds
moduleNamespacePromiseDynValue, one lifted async mapper per module for
the promise slots.
Zero cost: a program that never boxes a namespace links exactly what it
did. The zero-cost gate reports every program identical, with
byte-identical IR and LLVM; the runtime pack's only difference is the
new unit (0 changed runtime functions), and among the namespace corpus
programs only 8412 links it.
Evidence: tests/corpus/8412-module-namespace-untyped is a differential
program against Node in both the plain and sanitized lanes, and the
diagnostics snapshots and TypeScript 7 baselines stay unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
After a fetch body's handle gained its web view, a read, cancel or pipe
started on the view left Response.bodyUsed false until the view's pull
reached the body, where Node's bodyUsed reads the stream's disturbed state
at once. bodyUsed now also answers the view's disturbed state, which the
web streams' bridge reports (ScrWebStreamBridge.disturbed).
A program that names none of the web stream classes linked them only for
member accesses the checker resolves to ReadableStream, so body.tee() or
for await over a body inside a function whose parameters have no type
threw a TypeError. The same member names on an untyped receiver, and
for await over an untyped value, now link the classes in a program that
links the fetch unit.
Adds corpus program fetch-body-web-stream-helpers.mjs, with its corpus
cost and preflight/order baseline, and bodyUsed cases to
fetch-body-web-stream-linked.mjs.
In JavaScript, a Response or Request body answered getReader, cancel and
locked, and every other ReadableStream member was undefined: tee,
pipeTo, pipeThrough, values and for await threw TypeErrors where Node's
body is a ReadableStream.
In programs that link the web stream classes, the body's handle now
stands for a ReadableStream once any other member is used
(sf_stream_web_view): a stream with a zero high-water mark whose native
source reads the body (scr_webstream_readable_over). From then on every
member goes to it, the class's method values and getters run on it
through the alias's view (scr_dyn_handle_class_receiver), and text(),
clone(), new Response(body) and fetch uploads read the body through it,
so its lock and bodyUsed follow it as in Node. A body locked when the
view is made keeps the view locked until the lock is released. Once
only the view's source holds the body, the body lets the view go.
The compiler links the classes into a program whose JavaScript reads a
ReadableStream member that the fetch profile lists as unsupported by the
native stream, or iterates a ReadableStream with for await, and that
names none of the classes: %main calls webstream.install first.
Programs without web streams build to the same binaries as before.
Adds corpus programs fetch-body-web-stream-linked.mjs,
fetch-body-web-stream-unlinked.mjs and fetch-body-web-stream-network.mjs,
with their corpus costs and preflight/order baselines.
The backend declares scr_unwind_resume, so scr_exception.h carries its
prototype, as the runtime ABI test requires. The unwind-ABI pattern names
both entry points in full, so the test reads no partial symbol from it.
A new plan was copied into its slot as a whole struct, which wrote the
slot's pc with a plain store while other threads' lookups read it
atomically: a data race. The insert now copies the fields after pc and
publishes pc only with the release store.
The unwinder test also runs its threaded cases under ThreadSanitizer,
which reports the race without this change.
text(), json(), bytes() and the other body readers of a Response or
Request whose body is a ReadableStream asked the stream's source for more
chunks after draining its queue only when the stream had a zero
high-water mark. When the queue had filled before the read began (a timer
ran after the Response was made), the reader took the queued chunk and
then waited for a close that no pull would bring, and the read never
settled. The reader now pulls whenever it empties the queue, so it reads
to the end, as Node does. This applies to streams from ReadableStream.from,
from an underlying source, and to JavaScript ReadableStreams.
Adds corpus programs fetch-body-queued-before-read.ts and
fetch-body-queued-before-read-untyped.mjs, with their corpus costs and
preflight/order baselines.
scr_unwind_raise walks the stack with plans cached per return address.
The first raise through a return address finds the function's FDE
through the .eh_frame_hdr of the object that contains it, runs the CFI
up to the call, and finds the call site's landing pad in the LSDA; the
plan records the frame's CFA rule, where its callee-saved registers are
saved, and the landing pad. Later raises look the plans up, restore the
registers, and jump to the first landing pad. Plans are inserted under
a lock and read without one.
Landing pads continue the unwind with scr_unwind_resume instead of
`resume`, so the walk goes on from their frame. A frame whose CFI,
pointer encodings or personality the plans do not express, or a return
address without an FDE, leaves the unwind to libgcc's forced unwind
from the frame the walk started at, as before. Other Linux targets keep
libgcc's unwinder; their scr_unwind_resume calls _Unwind_Resume.
bench:runtime gains layered-errors, a workload whose errors are thrown
below small functions that hold references and caught per entry in a
loop, and the corpus gains unwind-landing-pad-chains, which pins landing
pads that continue the unwind, including in frames LLVM inlines into a
caller whose handler takes the exception.
The island host replaced process.on, process.addListener and
process.prependListener with separate wrappers, and process.emit with its
own, as own enumerable properties of process. Island code could see the
difference: Object.keys(process) listed the four, process.on was no
longer process.addListener, prependListener was named addListener, and
each was an own property where Node inherits it.
One wrapper now serves on and addListener, and prependListener and emit
keep their names. They are defined on process's own prototype, which
Node gives process below EventEmitter.prototype, with
EventEmitter.prototype's attributes. process has no own property for
them, so Object.keys, Object.hasOwn and for...in see what they see in
Node. for...in lists a prototype's keys before its parent's, so the
keys EventEmitter.prototype lists before the wrapped ones (_events,
_eventsCount, _maxListeners, setMaxListeners, getMaxListeners) are
defined there too, with the same values, in their order.
6891 compares the identities, names, lengths, own-ness, Object.keys and
for...in over process against Node.
