Backport async hooks to v8.x by AndreasMadsen · Pull Request #18179 · nodejs/node

yhwang

and others added 22 commits

January 16, 2018 13:00

Previously, built-in modules are registered before main() via
__attribute__((constructor)) mechanism in GCC and similiar
mechanism in MSVC. This causes some issues when node is built as
static library. Calling module registration function for built-in
modules in node::Init() helps to avoid the issues.

Signed-off-by: Yihong Wang <yh.wang@ibm.com>
PR-URL: #16565
Refs: #14986 (comment)
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

Use std::unique_ptr instead of raw pointers for the
tracing_agent_ in node.cc. This makes ownership clearer and we
don't risk a memory leak.

PR-URL: #17012
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>

This will allow trace event to record timing information for all
asynchronous operations that are observed by async_hooks.

PR-URL: #15538
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>

In cases where libraries create AsyncResources which may be emitting
more events depending on usage, the only way to ensure that destroy is
called properly is by calling it when the resource gets garbage
collected.

Fixes: #16153
PR-URL: #16998
Fixes: #16153
Reviewed-By: Andreas Madsen <amwebdk@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

Commit d217b28 ("async_hooks: add trace events to async_hooks")
used `NODE_MODULE_CONTEXT_AWARE_BUILTIN()` instead.

After commit 8680bb9 ("src: explicitly register built-in modules")
it no longer works for static library builds so remove it.

PR-URL: #17071
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>

This commit renames async-wrap to async_wrap for consitency with other
c++ source files.

PR-URL: #17022
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>

This removes the `process._needImmediateCallback` property
and its semantics of having a 1/0 switch that tells C++ whether
immediates are currently scheduled.

Instead, a counter keeping track of all immediates is created,
that can be increased on `setImmediate()` or decreased when an
immediate is run or cleared.

This is faster, because rather than reading/writing a C++ getter,
this operation can be performed as a direct memory read/write via
a typed array. The only C++ call that is left to make is
activating the native handles upon creation of the first
`Immediate` after the queue is empty.

One other (good!) side-effect is that `immediate._destroyed` now
reliably tells whether an `immediate` is still scheduled to run or not.

Also, as a nice extra, this should make it easier to implement
an internal variant of `setImmediate` for C++ that piggybacks
off the same mechanism, which should be useful at least for
async hooks and HTTP/2.

Benchmark results:

    $ ./node benchmark/compare.js --new ./node --old ./node-master-1b093cb93df0 --runs 10 --filter immediate timers | Rscript benchmark/compare.R
    [00:08:53|% 100| 4/4 files | 20/20 runs | 1/1 configs]: Done
                                                         improvement confidence      p.value
     timers/immediate.js type="breadth" thousands=2000      25.61 %         ** 1.432301e-03
     timers/immediate.js type="breadth1" thousands=2000      7.66 %            1.320233e-01
     timers/immediate.js type="breadth4" thousands=2000      4.61 %            5.669053e-01
     timers/immediate.js type="clear" thousands=2000       311.40 %        *** 3.896291e-07
     timers/immediate.js type="depth" thousands=2000        17.54 %         ** 9.755389e-03
     timers/immediate.js type="depth1" thousands=2000       17.09 %        *** 7.176229e-04
     timers/set-immediate-breadth-args.js millions=5        10.63 %          * 4.250034e-02
     timers/set-immediate-breadth.js millions=10            20.62 %        *** 9.150439e-07
     timers/set-immediate-depth-args.js millions=10         17.97 %        *** 6.819135e-10

PR-URL: #17064
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

PR-URL: #17117
Reviewed-By: James M Snell <jasnell@gmail.com>

PR-URL: #17117
Reviewed-By: James M Snell <jasnell@gmail.com>

async_hooks emits trace_events. This adds the executionAsyncId to the
init events. In theory this could be inferred from the before and after
events but this is much simpler and doesn't require knowledge of all
events.

PR-URL: #17196
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

Adds `TCPSERVERWRAP` and `PIPESERVERWRAP` as provider types. This
makes it possible to distinguish servers from connections.

PR-URL: #17157
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>

rename initTriggerId to defaultTriggerAsyncId such it matches the rest
of our naming.

PR-URL: #17273
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

When context is missing the executionAsyncId will be zero. For the
default triggerAsyncId the zero value was used to default to the
executionAsyncId. While this was not technically wrong because the
functions are different themself, it poorly separated the two concepts.

PR-URL: #17273
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

Previously the getter would mutate the kDefaultTriggerAsncId value. This
refactor changes the setter to bind the current kDefaultTriggerAsncId to
a scope, such that the getter doesn't have to mutate its own value.

PR-URL: #17273
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

PR-URL: #17757
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Jon Moss <me@jonathanmoss.me>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: vdeturckheim <vlad2t@hotmail.com>

SetupHooks is only available via `process.binding('async_wrap')`, so
there's no reason it shouldn't be called with the appropriate arguments,
since it is an internal-only function. The only place this function is
used is `lib/internal/async_hooks.js`.

PR-URL: #17832
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>

- Communicate the current async stack length through a
  typed array field rather than a native binding method
- Add a new fixed-size `async_ids_fast_stack` typed array
  that contains the async ID stack up to a fixed limit.
  This increases performance noticeably, since most of the time
  the async ID stack will not be more than a handful of
  levels deep.
- Make the JS `pushAsyncIds()` and `popAsyncIds()` functions
  do the same thing as the native ones if the fast path
  is applicable.

Benchmarks:

    $ ./node benchmark/compare.js --new ./node --old ./node-master --runs 10 --filter next-tick process | Rscript benchmark/compare.R
    [00:03:25|% 100| 6/6 files | 20/20 runs | 1/1 configs]: Done
                                                   improvement confidence      p.value
     process/next-tick-breadth-args.js millions=4     19.72 %        *** 3.013913e-06
     process/next-tick-breadth.js millions=4          27.33 %        *** 5.847983e-11
     process/next-tick-depth-args.js millions=12      40.08 %        *** 1.237127e-13
     process/next-tick-depth.js millions=12           77.27 %        *** 1.413290e-11
     process/next-tick-exec-args.js millions=5        13.58 %        *** 1.245180e-07
     process/next-tick-exec.js millions=5             16.80 %        *** 2.961386e-07

PR-URL: #17780
Reviewed-By: James M Snell <jasnell@gmail.com>

PR-URL: #18005
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>

The existing version of defaultTriggerAsyncIdScope creates an Array
for the callback's arguments which is highly inefficient. Instead,
use rest syntax and allow V8 to do that work for us. This yields
roughly 2x performance for this particular function.

PR-URL: #18004
Reviewed-By: Andreas Madsen <amwebdk@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

This allows more easy tracking of where HTTP requests come from. Before
this change the HTTPParser would have the HTTPServer as the
triggerAsyncId.

The HTTPParser will still have the executionAsyncId set to the HTTP
Server so that information is still directly available. Indirectly, the
TCP socket itself also has its triggerAsyncId set to the TCP Server.

PR-URL: #18003
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jon Moss <me@jonathanmoss.me>
Reviewed-By: Daijiro Wachi <daijiro.wachi@gmail.com>

If IPv6 is not supported on a machine, the IPv6 handle will first be
created, this will then fail and default to an IPv4 handle. This causes
the graph to change, as there now is an extra handle.

PR-URL: #18143
Fixes: #18003
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

This was referenced

Jan 16, 2018

gibfahn pushed a commit to gibfahn/node that referenced this pull request

Jan 17, 2018

- Communicate the current async stack length through a
  typed array field rather than a native binding method
- Add a new fixed-size `async_ids_fast_stack` typed array
  that contains the async ID stack up to a fixed limit.
  This increases performance noticeably, since most of the time
  the async ID stack will not be more than a handful of
  levels deep.
- Make the JS `pushAsyncIds()` and `popAsyncIds()` functions
  do the same thing as the native ones if the fast path
  is applicable.

Benchmarks:

    $ ./node benchmark/compare.js --new ./node --old ./node-master --runs 10 --filter next-tick process | Rscript benchmark/compare.R
    [00:03:25|% 100| 6/6 files | 20/20 runs | 1/1 configs]: Done
                                                   improvement confidence      p.value
     process/next-tick-breadth-args.js millions=4     19.72 %        *** 3.013913e-06
     process/next-tick-breadth.js millions=4          27.33 %        *** 5.847983e-11
     process/next-tick-depth-args.js millions=12      40.08 %        *** 1.237127e-13
     process/next-tick-depth.js millions=12           77.27 %        *** 1.413290e-11
     process/next-tick-exec-args.js millions=5        13.58 %        *** 1.245180e-07
     process/next-tick-exec.js millions=5             16.80 %        *** 2.961386e-07

PR-URL: nodejs#17780
Backport-PR-URL: nodejs#18179
Reviewed-By: James M Snell <jasnell@gmail.com>

gibfahn pushed a commit to gibfahn/node that referenced this pull request

Jan 17, 2018

PR-URL: nodejs#18005
Backport-PR-URL: nodejs#18179
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>

gibfahn pushed a commit to gibfahn/node that referenced this pull request

Jan 17, 2018

The existing version of defaultTriggerAsyncIdScope creates an Array
for the callback's arguments which is highly inefficient. Instead,
use rest syntax and allow V8 to do that work for us. This yields
roughly 2x performance for this particular function.

PR-URL: nodejs#18004
Backport-PR-URL: nodejs#18179
Reviewed-By: Andreas Madsen <amwebdk@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

gibfahn pushed a commit to gibfahn/node that referenced this pull request

Jan 17, 2018

This allows more easy tracking of where HTTP requests come from. Before
this change the HTTPParser would have the HTTPServer as the
triggerAsyncId.

The HTTPParser will still have the executionAsyncId set to the HTTP
Server so that information is still directly available. Indirectly, the
TCP socket itself also has its triggerAsyncId set to the TCP Server.

PR-URL: nodejs#18003
Backport-PR-URL: nodejs#18179
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jon Moss <me@jonathanmoss.me>
Reviewed-By: Daijiro Wachi <daijiro.wachi@gmail.com>

gibfahn pushed a commit to gibfahn/node that referenced this pull request

Jan 17, 2018

If IPv6 is not supported on a machine, the IPv6 handle will first be
created, this will then fail and default to an IPv4 handle. This causes
the graph to change, as there now is an extra handle.

PR-URL: nodejs#18143
Backport-PR-URL: nodejs#18179
Fixes: nodejs#18003
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: James M Snell <jasnell@gmail.com>