PR-URL: https://github.com/nodejs/node/pull/31624
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
HTTP header values can have trailing OWS, but it should be stripped. It
is not semantically part of the header's value, and if treated as part
of the value, it can cause spurious inequality between expected and
actual header values.
Note that a single SPC of leading OWS is common before the field-value,
and it is already handled by the HTTP parser by stripping all leading
OWS. It is only the trailing OWS that must be stripped by the parser
user.
header-field = field-name ":" OWS field-value OWS
; https://tools.ietf.org/html/rfc7230#section-3.2
OWS = *( SP / HTAB )
; https://tools.ietf.org/html/rfc7230#section-3.2.3
Fixes: https://hackerone.com/reports/730779
PR-URL: https://github.com/nodejs-private/node-private/pull/189
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
wrk did not support sending optional headers, they were silently
ignored. This appears to be an oversight, they were supported for
autocannon.
PR-URL: https://github.com/nodejs-private/node-private/pull/189
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
This removes 'to Number' casting in multiple benchmarks (which is
handled by the benchmark runner) and cleans up some var usage in changed
benchmarks.
PR-URL: https://github.com/nodejs/node/pull/31581
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Add a raw `MessagePort` benchmark that does not ping back and forth
between different threads, unlike the `echo.js` benchmark, as there
are some performance differences between single-threaded and multi-
threaded operation, and a single-threaded Environment can be somewhat
easier to work with when profiling.
PR-URL: https://github.com/nodejs/node/pull/31568
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Use let and const in domain, es, events, fixtures, fs, http,
http2 and misc.
PR-URL: https://github.com/nodejs/node/pull/31518
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/31522
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
This implementation provides optimizations not included upstream.
PR-URL: https://github.com/nodejs/node/pull/31201
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
8fb5fe28a45cc884567cd39e3b2f6b4272917af broke the benchmark for
getStringWidth(). This fixes it up by updating the argument to
`require()` to retrieve `getStringWidth()` from the new internal module
location.
PR-URL: https://github.com/nodejs/node/pull/31476
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
This fixes a benchmark test that was recently broken by a breaking
change on the master branch.
Fixes: https://github.com/nodejs/node/issues/31372
PR-URL: https://github.com/nodejs/node/pull/31377
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
This also allows us to remove backslash escaping for `[` and `]`
inside of header code, which makes the bare markdown more readable.
Refs: https://github.com/nodejs/node/pull/31086
PR-URL: https://github.com/nodejs/node/pull/31149
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anto Aravinth <anto.aravinth.cse@gmail.com>
Reviewed-By: Beth Griggs <Bethany.Griggs@uk.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Currently we only consider write when benchmarking.
PR-URL: https://github.com/nodejs/node/pull/31066
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
adds clear connections to the secure-pair performance test to prove
that in some cases (when the sender send the data in small chunks)
clear connections perform worse than TLS connections
Also add a byte chunk size test to benchmark/net/net-pipe.js
Refs: https://github.com/nodejs/node/issues/27970
PR-URL: https://github.com/nodejs/node/pull/27971
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Make manywrites benchmark a bit more realistic by
taking back pressure into account. Otherwise
memory usage would no correspond well with
real world usage.
PR-URL: https://github.com/nodejs/node/pull/30977
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
* Let users provide more than one pattern by repeating the flag
* Add new flag --exclude to exclude patterns
* Add tests for --filter
* Document --filter
This commit also fixes a bug where things like
`compare.js --new --old binary --new binary` was acceptable (now the
script will exit and print the usage message).
PR-URL: https://github.com/nodejs/node/pull/29987
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
This makes sure we do not retrieve the handler in case it's not
required. This improves the performance a tiny bit for these cases.
PR-URL: https://github.com/nodejs/node/pull/30767
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
This adds a couple of benchmarks to check different options and code
paths.
PR-URL: https://github.com/nodejs/node/pull/30767
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
The doc/guides directory looks like a place for guides to using Node.js
but it's actually where we store development/internal documentation.
Move the benchmark guide to the benchmark directory where someone (who
doesn't know that bit of somewhat-surprising information) is more likely
to find it.
PR-URL: https://github.com/nodejs/node/pull/30781
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Improve performance and reduce memory usage when a writable stream
is written to with the same callback (which is the most common case)
and when the write operation finishes synchronously (which is also
often the case).
confidence improvement accuracy (*) (**) (***)
streams/writable-manywrites.js sync='no' n=2000000 0.99 % ±3.20% ±4.28% ±5.61%
streams/writable-manywrites.js sync='yes' n=2000000 *** 710.69 % ±19.65% ±26.47% ±35.09%
Refs: https://github.com/nodejs/node/issues/18013
Refs: https://github.com/nodejs/node/issues/18367
PR-URL: https://github.com/nodejs/node/pull/30710
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Upcoming lint rule will require a blank line between consecutive
functions. Add it in the places where we don't have it already.
PR-URL: https://github.com/nodejs/node/pull/30696
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Add an `serialization` option that allows child process IPC to
use the (typically more powerful) V8 serialization API.
Fixes: https://github.com/nodejs/node/issues/10965
PR-URL: https://github.com/nodejs/node/pull/30162
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Warn when a non-existent property of an unfinished module.exports
object is being accessed, as that very often indicates the presence
of a hard-to-detect and hard-to-debug problem.
This mechanism is only used if `module.exports` is still a
regular object at the point at which the second, circular `require()`
happens.
The downside is that, temporarily, `module.exports` will have a
prototype other than `Object.prototype`, and that there may
be valid uses of accessing non-existent properties of unfinished
`module.exports` objects.
Performance of circular require calls in general is not
noticeably impacted.
confidence improvement accuracy (*) (**) (***)
module/module-loader-circular.js n=10000 3.96 % ±5.12% ±6.82% ±8.89%
Example:
$ cat a.js
'use strict';
const b = require('./b.js');
exports.fn = () => {};
$ cat b.js
'use strict';
const a = require('./a.js');
a.fn();
$ node a.js
(node:1617) Warning: Accessing non-existent property 'fn' of module exports inside circular dependency
/tmp/b.js:4
a.fn();
^
TypeError: a.fn is not a function
at Object.<anonymous> (/tmp/b.js:4:3)
[...]
PR-URL: https://github.com/nodejs/node/pull/29935
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Add an option that controls the size of the internal
buffer.
Fixes: https://github.com/nodejs/node/issues/29941
PR-URL: https://github.com/nodejs/node/pull/30114
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Prepare the final few documents that haven't been updated to always use
`[]` with reference links and to escape `[` and `]` for things that
aren't links in markdown files.
PR-URL: https://github.com/nodejs/node/pull/29963
Reviewed-By: Masashi Hirano <shisama07@gmail.com>
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Since worker threads are complete Node.js environments, including the
ability to load native addons, and since those native addons can
allocate resources to be freed when objects go out of scope, and since,
upon worker thread exit, the engine does not invoke the weak callbacks
responsible for freeing resources which still have references, this
modification introduces tracking for weak references such that a list
of outstanding weak references is maintained. This list is traversed
during environment teardown. The callbacks for the remaining weak
references are called.
This change is also relevant for Node.js embedder scenarios, because in
those cases the process also outlives the `node::Environment` and
therefore weak callbacks should also be rendered as environment cleanup
hooks to ensure proper cleanup after native addons. This changes
introduces the means by which this can be accomplished.
A benchmark is included which measures the time it takes to execute the
weak reference callback for a given number of weak references.
Re: https://github.com/tc39/proposal-weakrefs/issues/125#issuecomment-535832130
PR-URL: https://github.com/nodejs/node/pull/28428
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
This improves dgram performance by avoiding unnecessary async
operations.
One issue with this commit is that it seems hard to actually create
conditions under which the fallback path to the async case is
actually taken, for all supported OS, so an internal CLI option
is used for testing that path.
Another caveat is that the lack of an async operation means
that there are slight timing differences (essentially `nextTick()`
rather than `setImmediate()` for the send callback).
PR-URL: https://github.com/nodejs/node/pull/29832
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Benchmark different types of operations and make results comparable
by normalizing process.env for enumeartion.
PR-URL: https://github.com/nodejs/node/pull/29188
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
A lot of strings that are going to be passed to `getStringWidth()`
are ASCII strings, for which the calculation is rather easy and
calling into C++ can be skipped.
confidence improvement accuracy (*) (**) (***)
misc/getstringwidth.js n=100000 type='ascii' *** 328.99 % ±21.73% ±29.25% ±38.77%
misc/getstringwidth.js n=100000 type='emojiseq' 2.94 % ±7.66% ±10.19% ±13.26%
misc/getstringwidth.js n=100000 type='fullwidth' 4.70 % ±5.64% ±7.50% ±9.76%
PR-URL: https://github.com/nodejs/node/pull/29301
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
These are rendered as single breaks.
Addresses Markdownlint MD012 rule.
PR-URL: https://github.com/nodejs/node/pull/29352
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Co-Authored-By: Anna Henningsen <anna@addaleax.net>
PR-URL: https://github.com/nodejs/node/pull/25436
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
