Imprecision in process.hrtime() in some situations can result in a zero
duration being used as a denominator in benchmark tests. This would
almost certainly never happen in real benchmarks. It is only likely in
very short benchmarks like the type we run in our test suite to just
make sure that the benchmark code is runnable.
So, if the environment variable that we use in tests to indicate "allow
ludicrously short benchmarks" is set, convert a zero duration for
a benchmark to 1 nano-second.
PR-URL: https://github.com/nodejs/node/pull/13110
Fixes: https://github.com/nodejs/node/issues/13102
Fixes: https://github.com/nodejs/node/issues/12433
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
The immediate.js benchmark with `type` set to `depth` measures the same
thing as set-immediate-depth.js. Remove the redundancy.`
PR-URL: https://github.com/nodejs/node/pull/13009
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Having semicolons there runs counter to our documentation and illicits
warnings in pedantic mode. This removes semicolons from after uses of
NODE_MODULE and NODE_MODULE_CONTEXT_AWARE_BUILTIN.
PR-URL: https://github.com/nodejs/node/pull/12919
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Alexey Orlenko <eaglexrlnk@gmail.com>
Previously when there is an error in the chunky client of the
http benchmark, the server would not check the exit code and
thus produce invalid results.
PR-URL: https://github.com/nodejs/node/pull/12916
Fixes: https://github.com/nodejs/node/issues/12903
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Some benchmark tests are failing intermittently, possibly due to
hrtime() imprecision on particular hosts. This change will confirm or
refute that as the root cause the next time the test fails on CI. Either
way, it's a valid check.
PR-URL: https://github.com/nodejs/node/pull/12934
Ref: https://github.com/nodejs/node/issues/12497
Ref: https://github.com/nodejs/node/issues/12433
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
This is first in a hoped-for series of moves away from a monolithic
common.js that is loaded for every test and towards a more modular
approach. (In the end, common.js will hopefully contain checks for
variables leaking into the global space and perhaps some of the more
ubiquitous functions like common.mustCall().)
Move the WPT testing code to its own module.
PR-URL: https://github.com/nodejs/node/pull/12736
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
test-benchmark-child-process failures reveal that
child-process-exec-stdout benchmark sometimes leaves around a stray
yes.exe process. Add code to terminate the process.
PR-URL: https://github.com/nodejs/node/pull/12658
Ref: https://github.com/nodejs/node/issues/12560
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Use the more efficient module.exports = {} pattern,
restructure imports from bindings, requires.
PR-URL: https://github.com/nodejs/node/pull/12681
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/12429
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
When no matching benchmark files are found, a more sensible error
is shown now.
PR-URL: https://github.com/nodejs/node/pull/12421
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: Alexey Orlenko <eaglexrlnk@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/12429
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/12286
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Something unidentified at the moment is causing the arrays benchmarks to
deopt when run with the TurboFan compiler. Refactor the test to use an
inner function that can be correctly optimized by TurboFan and
Crankshaft.
PR-URL: https://github.com/nodejs/node/pull/11894
Ref: https://github.com/nodejs/node/issues/11851#issuecomment-287106714
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
The fast base64 decoder used to switch to the slow one permanently when
it saw a whitespace or other garbage character. Since the most common
situation such characters may be encountered in is line-wrapped base64
data, a more profitable strategy is to decode a single 24-bit group with
the slow decoder and then continue running the fast algorithm.
PR-URL: https://github.com/nodejs/node/pull/12146
Ref: https://github.com/nodejs/node/issues/12114
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Also add a benchmark to compare both ways to create strings.
PR-URL: https://github.com/nodejs/node/pull/12170
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
A doc suggested in an error message is no longer the place to get the
information about required http benchmarkers. Update the error message
to point to the current location for the information.
PR-URL: https://github.com/nodejs/node/pull/12026
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/11930
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
This allows running a benchmark with two or more values for the same
config rather than just one or all of them, for example:
```
node benchmark/buffers/buffer-creation.js type=buffer() type=fast-alloc type=fast-alloc-fill
```
PR-URL: https://github.com/nodejs/node/pull/11819
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Brian White <mscdex@mscdex.net>
Add a space for minimal readability.
PR-URL: https://github.com/nodejs/node/pull/11925
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Make it call-site-cwd-independent.
PR-URL: https://github.com/nodejs/node/pull/11904
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This removes common.v8ForceOptimization calls from url and vm benchmark
files.
PR-URL: https://github.com/nodejs/node/pull/11908
Fixes: https://github.com/nodejs/node/issues/11895
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Including:
* Move async *stat() functions to FillStatsArray() now used by the
sync *stat() functions
* Avoid creating fs.Stats instances for implicit async/sync *stat()
calls used in various fs functions
* Store reference to Float64Array data on C++ side for easier/faster
access, instead of passing from JS to C++ on every async/sync *stat()
call
PR-URL: https://github.com/nodejs/node/pull/11665
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Add missing 'i=0' from for-loops from punycode and get-ciphers
benchmarks.
PR-URL: https://github.com/nodejs/node/pull/11720
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
This removes all instances of %OptimizeFunctionOnNextCall from common.js
and README.md
PR-URL: https://github.com/nodejs/node/pull/9615
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
