Removes use of builtins that are unavailable for older clang. Per
benchmarks, only uses builtins on Windows, where speedup is
significant.
Also adds test for unaligned ucs2 buffer write. Between #3410
and #7645, bytes were swapped twice on bigendian platforms if buffer
was not two-byte aligned. See comment in #7645.
PR-URL: https://github.com/nodejs/node/pull/7645
Fixes: https://github.com/nodejs/node/issues/7618
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Refactor various functions that read values from the contextify
options object. Rather than passing args and the index, pass the
value at that index.
We use env->isolate() rather than args.GetIsolate(), but since env
was constructed from args, this is the same isolate.
PR-URL: https://github.com/nodejs/node/pull/8850
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
The list.length argument is missing from the comment and just adding
this for clarity.
PR-URL: https://github.com/nodejs/node/pull/8816
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
When V8 was updated on master to 5.4 there were ABI breaking changes.
In the past we have not landed these types of changes before a release,
and as such have only bumped the NODE_MODULE_VERSION number in the
release commit.
Since we are going to be keeping the V8 5.4 beta on master and in the
v7 betas I think it makes sense for us to bump the module number prior
to a release commit being made. It is possible that this commit should
be reverted prior to v7.0.0 being cut. Alternatively we may want to
modify our release process for V8 to include a NODE_MODULE_VERSION
bump before landing on master when applicable.
NODE_MODULE_VERSION is being bumped to 51 instead of 49 to avoid
conflicts with NODE_MODULE_VERSIONs being used in electron.
Ref: https://github.com/electron/electron/issues/5851#issuecomment-246920775
Ref: https://github.com/nodejs/node/pull/8317
PR-URL: https://github.com/nodejs/node/pull/8808
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Michaël Zasso <mic.besace@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Call `v8::Isolate::GetCurrent()->LowMemoryNotification()` when
an allocation fails to give V8 a chance to clean up and return
memory before retrying (and possibly giving up).
PR-URL: https://github.com/nodejs/node/pull/8482
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Provide shortcut `node::CheckedMalloc()` and friends that
replace `node::Malloc()` + `CHECK_NE(·, nullptr);` combinations
in a few places.
PR-URL: https://github.com/nodejs/node/pull/8482
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Pass the desired return type directly to the allocation functions,
so that the resulting `static_cast` from `void*` becomes unneccessary
and the return type can be use as a reasonable default value for the
`size` parameter.
PR-URL: https://github.com/nodejs/node/pull/8482
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Adds an optional second parameter to `node::Malloc()` and
an optional third parameter to `node::Realloc()` giving the
size/number of items to be allocated, in the style of `calloc(3)`.
Use a proper overflow check using division;
the previous `CHECK_GE(n * size, n);` would not detect all cases
of overflow (e.g. `size == SIZE_MAX / 2 && n == 3`).
PR-URL: https://github.com/nodejs/node/pull/8482
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Rename CHECK_NOT_OOB() to THROW_AND_RETURN_IF_OOB() because the old name
suggests it asserts and aborts when it is really a control flow macro.
PR-URL: https://github.com/nodejs/node/pull/8784
Reviewed-By: Anna Henningsen <anna@addaleax.net>
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: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Correcting a couple of minor spelling typos in comments.
PR-URL: https://github.com/nodejs/node/pull/8736
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Commit 46633934fe (src: pull
OnConnection from pipe_wrap and tcp_wrap) removed the private handle_
member from TCPWrap which should allow us to rename the private
handle__ member in HandleWrap.
PR-URL: https://github.com/nodejs/node/pull/8712
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Throw a JS exception instead of aborting so the user at least has a
fighting chance of understanding what went wrong.
Fixes: https://github.com/nodejs/node/issues/8699
PR-URL: https://github.com/nodejs/node/pull/8710
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Currently, there are a few places where macro functions passed to the
PER_ISOLATE_PRIVATE_SYMBOL_PROPERTIES and
PER_ISOLATE_STRING_PROPERTIES macros, don't use the StringValue parameter.
This commit removes the StringValue parameter where it is not used.
PR-URL: https://github.com/nodejs/node/pull/7905
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Add an error message in watchdog if we abort because uv_loop_init fails.
PR-URL: https://github.com/nodejs/node/pull/8634
Fixes: https://github.com/nodejs/node/issues/8555
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
CHECK includes node::Abort(), so that's unneeded to call ABORT after
CHECK.
PR-URL: https://github.com/nodejs/node/pull/8593
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
The use of libc++ is now forced on OSX.
PR-URL: https://github.com/nodejs/node/pull/8317
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Ali Ijaz Sheikh <ofrobots@google.com>
Change `Malloc()/Calloc()` so that size zero does not return a null
pointer, consistent with prior behavior.
Fixes: https://github.com/nodejs/node/issues/8571
PR-URL: https://github.com/nodejs/node/pull/8572
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@keybase.io>
Reviewed-By: Yorkie Liu <yorkiefixer@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
This commit prevents thrown JavaScript exceptions from crashing
the process in node_contextify's CopyProperties() function.
Fixes: https://github.com/nodejs/node/issues/8537
PR-URL: https://github.com/nodejs/node/pull/8649
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Moves inflateSetDictionary right after inflateInit2 when mode is
INFLATERAW, since without the wrapper in appears zlib won't return
Z_NEED_DICT as it would otherwise, and will thus attempt inflating
without the dictionary, leading to an error.
Fixes: https://github.com/nodejs/node/issues/8507
PR-URL: https://github.com/nodejs/node/pull/8512
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Ctor has to be added as memset to 0 is no longer an option, since
the structure now has std::vector member.
Attempt at fixing nodejs/node#8155 (so far I was not able to repro it)
PR-URL: https://github.com/nodejs/node/pull/8536
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
instead of creating own buffer, use MaybeStackBuffer on DoSend to
take advantage of its destructor to free up memory, so buffer
never leaks memory - even if code in DoSend throws.
Use MaybeStackBuffer in Writev to take advantage of destructor
on MaybeStackBuffer to clear itself up, rather than Writev managing
resources itself.
PR-URL: https://github.com/nodejs/node/pull/8626
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Some tools are now relying on 9229 to be node.js "inspector" port (I
see Chrome extensions, some online blog posts, etc.) Also, having same
default port values for old and new protocols may lead to some
confusion, e.g. when tools are trying to autodiscover debuggable Node
instances.
This is a partial revert of 9f1f7e2. This commit preserves the fix for
issue #8201 bringing back the behavior that the old and new protocols
run on different ports.run on different ports.
PR-URL: https://github.com/nodejs/node/pull/8550
Reviewed-By: ofrobots - Ali Ijaz Sheikh <ofrobots@google.com>
Reviewed-By: cjihrig - Colin Ihrig <cjihrig@gmail.com>
This test executes a simple debug session over the inspector protocol.
PR-URL: https://github.com/nodejs/node/pull/8429
Reviewed-By: ofrobots - Ali Ijaz Sheikh <ofrobots@google.com>
`FChown` and `Chown` test that the `uid` and `gid` parameters
they receive are unsigned integers, but `Stat()` and `FStat()`
would return the corresponding fields of `uv_stat_t` as signed
integers. Applications which pass those these values directly
to `Chown` may fail
(e.g. for `nobody` on OS X, who has an `uid` of `-2`, see e.g.
https://github.com/nodejs/node-v0.x-archive/issues/5890).
This patch changes the `Integer::New()` call for `uid` and `gid`
to `Integer::NewFromUnsigned()`.
All other fields are kept as they are, for performance, but
strictly speaking the respective sizes of those
fields aren’t specified, either.
Ref: https://github.com/npm/npm/issues/13918
PR-URL: https://github.com/nodejs/node/pull/8515
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
undo accidental change to other fields of uv_fs_stat
This commit consolidates the debugging port used by the
inspector and Node's legacy debugger.
Fixes: https://github.com/nodejs/node/issues/8201
PR-URL: https://github.com/nodejs/node/pull/8386
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
This commit attempts to address one of the items in
https://github.com/nodejs/node/issues/4641 which is related to
src/pipe_wrap.cc and src/tcp_wrap.cc.
Currently both pipe_wrap.cc and tcp_wrap.cc contain an AfterConnect
function that are almost identical. This commit extracts this function
into ConnectionWrap so that that both can share it.
PR-URL: https://github.com/nodejs/node/pull/8448
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
They seem to have been introduced as "convenience methods" in commit
75adde0 ("src: remove `node_isolate` from source") for reasons I can
only guess at but they can be removed without much hassle.
PR-URL: https://github.com/nodejs/node/pull/8427
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/8450
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reinstate crypto.timingSafeEqual() which was reverted due to test
issues. The flaky test issues are resolved in this new changeset.
PR-URL: https://github.com/nodejs/node/pull/8304
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Commit a01e8bc (src: moving f function call comment, 2016-09-04) moved
a comment to be closer to the place in the code it was describing.
While working on this there was feedback on the comment
itself:
https://github.com/nodejs/node/commit/
a01e8bcf189bd598d496f347f60007bc2211e528#comments
This PR includes the suggestions in the above comments.
PR-URL: https://github.com/nodejs/node/pull/8416
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Franziska Hinkelmann <franzih@chromium.org>
Reviewed-By: James M Snell <jasnell@gmail.com>
`malloc(0)` may return NULL on some platforms. Do not report
out-of-memory error unless `malloc` was passed a number greater than
`0`.
PR-URL: https://github.com/nodejs/node/pull/8352
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
malloc(0) and realloc(ptr, 0) have implementation-defined behavior in
that the standard allows them to either return a unique pointer or a
nullptr for zero-sized allocation requests. Normalize by always using
a nullptr.
- Introduce node::malloc, node::realloc and node::calloc that should
be used throught our source.
- Update all existing node source files to use the new functions
instead of the native allocation functions.
Fixes: https://github.com/nodejs/node/issues/7549
PR-URL: https://github.com/nodejs/node/pull/7564
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
It seems that it is possible with some toolchains for both `__GLIBC__`
and `__UCLIBC__` to be defined, confusing our "do we have execinfo.h?"
logic.
Assume that when `__UCLIBC__` is defined, we are dealing with a libc
that does not have execinfo.h.
Fixes: https://github.com/nodejs/node/issues/8233
PR-URL: https://github.com/nodejs/node/pull/8308
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
The comment about calling the f function seems to have drifted
a little. Moving it to be closer to the actual call.
PR-URL: https://github.com/nodejs/node/pull/8405
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Changes inspector integration to use Node.js script file name as target
title (reported in JSON and shown in developer tools UIs). It will also
report file:// URL for the script as some tools seem to use that field
to open the script in the editor.
PR-URL: https://github.com/nodejs/node/pull/8243
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
node.cc had two functions with the name AtExit with entirely different
purposes:
* node::AtExit(): file static; used to register the atexit(3) handler
for the Node process.
* node::AtExit(void (*)(void*), void*): publicly exported symbol that
addons can use to request callbacks upon exit.
For code readability it is better to avoid the unintentional overload.
PR-URL: https://github.com/nodejs/node/pull/8273
Reviewed-By: addaleax - Anna Henningsen <anna@addaleax.net>
Reviewed-By: jasnell - James M Snell <jasnell@gmail.com>
Reviewed-By: cjihrig - Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
This fixes a race condition when messages are coming while V8 is still
dispatching the previous batch.
PR-URL: https://github.com/nodejs/node/pull/8264
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
This change simplifies buffer management to address a number of issues
that original implementation had.
Original implementation was trying to reduce the number of allocations
by providing regions of the internal buffer to libuv IO code. This
introduced some potential use after free issues if the buffer grows
(or shrinks) while there's a pending read. It also had some confusing
math that resulted in issues on Windows version of the libuv.
PR-URL: https://github.com/nodejs/node/pull/8257
Fixes: https://github.com/nodejs/node/issues/8155
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
If `IsWindows7OrGreater()` returns `false`, the Node.js program should
exit with a more specific code ERROR_EXE_MACHINE_TYPE_MISMATCH instead
the code 0x1(ERROR_INVALID_FUNCTION)
PR-URL: https://github.com/nodejs/node/pull/8204
Reviewed-By: João Reis <reis@janeasystems.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Zach Bjornson
Brian White
Franziska Hinkelmann
Daniel Bevenius
Myles Borins
Anna Henningsen
Ben Noordhuis
Yosuke Furukawa
Eugene Ostroukhov
yorkie
Michaël Zasso
Rich Trott
cjihrig
Tarjei Husøy
Paul Kiddie
Ojas Shirekar
Santiago Gimeno
not-an-aardvark
Michael Dawson
Ali Ijaz Sheikh