Instead of using kOnTimeout index to track a special list
processing function, just pass in a function to C++ at
startup that executes all handles and determines which
function to call.
This change improves the performance of unpooled timeouts
by roughly 20%, as well as makes the unref/ref processing
easier to follow.
PR-URL: https://github.com/nodejs/node/pull/18582
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Vladimir de Turckheim <vlad2t@hotmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
The current output uses JSON.stringify to escape the config values.
This switches to util.inspect to have a better readable output.
PR-URL: https://github.com/nodejs/node/pull/18597
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Andreas Madsen <amwebdk@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Initial set of fs.promises APIs with documentation and one
benchmark.
PR-URL: https://github.com/nodejs/node/pull/18297
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Benchmark comparing `util._extend()`, `Object.assign()`,
and the spread operator for object assignment.
`util._extend()` still wins currently.
PR-URL: https://github.com/nodejs/node/pull/18442
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Weijia Wang <starkwang@126.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Shorten the config name in check_invalid_header_char so it would
not result in long lines that make the benchmark result hard to read.
PR-URL: https://github.com/nodejs/node/pull/18452
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Weijia Wang <starkwang@126.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
PR-URL: https://github.com/nodejs/node/pull/18379
Reviewed-By: Jon Moss <me@jonathanmoss.me>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Problem:
Node implements fs.readFile as:
- a call to stat, then
- a C++ -> libuv request to read the entire file using the stat size
Why is this bad?
The effect is to place on the libuv threadpool a potentially-large
read request, occupying the libuv thread until it completes.
While readFile certainly requires buffering the entire file contents,
it can partition the read into smaller buffers
(as is done on other read paths)
along the way to avoid threadpool exhaustion.
If the file is relatively large or stored on a slow medium, reading
the entire file in one shot seems particularly harmful,
and presents a possible DoS vector.
Solution:
Partition the read into multiple smaller requests.
Considerations:
1. Correctness
I don't think partitioning the read like this raises
any additional risk of read-write races on the FS.
If the application is concurrently readFile'ing and modifying the file,
it will already see funny behavior. Though libuv uses preadv where
available, this doesn't guarantee read atomicity in the presence of
concurrent writes.
2. Performance
Downside: Partitioning means that a single large readFile will
require into many "out and back" requests to libuv,
introducing overhead.
Upside: In between each "out and back", other work pending on the
threadpool can take a turn.
In short, although partitioning will slow down a large request,
it will lead to better throughput if the threadpool is handling
more than one type of request.
Fixes: https://github.com/nodejs/node/issues/17047
PR-URL: https://github.com/nodejs/node/pull/17054
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Tiancheng "Timothy" Gu <timothygu99@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Long ago, V8 was much faster switching on string lengths than values.
That is no longer the case, so we can simplify a couple of methods.
PR-URL: https://github.com/nodejs/node/pull/18351
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Kyle Farnung <kfarnung@microsoft.com>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jon Moss <me@jonathanmoss.me>
This applies to all `async_hooks`, `dns`, `cluster`, `domain` and
`module` benchmarks.
PR-URL: https://github.com/nodejs/node/pull/18250
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
This is made to mirror the connection-bound net benchmarks.
PR-URL: https://github.com/nodejs/node/pull/17983
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
