The Windows ETW code is not written to be compatible with multi
threading, and in particular it relies on global state like a
single static `uv_async_t`. Adding that to multiple threads
would corrupt the corresponding loops' handle queues.
This addresses the flakiness of at least
`test-worker-exit-code` and very likely other flaky tests that
relate to Worker threads on Windows as well.
Fixes: https://github.com/nodejs/node/issues/25847
Fixes: https://github.com/nodejs/node/issues/25702
Fixes: https://github.com/nodejs/node/issues/24005
Fixes: https://github.com/nodejs/node/issues/23873
PR-URL: https://github.com/nodejs/node/pull/25907
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Use MaybeLocal::ToLocal and don't crash X509ToObject on error.
PR-URL: https://github.com/nodejs/node/pull/25717
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
EC_KEY_key2buf returns an OPENSSL_malloc'd pointer so it shouldn't be
passed into Buffer::New, which expect a libc malloc'd pointer. Instead,
factor out the ECDH::GetPublicKey code which uses EC_POINT_point2oct.
This preserves the existing behavior where encoding failures are
silently ignored, but it is probably safe to CHECK fail them instead.
PR-URL: https://github.com/nodejs/node/pull/25717
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Improve the material in the Collaborator Guide on unintended breaking
changes.
PR-URL: https://github.com/nodejs/node/pull/25887
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anto Aravinth <anto.aravinth.cse@gmail.com>
OpenSSL is packaging its git and travis configuration files. Remove
them, Node.js has its own.
PR-URL: https://github.com/nodejs/node/pull/25689
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Ujjwal Sharma <usharma1998@gmail.com>
Libuv does not guarantee that handles have their close
callbacks called in the order in which they were added
(and in fact, currently calls them in reverse order).
This patch ensures that the `flush_signal_` handle
is no longer in use (i.e. its close callback has already
been run) when we signal to the main thread that
`~NodeTraceBuffer` may be destroyed.
The same applies for `~NodeTraceWriter`.
Credit for debugging goes to Gireesh Punathil.
Fixes: https://github.com/nodejs/node/issues/25512
Co-authored-by: Gireesh Punathil <gpunathi@in.ibm.com>
PR-URL: https://github.com/nodejs/node/pull/25896
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Don't throw on invalid property access if options is not provided, and
ensure callback is a function.
PR-URL: https://github.com/nodejs/node/pull/25876
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Signed-off-by: gengjiawen <technicalcute@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/25732
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Signed-off-by: gengjiawen <technicalcute@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/25820
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, this test fails when configured --without-ssl:
=== release test-worker-cleanexit-with-moduleload ===
Path: parallel/test-worker-cleanexit-with-moduleload
events.js:173
throw er; // Unhandled 'error' event
^
internal/util.js:101
throw new ERR_NO_CRYPTO();
^
Error [ERR_NO_CRYPTO]:
Node.js is not compiled with OpenSSL crypto support
This commit as a check for crypto so that this test is skipped if there
is no crypto support.
PR-URL: https://github.com/nodejs/node/pull/25811
Reviewed-By: Gireesh Punathil <gpunathi@in.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
For `globals`, booleans are deprecated in favor strings
`'readable'`/`'writeable'`.
PR-URL: https://github.com/nodejs/node/pull/25877
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Yuta Hiroto <hello@hiroppy.me>
Version 2.1.0 of dmn will get rid of .map and .ts files.
PR-URL: https://github.com/nodejs/node/pull/25877
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Yuta Hiroto <hello@hiroppy.me>
There are no non-internal builtin modules left, so this
should be safe to remove to a large degree.
PR-URL: https://github.com/nodejs/node/pull/25829
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anto Aravinth <anto.aravinth.cse@gmail.com>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Report the actual source code when running with `--eval` and
`--inspect-brk`, by telling the vm module to break on the
first line of the script being executed rather than wrapping
the source code in a function.
PR-URL: https://github.com/nodejs/node/pull/25832
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
This just makes the code a bit more obvious (subjectively).
PR-URL: https://github.com/nodejs/node/pull/25869
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Files in benchmark/assert/* were sometimes using trailing commas for
multi-line objects and sometimes not, mixing the approaches in the same
file sometimes. Standardize these files to always use trailing commas in
multi-line objects.
Additionally, remove some unnecessary line-wrapping (so that there are
fewer multi-line objects).
PR-URL: https://github.com/nodejs/node/pull/25865
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
The prerequisistes for test-all-suites were running some tests
themselves. When one of those tests failed during a coverage run, it
resulted in artificially low coverage. Fix the prerequisites to only
build stuff, not test.
PR-URL: https://github.com/nodejs/node/pull/25892
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
PR-URL: https://github.com/nodejs/node/pull/25855
Refs: https://github.com/nodejs/node/pull/25755
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Anto Aravinth <anto.aravinth.cse@gmail.com>
Remove bullet points about how breaking changes are handled in Current
and LTS branches.
* This information is covered in
https://github.com/nodejs/Release#release-plan.
* Having it here and in the above link means that the two may get out of
sync, resulting in confusion.
* The above link appears later in the doc in the LTS section.
* Most Collaborators should not land *anything* in LTS branches and will
almost never have a reason to land things into Current branches. They
land stuff on the master branch. Adding this material here is
confusing because it implies that Collaborators do land stuff on those
branches. Save talk about those branches and how they're managed for
later in the LTS section where it can be made clear that most
Collaborators should not be landing code there anyway.
PR-URL: https://github.com/nodejs/node/pull/25842
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
OpenSSL has supported async notification of sessions and tickets since
1.1.0 using SSL_CTX_sess_set_new_cb(), for all versions of TLS. Using
the async API is optional for TLS1.2 and below, but for TLS1.3 it will
be mandatory. Future-proof applications should start to use async
notification immediately. In the future, for TLS1.3, applications that
don't use the async API will silently, but gracefully, fail to resume
sessions and instead do a full handshake.
See: https://wiki.openssl.org/index.php/TLS1.3#Sessions
PR-URL: https://github.com/nodejs/node/pull/25831
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
There is currently no Makefile target that runs every test suite. This
adds one.
PR-URL: https://github.com/nodejs/node/pull/25799
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Signed-off-by: gengjiawen <technicalcute@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/25790
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This has already been practically end-of-life since `node --debug`
alone would exit the process. This patch drops support of
`node --inspect --debug-brk` as well.
`node --inspect --debug-brk` has been deprecated since v8,
it has been maintained so that vendors can target Node.js
v6 and above without detecting versions.
The support of `--inspect`, which starts from v6, will reach
end-of-life in April 2019, it should be safe to drop the support
of `--inspect --debug-brk` altogether in v12.
Also removes `process._deprecatedDebugBrk`
PR-URL: https://github.com/nodejs/node/pull/25828
Refs: https://github.com/nodejs/node/pull/12949
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Moves the exit of `--debug` and `--debug-brk` earlier, that is,
after the option parsing is done in the C++ land.
Also removes `process._invalidDebug`.
PR-URL: https://github.com/nodejs/node/pull/25828
Refs: https://github.com/nodejs/node/pull/12949
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Exclude async-hooks/test-callback-error as it seems
to fail consistently when run with coverage on
the docker ubuntu16 machines we plan to run the
coverage sanity test on.
PR-URL: https://github.com/nodejs/node/pull/25833
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Ben Coe <bencoe@gmail.com>
This patch moves the part in the report initialization
that mutates the process object into bootstrap/node.js
so it's easier to tell the side effect of the initialization
on the global state during bootstrap.
PR-URL: https://github.com/nodejs/node/pull/25821
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Instead of using process.config.variables.v8_enable_inspector
to detect whether inspector is enabled in the build.
PR-URL: https://github.com/nodejs/node/pull/25819
Refs: https://github.com/nodejs/node/issues/25343
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This patches changes the `safe_globals` internal module into a
script that gets run during bootstrap and saves JavaScript builtins
(primordials) into an object that is available for all other builtin
modules to access lexically later.
PR-URL: https://github.com/nodejs/node/pull/25816
Refs: https://github.com/nodejs/node/issues/18795
Reviewed-By: Bradley Farias <bradley.meck@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Gus Caplan <me@gus.host>
Unless configured, V8 defaults to limiting the max heaps size to 700 MB
or 1400MB on 32 and 64-bit platforms respectively. This default is
based on the browser use-cases and doesn't make a lot of sense
generally. This change properly configures the heap size based on
actual available memory.
PR-URL: https://github.com/nodejs/node/pull/25576
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Yang Guo <yangguo@chromium.org>
Reviewed-By: Ujjwal Sharma <usharma1998@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Report release metadata separately from `componentVersions`.
Test `componentVersions` and `release` values in the report.
PR-URL: https://github.com/nodejs/node/pull/25826
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
- Assign thread IDs to `Environment` instances, rather than Workers.
This is more embedder-friendly than the current system, in which
all “main threads” (if there are multiple ones) would get the
id `0`.
- Because that means that `isMainThread === (threadId === 0)` no longer
holds, refactor `isMainThread` into a separate entity. Implement it
in a way that allows for future extensibility, because we use
`isMainThread` in multiple different ways (determining whether there
is a parent thread; determining whether the current thread has control
of the current process; etc.).
PR-URL: https://github.com/nodejs/node/pull/25796
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Denys Otrishko <shishugi@gmail.com>
Instead of returning a generic 400 response when the
max header size is reached, return a 431 Request Header
Fields Too Large.
This is a semver-major because it changes the HTTP
status code for requests that trigger the header
overflow error.
PR-URL: https://github.com/nodejs/node/pull/25605
Fixes: https://github.com/nodejs/node/issues/25528
Refs: https://tools.ietf.org/html/rfc6585#section-5
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
If COV_ENFORCE_THRESHOLD is set, tests run in coverage mode will
exit with an error, if line coverage is below the percentage
threshold.
PR-URL: https://github.com/nodejs/node/pull/25675
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Throw a `DataCloneError` exception when encountering duplicate
`ArrayBuffer`s or `MessagePort`s in the transfer list.
Fixes: https://github.com/nodejs/node/issues/25786
PR-URL: https://github.com/nodejs/node/pull/25815
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, there are a number of compiler warnings generated when
building the addons on Linux, for example:
make[1]: Entering directory '/node/test/addons/zlib-binding/build'
CXX(target) Release/obj.target/binding/binding.o
SOLINK_MODULE(target) Release/obj.target/binding.node
COPY Release/binding.node
make[1]: Leaving directory '/node/test/addons/zlib-binding/build'
In file included from ../binding.cc:1:
/node/src/node.h:515:51: warning:
cast between incompatible function types from
'void (*)(v8::Local<v8::Object>,
v8::Local<v8::Value>,
v8::Local<v8::Context>)' to
'node::addon_context_register_func' {aka
'void (*)(v8::Local<v8::Object>,
v8::Local<v8::Value>,
v8::Local<v8::Context>,
void*)'} [-Wcast-function-type]
(node::addon_context_register_func) (regfunc), \
^
/node/src/node.h:533:3:
note: in expansion of macro 'NODE_MODULE_CONTEXT_AWARE_X'
NODE_MODULE_CONTEXT_AWARE_X(modname, regfunc, NULL, 0)
^~~~~~~~~~~~~~~~~~~~~~~~~~~
../binding.cc:58:1:
note: in expansion of macro 'NODE_MODULE_CONTEXT_AWARE'
NODE_MODULE_CONTEXT_AWARE(NODE_GYP_MODULE_NAME, Initialize)
^~~~~~~~~~~~~~~~~~~~~~~~~
This commit adds the flag -Wno-cast-function-type to suppress these
warnings. With this change the warnings are not displayed anymore and
the output matches that of osx when running
'make -j8 test/addons/.buildstamp'.
PR-URL: https://github.com/nodejs/node/pull/25663
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
PKCS#8 describes an encoding for private keys, not public keys. Using
the name as a format usually refers to the PrivateKeyInfo structure it
defines. (Or sometimes the EncryptedPrivateKeyInfo variant or the
little-used OneAsymmetricKey extension defined in RFC 5958.) None of
these structures can hold public keys to begin with.
d2i_PUBKEY actually parses the public key analogous format defined in
X.509, SubjectPublicKeyInfo. This function already uses "X.509" to refer
to an entire certificate, so say SubjectPublicKeyInfo in the comment.
(The abbreviation SPKI sometimes refers to SubjectPublicKeyInfo, e.g. in
RFC 7469, but it can also an alternate S-expression-based PKI to X.509,
Simple Public Key Infrastructure. To avoid confusion, stick with the
expanded name.)
PR-URL: https://github.com/nodejs/node/pull/25736
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Refactor benchmark/common.js to consistently use arrow functions for
anonymous callbacks and trailing commas.
PR-URL: https://github.com/nodejs/node/pull/25805
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Beth Griggs <Bethany.Griggs@uk.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
