Currently, while FIPS is not supported yet for this release there might
be an option to dynamically link against a FIPS compatible OpenSSL
version.
This commit fixes the compiler errors.
PR-URL: https://github.com/nodejs/node/pull/25412
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Remove the following properties:
- `preserveSymlinks`
- `preserveSymlinksMain`
- `experimentalModules`
- `userLoader`
- `experimentalVMModules`
- `experimentalREPLAwait`
- `exposeInternals`
We used to use them to pass cli option values from C++ into
JS, but now the values are obtained in JS land using
`require('internal/options').getOptionValue` instead so they
are unused.
Also removes `test/parallel/test-internal-modules-expose.js`
which tests `--expose-internals`.
We already have hundreds of tests depending on `--expose-internals`,
they are more than enough to test the functionality of the flag.
PR-URL: https://github.com/nodejs/node/pull/25463
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Instead of using `internalBinding('config')` which should be used
to carry information about build-time options, directly pass the
run-time cli options into bootstrap/loaders.js lexically via
function arguments.
PR-URL: https://github.com/nodejs/node/pull/25463
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Refael Ackermann <refack@gmail.com>
The idea is to allow the C++ layer to run arbitrary scripts
as the main script. This paves the way for
- cctest of the execution of Node.js instances
- Earlier handling of per-process CLI options that affect
execution modes (those usually do not make sense for the
embedders).
- Targets like mkcodecache or mksnapshot.
Also moves the handling of `_third_party_main.js` into C++.
PR-URL: https://github.com/nodejs/node/pull/25474
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Move `node::errno_string` into node_errors.h/cc and move it into
the `node:errors` namespace to reduce the size of the header.
It's not on any performance-critical code path so does not need
to be inlined.
PR-URL: https://github.com/nodejs/node/pull/25396
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
- Remove `NativeModule._source` - the compilation is now entirely
done in C++ and `process.binding('natives')` is implemented
directly in the binding loader so there is no need to store
additional source code strings.
- Instead of using an object as `NativeModule._cached` and insert
into it after compilation of each native module, simply prebuild
a JS map filled with all the native modules and infer the
state of compilation through `mod.loading`/`mod.loaded`.
- Rename `NativeModule.nonInternalExists` to
`NativeModule.canBeRequiredByUsers` and precompute that
property for all the native modules during bootstrap instead
of branching in every require call during runtime. This also fixes
the bug where `worker_threads` can be made available with
`--expose-internals`.
- Rename `NativeModule.requireForDeps` to
`NativeModule.requireWithFallbackInDeps`.
- Add a test to make sure we do not accidentally leak any module
to the global namespace.
PR-URL: https://github.com/nodejs/node/pull/25352
Reviewed-By: Anna Henningsen <anna@addaleax.net>
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>
Changes `SetupProcessObject` to `CreateProessObject` which creates
the process object from scratch and return it to `Environment::Start`
to be stored in the Environment object.
PR-URL: https://github.com/nodejs/node/pull/25397
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This makes the STL variant of `node::Init()` a bit more
suitable for inclusion in a proper embedder API, as errors
or other output are reported to the caller rather than
directly being printed, and the process is not exited
directly either.
PR-URL: https://github.com/nodejs/node/pull/25370
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Instead of exposing it in `lib/internal/buffer.js` after deleting
it from the binding and then do the initialization in
`lib/buffer.js`, which results in an implicit dependency on
the order in which these modules are loaded.
PR-URL: https://github.com/nodejs/node/pull/25292
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Having an experimental feature behind a flag makes change
if we are expecting significant breaking changes to its API.
Since the Worker API has been essentially stable since
its initial introduction, and no noticeable doubt about
possibly not keeping the feature around has been voiced,
removing the flag and thereby reducing the barrier to experimentation,
and consequently receiving feedback on the implementation,
seems like a good idea.
PR-URL: https://github.com/nodejs/node/pull/25361
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Yuta Hiroto <hello@hiroppy.me>
Reviewed-By: Shingo Inoue <leko.noor@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Tiancheng "Timothy" Gu <timothygu99@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Masashi Hirano <shisama07@gmail.com>
Reviewed-By: Weijia Wang <starkwang@126.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
So that it's easier to tell whether we are manipulating per-process
global states that may need to be treated with care to avoid races.
Also added comments about these variables and moved some of them
to a more suitable compilation unit:
- Move `v8_initialized` to `util.h` since it's only used in
`util.cc` and `node.cc`
- Rename `process_mutex` to `tty_mutex` and move it into
`node_errors.cc` since that's the only place it's used
to guard the tty.
- Move `per_process_opts_mutex` and `per_process_opts`
into `node_options.h` and rename them to
`per_process::cli_options[_mutex]`
- Rename `node_isolate[_mutex]` to `per_process::main_isolate[_mutex]`
PR-URL: https://github.com/nodejs/node/pull/25302
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Currently the following compiler warning is generated:
../src/node_task_queue.cc:93:12: warning:
unused variable 'isolate' [-Wunused-variable]
Isolate* isolate = env->isolate();
^
1 warning generated.
This commit removes the unused variable.
PR-URL: https://github.com/nodejs/node/pull/25368
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
- Rename `internalTickCallback` to `processTicksAndRejections`, make
sure it does not get called if it's not set in C++.
- Rename `emitPromiseRejectionWarnings` to `processPromiseRejections`
since it also emit events that are not warnings.
- Sets `SetPromiseRejectCallback` in the `Environment` constructor
to make sure it only gets called once per-isolate, and make
sure it does not get called if it's not set in C++.
- Wrap promise rejection callback initialization into
`listenForRejections()`.
- Add comments.
PR-URL: https://github.com/nodejs/node/pull/25200
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
- Wrap access to tickInfo fields in functions
- Rename `kHasScheduled` to `kHasTickScheduled` and
`kHasPromiseRejections` to `kHasRejectionToWarn` for clarity - note
the latter will be set to false if the rejection does not lead to
a warning so the previous description is not accurate.
- Set `kHasRejectionToWarn` in JS land of relying on C++ to use
an implict contract (return value of the promise rejection handler)
to set it, as the decision is made entirely in JS land.
- Destructure promise reject event constants.
PR-URL: https://github.com/nodejs/node/pull/25200
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Enable in-memory transfer of WASM modules without recompilation.
Previously, the serialization step worked, but deserialization failed
because we did not explicitly enable decoding inline WASM modules,
and so the message was not successfully received.
PR-URL: https://github.com/nodejs/node/pull/25314
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Since the API we expose through the `v8` module is Buffer-based,
we cannot transfer WASM modules directly. Instead, we enable
the V8-provided inline WASM (de)serialization for WASM modules.
PR-URL: https://github.com/nodejs/node/pull/25313
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Gus Caplan <me@gus.host>
Store the full information we have on a given `SharedArrayBuffer`,
and use the deleter provided by the JS engine to free the memory
when that is needed.
This fixes memory lifetime management for WASM buffers that are
passed through a `MessageChannel` (e.g. between threads).
PR-URL: https://github.com/nodejs/node/pull/25307
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Tiancheng "Timothy" Gu <timothygu99@gmail.com>
- Cache the `isolate` value between calls
- Introduce an overload for dealing with integers/numbers
- Use the vectored `v8::Array::New` constructor + `MaybeStackBuffer`
for faster array creation
PR-URL: https://github.com/nodejs/node/pull/25288
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
This commit fixes the following warning:
./src/cares_wrap.cc:1268:5: warning: comparison of integers of
different signs: 'uint32_t' (aka 'unsigned int') and 'int'
[-Wsign-compare]
CHECK_EQ(ret->Length(), a_count + aaaa_count);
PR-URL: https://github.com/nodejs/node/pull/25230
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
This patch:
- Moves `tryGetCwd`, `evalScript` and `fatalException` from
`bootstrap/node.js` into `process/execution.js` so that
they do have to be passed into the worker thread
setup function, instead the worker code can require them
when necessary.
- Moves `setUncaughtExceptionCaptureCallback` and
`hasUncaughtExceptionCaptureCallback` along with the two
global state `exceptionHandlerState` and
`shouldAbortOnUncaughtToggle` info `process.execution.js`
as those are only used by the fatalException and these
two accessors as one self-contained unit.
PR-URL: https://github.com/nodejs/node/pull/25199
Reviewed-By: James M Snell <jasnell@gmail.com>
cjihrig