child_process: pass spawn options to the binding positionally#63930
Open
anonrig wants to merge 1 commit into
Open
child_process: pass spawn options to the binding positionally#63930anonrig wants to merge 1 commit into
anonrig wants to merge 1 commit into
Conversation
nodejs-github-bot
added
c++
anonrig
force-pushed
the
process-wrap-fast-spawn-options
branch
from
7b1bdfc to
91a67db
Compare
ChildProcess.prototype.spawn() handed a single options object to the ProcessWrap::Spawn() binding, which then read about a dozen properties back out individually with Object::Get(). Each of those is a property lookup across the JS/C++ boundary on every spawn. Pass file, args, cwd, envPairs, stdio, uid and gid as positional arguments and pack the boolean flags (detached, windowsHide, windowsVerbatimArguments) into a single integer whose bit values are exported from the binding as `constants`. The native side then reads each value directly from the call arguments. Add internal typings for the process_wrap binding (previously untyped) describing the new positional spawn() signature and the exported constants. There is no observable behavior change. Spawn wall-clock time is dominated by the operating system process-creation cost and is unchanged; this reduces the per-spawn work done on the main thread and clarifies the contract between the JS and C++ layers. Signed-off-by: Yagiz Nizipli <yagiz@nizipli.com>
anonrig
force-pushed
the
process-wrap-fast-spawn-options
branch
from
91a67db to
883bf77
Compare
panva
approved these changes
Jun 15, 2026
anonrig
added
request-ci
github-actions
Bot
removed
the
request-ci
Collaborator
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
ChildProcess.prototype.spawn()handed a single options object to theProcessWrap::Spawn()binding, which then read roughly a dozen properties backout one at a time with
Object::Get(). Each of those is a property lookupacross the JS/C++ boundary on every spawn.
This passes
file,args,cwd,envPairs,stdio,uidandgidaspositional arguments, and packs the boolean flags (
detached,windowsHide,windowsVerbatimArguments) into a single integer whose bit values are exportedfrom the binding as
constants(keeping a single source of truth between the JSand C++ sides). The native side now reads each value directly from the call
arguments instead of looking them up by name.
Scope / expectations
This is a refactor, not a measurable speedup on its own. End-to-end spawn
wall-clock time is dominated by the operating system's process-creation cost, so
spawnthroughput is unchanged within noise (verified by an A/B build, seebelow). The benefit is reduced per-spawn work on the main thread and a clearer,
narrower contract between the JS and C++ layers — useful groundwork for moving
more of the
child_processhandling into the native layer.Verification
lib/internal/child_process.js+src/process_wrap.ccbuild cleanly.test/parallel/test-child-process-*(108 tests), thetest/sequentialchild-process tests, and
test/parallel/test-cluster-*all pass.cpplint,eslint, andgit-clang-formatare clean on the changed lines.child-process-spawn-options.js,envc=1024 argc=64,n=3000): old ≈ 422 spawns/s vs new ≈ 405 spawns/s — i.e. no change beyondrun-to-run noise, confirming there is no regression.
There is no observable behavior change.