| Age | Commit message (Collapse) | Author |
|---|
|
Closing in the Agent destructor is too late, because that happens
when the Environment is destroyed, not when libuv handles are closed.
This fixes a situation in which the same libuv loop is re-used for
multiple Environment instances sequentially, e.g. in our cctest.
PR-URL: https://github.com/nodejs/node/pull/30612
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
|
|
Run inspector cleanup code on Environment teardown.
This is part of a series of changes to make embedding easier, by
requiring fewer internal methods to build a fully functioning
Node.js instance.
PR-URL: https://github.com/nodejs/node/pull/30229
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Shelley Vohr <codebytere@gmail.com>
|
|
Turn tasks scheduled on the `v8::Isolate` or on the given platform
into no-ops if the underlying `MainThreadInterface` has gone away
before the task could be run (which would happen when the
`Environment` instance and with it the `inspector::Agent` instance
are destroyed).
This addresses an issue that Electron has been having with
inspector support, and generally just seems like the right thing
to do, as we may not fully be in control of the relative timing
of Environment teardown, platform tasksexecution, and the
execution of `RequestInterrupt()` callbacks (although
the former two always happen in the same order in our own code).
PR-URL: https://github.com/nodejs/node/pull/30031
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Shelley Vohr <codebytere@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
|
|
This API is designed to enable worker threads use Inspector protocol
on main thread (and other workers through NodeWorker domain).
Note that worker can cause dead lock by suspending itself. I will
work on a new API that will allow workers to be hidden from the
inspector.
Fixes: https://github.com/nodejs/node/issues/28828
PR-URL: https://github.com/nodejs/node/pull/28870
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/29076
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
|
|
Main thread (the one that WS endpoint connects to) should be able
to report all workers.
PR-URL: https://github.com/nodejs/node/pull/28872
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
|
|
Fixes: https://github.com/nodejs/node/issues/28741
PR-URL: https://github.com/nodejs/node/pull/28756
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anto Aravinth <anto.aravinth.cse@gmail.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
Fixes: https://github.com/nodejs/node/issues/28528
PR-URL: https://github.com/nodejs/node/pull/28613
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
This is a cleanup, allowing for a better separation of concerns.
PR-URL: https://github.com/nodejs/node/pull/28526
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
|
|
Previously the enhancement were done right after emitting
`'uncaughtException'`, which meant by the time we knew the
exception was fatal in C++, the error.stack had already been
patched.
This patch moves those routines to be called later during the
fatal exception handling, and split them into two stages:
before and after the inspector is notified by the invocation of
`V8Inspector::exceptionThrown`. We now expand the stack to include
additional informations about unhandled 'error' events before
the inspector is notified, but delay the highlighting of the
frames until after the inspector is notified, so that the
ANSI escape sequences won't show up in the inspector console.
PR-URL: https://github.com/nodejs/node/pull/28308
Fixes: https://github.com/nodejs/node/issues/28287
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/28036
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Yongsheng Zhang <zyszys98@gmail.com>
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
|
|
This flag specifies how inspector websocket url should be reported.
Tthre options are supported:
- stderr - reports websocket as a message to stderr,
- http - exposes /json/list endpoint that contains inspector websocket
url,
- binding - require('inspector').url().
Related discussion: https://github.com/nodejs/diagnostics/issues/303
PR-URL: https://github.com/nodejs/node/pull/27741
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
|
|
NodeRuntime domain was introduced to give inspector client way to
fetch captured information before Node process is gone. We need
similar capability for work.
With current protocol inspector client can force worker to wait
on start by passing waitForDebuggerOnStart flag to NodeWorker.enable
method. So client has some time to setup environment, e.g. start
profiler. At the same time there is no way to prevent worker from
being terminated. So we can start capturing profile but we can not
reliably get captured data back.
This PR implemented NodeRuntime.notifyWhenWaitingForDisconnect
method for worker. When NodeRuntime.waitingForDisconnect notification
is enabled, worker will wait for explicit NodeWorker.detach call.
With this PR worker tooling story is nicely aligned with main thread
tooling story. The only difference is that main thread by default is
waiting for disconnect but worker thread is not waiting.
Issue: https://github.com/nodejs/node/issues/27677
PR-URL: https://github.com/nodejs/node/pull/27706
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
|
|
PTHREAD_STACK_MIN is 2 KB with musl, which is too small to safely
receive signals. PTHREAD_STACK_MIN + MINSIGSTKSZ is 8 KB on arm64,
which is the musl architecture with the biggest MINSIGSTKSZ so let's
use that as a lower bound and let's quadruple it just in case.
PR-URL: https://github.com/nodejs/node/pull/27855
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
|
|
Inline headers should only be included into the .cc files that use them.
PR-URL: https://github.com/nodejs/node/pull/27755
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
This uses SIGSEGV handlers to catch WASM out of bound (OOB) memory
accesses instead of inserting OOB checks inline, resulting in a 25%-30%
speed increase.
Note that installing a custom SIGSEGV handler will break this, resulting
in potentially scary behaviour.
Refs: https://github.com/nodejs/node/issues/14927
PR-URL: https://github.com/nodejs/node/pull/27246
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
Historically Node process sends Runtime.executionContextDestroyed
with main context as argument when it is finished.
This approach has some disadvantages. V8 prevents running some
protocol command on destroyed contexts, e.g. Runtime.evaluate
will return an error or Debugger.enable won't return a list of
scripts.
Both command might be useful for different tools, e.g. tool runs
Profiler.startPreciseCoverage and at the end of node process it
would like to get list of all scripts to match data to source code.
Or some tooling frontend would like to provide capabilities to run
commands in console when node process is finished to allow user to
inspect state of the program at exit.
This PR adds new domain: NodeRuntime. After
NodeRuntime.notifyWhenWaitingForDisconnect is enabled by at least one
client, node will send NodeRuntime.waitingForDebuggerToDisconnect
event instead of Runtime.executionContextDestroyed. Based on this
signal any protocol client can capture all required information and
then disconnect its session.
PR-URL: https://github.com/nodejs/node/pull/27600
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
|
|
Its intended that *-inl.h header files are only included into the src
files that call the inline methods. Explicitly include it into the files
that need it.
PR-URL: https://github.com/nodejs/node/pull/27631
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
|
|
1. Do not rely on a string comparison to identify when
the frontend is ready to run and override a callback instead.
2. Remove unused boolean flag.
PR-URL: https://github.com/nodejs/node/pull/27591
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
|
|
Fix unnecessary inclusion of node_options-inl.h into env.h via
inspector_agent.h, causing almost all of src/ to recompile on any change
to the options parser. Its intended that *-inl.h header files are only
included into the src files that call the inline methods.
PR-URL: https://github.com/nodejs/node/pull/27538
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
`v8::Global` is essentially a nicer variant of `node::Persistent` that,
in addition to reset-on-destroy, also implements move semantics.
This commit makes the necessary replacements, removes
`node::Persistent` and (now-)unnecessary inclusions of the
`node_persistent.h` header, and makes some of the functions that
take Persistents as arguments more generic so that they work with all
`v8::PersistentBase` flavours.
PR-URL: https://github.com/nodejs/node/pull/27287
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/27264
Reviewed-By: Refael Ackermann <refack@gmail.com>
|
|
FromJust() is often used not for its return value, but for its
side-effects. In these cases, Check() exists, and is more clear as to
the intent. From its comment:
To be used, where the actual value of the Maybe is not needed, like
Object::Set.
See: https://github.com/nodejs/node/pull/26929/files#r269256335
PR-URL: https://github.com/nodejs/node/pull/27162
Reviewed-By: Anatoli Papirovski <apapirovski@mac.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Yongsheng Zhang <zyszys98@gmail.com>
|
|
This patch refactors `AppendExceptionLine` and `PrintSyncTrace`
to reuse the error formatting logic and use them to print
uncaught error in ``ToggleAsyncHook`
PR-URL: https://github.com/nodejs/node/pull/26859
Refs: https://github.com/nodejs/node/issues/26798
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
|
|
If Debugger.setAsyncCallStackDepth is sent during bootstrap,
we cannot immediately call into JS to enable the hooks, which could
interrupt the JS execution of bootstrap. So instead we save the
notification in the inspector agent if it's sent in the middle of
bootstrap, and process the notification later here.
Refs: https://github.com/nodejs/node/issues/26798
PR-URL: https://github.com/nodejs/node/pull/26935
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
|
|
Instead of patching process._breakFirstLine to inform the JS land
to wait for the debugger, check that the JS land has not yet
serialized the options and then patch the debug options from C++.
The changes will be carried into JS later during option serialization.
PR-URL: https://github.com/nodejs/node/pull/26602
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
|
|
PR-URL: https://github.com/nodejs/node/pull/26306
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
|
|
Fixes: https://github.com/nodejs/node/issues/25808
PR-URL: https://github.com/nodejs/node/pull/26303
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Refael Ackermann <refack@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/26159
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/26103
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
It is not obvious that timer handles are cleaned up properly
when the current `Environment` exits, nor that the `Environment`
knows to keep track of the closing handles.
This change may not be necessary, because timer handles
close without non-trivial delay (i.e. at the end of the current
event loop term), and JS-based inspector sessions (which are
the only ones we can easily test) are destroyed when cleaning up,
closing the timers as a result. I don’t know what happens
for other kinds of inspector sessions, though.
PR-URL: https://github.com/nodejs/node/pull/26088
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
|
|
Refs: https://github.com/nodejs/node/issues/24016
PR-URL: https://github.com/nodejs/node/pull/26011
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/25916
Refs: https://github.com/nodejs/node/pull/20914
Reviewed-By: Masashi Hirano <shisama07@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
|
|
Embedders may want to control whether a Node.js instance
controls the current process, similar to what we currently
have with `Worker`s.
Previously, the `isMainThread` flag had a bit of a double usage,
both for indicating whether we are (not) running a Worker and
whether we can modify per-process state.
PR-URL: https://github.com/nodejs/node/pull/25881
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
Add a check to make sure start_io_thread_async is not
accidentally re-used or used when uninitialized.
(This is a bit of an odd check imo, but it helped me figure
out a real issue and it might do so again, so… why not?)
PR-URL: https://github.com/nodejs/node/pull/25777
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
Fixes: https://github.com/nodejs/node/issues/23185
PR-URL: https://github.com/nodejs/node/pull/24814
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
- Instead of creating the console extensions eagerly during bootstrap
and storing them on an object, wrap the code into a function to be
called during `installAdditionalCommandLineAPI` only when the
extensions are actually needed, which may not even happen if the
user do not use the console in an inspector session, and does not
need to happen during bootstrap unconditionally.
- Simplify the console methods wrapping and move the `consoleFromVM`
storage to `internal/util/inspector.js`
PR-URL: https://github.com/nodejs/node/pull/25450
Reviewed-By: John-David Dalton <john.david.dalton@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
|
|
Move the C++ `process.emit` and `process.emitWarning` methods
from `node.cc` into into `node_process_events.cc`, and
reuse the implementation in other places that need to do
`process.emit` in C++.
PR-URL: https://github.com/nodejs/node/pull/25397
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
Instead of in node_internals.h. Also move process property
accessors that are not reused into node_process_object.cc
and make them static.
PR-URL: https://github.com/nodejs/node/pull/25397
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
Previously we create env->inspector_console_api_object() when
`process.binding('inspector')` is called, which may be too late
if the inspector console is used before the first call to
`process.binding('inspector')` - that is possible when
using `--inspect-brk-node`. Setting a breakpoint and using the
inspector console before that would crash the process.
This patch moves the initialization of the console API object to
the point when Environment is initialized so that
`installAdditionalCommandLineAPI()` can be essentially a noop
if we use the inspector console before the inspector binding
is initialized instead of crashing on an empty object.
PR-URL: https://github.com/nodejs/node/pull/24906
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
|
|
The V8 platform's CallOnForegroundThread() method is deprecated.
This commit replaces its use with GetForegroundTaskRunner()
functionality instead.
PR-URL: https://github.com/nodejs/node/pull/24925
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
Instead of using a shared pointer of the entire debug option set,
pass the parsed debug option to inspector classes by value because
they are set once the CLI argument parsing is done. Add another shared
pointer to HostPort being used by the inspector server, which is copied
from the one in the debug options initially. The port of the shared
HostPort is 9229 by default and can be specified as 0 initially but
will be set to the actual port of the server once it starts listening.
This makes the shared state clearer and makes it possible to use
`require('internal/options')` in JS land to query the CLI options
instead of using `process._breakFirstLine` and other underscored
properties of `process` since we are now certain that these
values should not be altered once the parsing is done and can be
passed around in copies without locks.
PR-URL: https://github.com/nodejs/node/pull/24772
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
In recent gcc, -Wrestrict warns when an argument passed to a
restrict-qualified parameter aliases with another argument.
PR-URL: https://github.com/nodejs/node/pull/24810
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/24427
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
|
|
Ref: https://github.com/nodejs/node/commit/283a967e356311a467113eea450a81827d43c969
PR-URL: https://github.com/nodejs/node/pull/24132
Refs: https://github.com/nodejs/node/commit/283a967e356311a467113eea450a81827d43c969
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
|
|
Move the following code into a new node_errors.cc file and
declare them in node_errors.h for clarity and make it possible
to include them with node_errors.h.
- AppendExceptionLine()
- DecorateErrorStack()
- FatalError()
- OnFatalError()
- PrintErrorString()
- FatalException()
- ReportException()
- FatalTryCatch
And move the following definitions (declared elsewhere than
node_errors.h) to node_errors.cc:
- Abort() (in util.h)
- Assert() (in util.h)
PR-URL: https://github.com/nodejs/node/pull/24058
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
|
|
Introduce a NodeTarget inspector domain modelled after ChromeDevTools
Target domain. It notifies inspector frontend attached to a main V8
isolate when workers are starting and allows passing messages to
inspectors on their isolates. All inspector functionality is enabled on
worker isolates.
PR-URL: https://github.com/nodejs/node/pull/21364
Reviewed-By: Aleksei Koziatinskii <ak239spb@gmail.com>
Reviewed-By: Jan Krems <jan.krems@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
|
|
PR-URL: https://github.com/nodejs/node/pull/22769
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
|
This method is required by inspector to report normalized urls over
the protocol.
Fixes https://github.com/nodejs/node/issues/22223
PR-URL: https://github.com/nodejs/node/pull/22251
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Tiancheng "Timothy" Gu <timothygu99@gmail.com>
|
|
PR-URL: https://github.com/nodejs/node/pull/22531
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
|
