When the last signal listener is removed, the signal wrap should be
closed, restoring the default signal handling behaviour. This is done in
a (patched) process.removeListener(). However, events.removeAllListeners
has an optimization to avoid calling removeListener() if there are no
listeners for the 'removeListener' event, introduced in 56668f54d1. That
caused the following code to fail to terminate:
process.stdin.resume();
function listener() {};
process.on('SIGINT', listener);
process.removeAllListeners('SIGINT');
process.kill(process.pid, 'SIGINT')
while the following will terminate:
process.stdin.resume();
function listener() {};
process.on('SIGINT', listener);
process.removeListener('SIGINT', listener);
process.kill(process.pid, 'SIGINT')
Replace the method patching with use of the 'newListener' and
'removeListener' events, which will fire no matter which methods are
used to add or remove listeners.
PR-URL: https://github.com/iojs/io.js/pull/687
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Removed DTRACE_NET_SOCKET_READ and DTRACE_NET_SOCKET_WRITE
as they were never called from within the source code.
PR-URL: https://github.com/iojs/io.js/pull/694
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
* Include a description for the error message
* For rename, link, and symlink, include both the source and destination
path in the error message.
* Expose the destination path as the `dest` property on the error object.
* Fix a bug where `ThrowUVException()` would incorrectly delegate to
`Environment::TrowErrnoException()`.
API impact:
* Adds an extra overload for node::UVException() which takes 6
arguments.
PR: https://github.com/iojs/io.js/pull/675
Fixes: https://github.com/iojs/io.js/issues/207
Closes: https://github.com/iojs/io.js/pull/293
Reviewed-by: Ben Noordhuis <info@bnoordhuis.nl>
Boosts speed up to 10% on primitive `createHash().update().digest()`
benchmark.
PR-URL: https://github.com/iojs/io.js/pull/664
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Handle most popular cases in a trie-style, branching on a first
character.
Remove useless HandleScope which was only eating time without producing
any value.
PR-URL: https://github.com/iojs/io.js/pull/664
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Signal dispositions are inherited by child processes. Restore ours to
sane defaults in case our parent process changed it, to prevent quirky
behavior when the parent does something silly like ignoring SIGSEGV.
PR-URL: https://github.com/iojs/io.js/pull/615
Reviewed-By: Sam Roberts <sam@strongloop.com>
Instead of installing an early debug signal handler, simply block the
SIGUSR1 signal at start-up and unblock it when the debugger is ready.
Both approaches are functionally equivalent but blocking the signal
accomplishes it in fewer lines of code.
PR-URL: https://github.com/iojs/io.js/pull/615
Reviewed-By: Sam Roberts <sam@strongloop.com>
Execute the per-platform initialization logic as early as possible,
for two reasons:
1. It opens the way for an upcoming commit to simplify early SIGUSR1
handling.
2. It should make life easier for embedders because io.js no longer
mucks around with the file descriptor limit or signal disposition
of the process.
PR-URL: https://github.com/iojs/io.js/pull/615
Reviewed-By: Sam Roberts <sam@strongloop.com>
Switch from running the loop with UV_RUN_ONCE to UV_RUN_DEFAULT, because
it's possible that the poll returns earlier than expected and thus the
timer is not run on a single interation.
The loop is not stopped either from the timer callback or from the async
handle's.
PR-URL: https://github.com/iojs/io.js/pull/622
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Rod Vagg <rod@vagg.org>
Notable changes:
* npm upgrade to 2.3.0 fixes Windows "uid is undefined" errors
* crypto.pseudoRandomBytes() is now an alias for crypto.randomBytes()
and will block if there is insufficient entropy to produce secure
values. See https://github.com/iojs/io.js/commit/e5e5980 for details.
* Patch for V8 to properly detect ARMv6; binaries now work again on
ARMv6 (Raspberry Pi etc.)
* Minor V8 upgrade from 4.1.0.7 to 4.1.0.12
* 'punycode' core module bumped from stability level 2-Unstable,
to 3-Stable
* Added new collaborators:
- Thorsten Lorenz (@thlorenz)
- Stephen Belanger (@qard)
- Jeremiah Senkpiel (@fishrock123)
- Evan Lucas (@evanlucas)
- Brendan Ashworth (@brendanashworth)
Previously pseudoRandomBytes worked similarly to randomBytes but in the
event of insufficient entropy would silently return non-secure values.
As of f68a116, the entropy pool blocks if there is insufficient entropy
instead of giving an error so there is now no longer a case where
pseudoRandomBytes would act differently than randomBytes.
Docs are updated to remove pseudoRandomBytes and to clarify that
randomBytes now does block instead of erring when entropy is low.
PR-URL: https://github.com/iojs/io.js/pull/557
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Since setting object properties in C++ can be slow, pass
data to JS using preallocated smalloc buffer and create
object in JS instead.
PR-URL: https://github.com/iojs/io.js/pull/469
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Mark several methods "override" in order to remove build warnings.
PR-URL: https://github.com/iojs/io.js/pull/531
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This can make node_is_initialized correctly set to true for applications
that use node::Init to embed iojs.
PR-URL: https://github.com/iojs/io.js/pull/225/
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Remove a few unused or barely used macros from src/node_file.cc.
PR-URL: https://github.com/iojs/io.js/pull/529
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Rename the misnomers ASSERT_IS_STRING_OR_BUFFER and ASSERT_IS_BUFFER.
Said macros don't assert, they throw a TypeError and return.
PR-URL: https://github.com/iojs/io.js/pull/529
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Add ASSERT counterparts to the CHECK_EQ/CHECK_NE/etc. family of macros.
PR-URL: https://github.com/iojs/io.js/pull/529
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Notable changes
* V8 upgrade from 3.31 to 4.1, this is not a major upgrade, the version number "4.1" signifies tracking towards Chrome 41. The 3.31 branch is now not tracking towards a stable release.
* Re-enable Windows XP / 2003 support
* npm upgrade to 2.2.0
* Improved FreeBSD support
Now that V8 has been upgraded, remove the --noharmony_classes and
--noharmony_object_literals workarounds from commits a2751e3e and
4e58211b.
PR-URL: https://github.com/iojs/io.js/pull/490
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Kenan Sulayman <kenan@sly.mn>
Reviewed-By: Rod Vagg <rod@vagg.org>
OpenBSD doesn't define AI_V4MAPPED by default. Defining it to 0
as done with other ports.
PR-URL: https://github.com/iojs/io.js/pull/427
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Use an array instead of an object to pass a parsed header chunk from c++
to javascript. This offers a 5-10% speedup on the http_simple benchmark,
as evidenced by running:
ab -k -t 100 -c 100 http://127.0.0.1:8000/bytes/100
PR: https://github.com/iojs/io.js/pull/292
Reviewed-by: Ben Noordhuis <info@bnoordhuis.nl>
The copyright and license notice is already in the LICENSE file. There
is no justifiable reason to also require that it be included in every
file, since the individual files are not individually distributed except
as part of the entire package.
* quote "NODE_VERSION_STRING" in node.rc to allow for complex version
strings
* change user-facing strings
* make sure .bat files are crlf
PR-URL: https://github.com/iojs/io.js/pull/291
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Reviewed-By: Nikolai Vavilov <vvnicholas@gmail.com>
io.js v1.0.0 is not ABI-compatible with joyent/node@v0.12, update
the NODE_MODULE_VERSION macro to reflect that. Update to a much
larger value to avoid future clashes with joyent/node.
PR-URL: https://github.com/iojs/io.js/pull/312
Reviewed-By: Bert Belder <bertbelder@gmail.com>
SSLv2 has been deprecated and known broken for nearly twenty years now.
I made SSLv2 support opt-in well over a year ago in commit 39aa894 and
now this commit removes it entirely.
PR-URL: https://github.com/iojs/io.js/pull/290
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Rod Vagg <rod@vagg.org>
removed because nightly builds use the TAG make variable
to create a prerelease tag after the version yet `-pre` is
also appended beyond that. `-pre` is unnecessary if another
prerelease tag is being used.
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
In very unlikely case, where `deflateInit2()` may return error (right
now happening only on exhausting all memory), the `ZCtx::Error()` will
be called and will try to `Unref()` the handle. But the problem is that
this handle was never `Ref()`ed, so it will trigger an assertion error
and crash the program.
Reviewed-by: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: https://github.com/joyent/node/pull/8687
Cherry-picked-from: 8c868989be
Sam Roberts
Glen Keane
Bert Belder
Fedor Indutny
Ben Noordhuis
Saúl Ibarra Corretgé
Johan Bergström
Rod Vagg
Calvin Metcalf
Vladimir Kurchatkin
Trevor Norris
Cheng Zhao
toastynerd
Jongyeol Choi
Yosuke Furukawa
Aaron Bieber
Shigeki Ohtsu
isaacs