Ben Noordhuis [Mon, 12 May 2014 03:38:47 +0000 (05:38 +0200)]
deps: fix up v8 postmortem codegen
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Trevor Norris [Tue, 9 Sep 2014 21:03:08 +0000 (14:03 -0700)]
v8: Upgrade 3.26.33 with 14 patches
V8 3.26.31 has received 14 patches since the upgrade to 3.26.33. Since
3.26.33 is technically a tag on the 3.27 branch, reverting back to
3.26.31 would remove now default functionality like WeakMaps. Because of
that the patches have simply been cherry-picked and squashed.
Here is a summary of all patches:
* Fix index register assignment in LoadFieldByIndex for arm, arm64, and
mips.
* Fix invalid attributes when generalizing because of incompatible map
change.
* Skip write barriers when updating the weak hash table.
* MIPS: Avoid HeapObject check in HStoreNamedField.
* Do GC if CodeRange fails to allocate a block.
* Array.concat: properly go to dictionary mode when required.
* Keep CodeRange::current_allocation_block_index_ in range.
* Grow heap slower if GC freed many global handles.
* Do not eliminate bounds checks for "<const> - x".
* Add missing map check to optimized f.apply(...).
* In GrowMode, force the value to the right representation to avoid
deopts between storing the length and storing the value.
* Reduce max executable size limit.
* Fix invalid condition in check elimination effects.
* Fix off-by-one error in Array.concat slow mode check.
For more information see: https://github.com/v8/v8/commits/3.26
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Trevor Livingston [Fri, 5 Sep 2014 14:56:55 +0000 (09:56 -0500)]
tls: `checkServerIdentity` option
Allow overriding `checkServerIdentity` function, when connecting to a
TLS server.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Ben Noordhuis [Fri, 5 Sep 2014 15:33:28 +0000 (17:33 +0200)]
src: remove Environment::GetCurrentChecked()
There is only one call site that uses it and that can do the checks
itself. Removes ~15 lines of code.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Fri, 22 Aug 2014 23:25:40 +0000 (01:25 +0200)]
lib: unexport http.parsers
Unexport the http.parsers freelist. It was originally exported by Ryan
in commit
0003c701 but the commit log doesn't mention why and it's never
been documented. It's unclear if there are any users.
The lifecycle of parser objects changed recently and it seems better to
not let people shoot themselves in the foot so easily.
If it turns out there are actually users, we can always re-export it
again - probably under a slightly different name, to force people to
update their code to the new way of things.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Fri, 22 Aug 2014 22:57:45 +0000 (00:57 +0200)]
lib: http: poison parser references after freeing
Make it a little harder to slip in use-after-free bugs by nulling out
references to the parser object after handing it off to freeParser().
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Fri, 22 Aug 2014 22:33:04 +0000 (00:33 +0200)]
src: update and expand comments in base-object.h
It's safe to call BaseObject::object() from your destructor _unless_
the handle is weak; then it's the weak callback that is calling your
destructor and the object will have been released by the time the
destructor runs.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Fri, 22 Aug 2014 22:31:44 +0000 (00:31 +0200)]
lib, src: don't make http parser handles weak
Weak handles put strain on the garbage collector and the parser handle
doesn't need to be weak in the first place. This change should improve
GC times on busy servers a little.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Thu, 4 Sep 2014 02:23:03 +0000 (04:23 +0200)]
src: add ClearWrap() to util.h
Counterpart to Wrap(), clears the previously assigned internal field.
Will be used in an upcoming commit.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Fri, 22 Aug 2014 21:57:46 +0000 (23:57 +0200)]
src: fix handle leak in Parser::Execute()
Fix a resource leak where an intermediate Local<Context> handle in
Environment::GetCurrent() got leaked into whatever HandleScope was
further up the stack.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Majid Arif Siddiqui [Thu, 14 Aug 2014 03:15:24 +0000 (11:15 +0800)]
lib: improved forEach object performance
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
cjihrig [Thu, 4 Sep 2014 15:10:44 +0000 (11:10 -0400)]
test: listen on exclusive port in cluster workers
Test that listening on exclusive ports with the cluster module works
correctly.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
cjihrig [Fri, 22 Aug 2014 20:51:53 +0000 (16:51 -0400)]
net,dgram: workers can listen on exclusive ports
Allow cluster workers to listen on exclusive ports for TCP and UDP,
instead of forcing all calls to go through the cluster master.
Fixes: #3856
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Trevor Norris [Wed, 3 Sep 2014 21:24:50 +0000 (14:24 -0700)]
src: be more intelligent about use of "arguments"
Use 'use strict' when there are named arguments and the arguments object
is passed to apply(). Also pass named arguments to call() when the named
argument is modified by the function.
Suggested in
https://github.com/joyent/node/pull/8302#issuecomment-
54331801
Confirmed in
https://github.com/joyent/node/pull/8302#issuecomment-
54364818
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Fedor Indutny [Mon, 1 Sep 2014 14:44:57 +0000 (18:44 +0400)]
tls: support multiple keys/certs
Required to serve website with both ECDSA/RSA certificates.
Fedor Indutny [Mon, 1 Sep 2014 14:44:07 +0000 (18:44 +0400)]
tls_wrap: fix use after free
Do not free TLSCallbacks from StreamWrap. TLSCallbacks is bound to a V8
object and should be collected by V8's GC.
Fedor Indutny [Mon, 1 Sep 2014 12:07:45 +0000 (16:07 +0400)]
crypto: use less memory for storing keys
Use `BIO_new_mem_buf` where possible to reduce memory usage and
initialization costs.
Jackson Tian [Mon, 1 Sep 2014 14:44:19 +0000 (22:44 +0800)]
net: Improve Socket.prototype.write()
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Trevor Norris [Wed, 3 Sep 2014 10:36:17 +0000 (03:36 -0700)]
stream_wrap: Add support to write binary strings
node::StringBytes::Write() has appropriate support to write strings with
'binary' encoding. So expose that API through StreamWrap and allow
inheriting classes to use it.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Trevor Norris [Tue, 2 Sep 2014 18:30:45 +0000 (11:30 -0700)]
node,async-wrap: verify domain enter/exit are set
The REPL global object lazy loads modules by placing getters for each.
This causes MakeDomainCallback() to be run if a native module is loaded
from the REPL, but if the domain module hasn't been loaded then there
are no enter/exit callbacks to be called. Causing an assert() to fail.
Fix the issue by conditionally running the callback instead of asserting
it is available. Also add "addon" test to verify the fix.
Fixes: #8231
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Jesús Leganés Combarro "piranna [Sat, 23 Aug 2014 21:01:06 +0000 (23:01 +0200)]
configure: generate a fully statically linked exec
Allow to create an executable with no external dynamic libraries, also the
ones from the system. This is somewhat dependent of the used C lib, for
example glibc has some internal dynamic libraries loaded by itself, but for
other ones like eglibc or dietlib, this would produce a true static linked
executable. This can be of interest for embebers or resource constraints
platforms, but the main reason for this is to allow to use a Javascript
file as Linux kernel 'init' on NodeOS.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Julien Gilli [Sat, 16 Aug 2014 00:20:32 +0000 (17:20 -0700)]
cluster: centralize removal from workers list.
Currently, cluster workers can be removed from the workers list in three
different places:
- In the exit event handler for the worker process.
- In the disconnect event handler of the worker process.
- In the disconnect event handler of the cluster master.
However, handles for a given worker are cleaned up only in one of these
places: in the cluster master's disconnect event handler.
Because these events happen asynchronously, it is possible that the
workers list is empty before we even clean up one handle. This makes
the assert that makes sure that no handle is left when the workers
list is empty fail.
This commit removes the worker from the cluster.workers list only when
the worker is dead _and_ disconnected, at which point we're sure that
its associated handles are cleaned up.
Fixes #8191 and #8192.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Brian White [Tue, 2 Sep 2014 13:27:33 +0000 (09:27 -0400)]
crypto: unsigned value can't be negative
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Brian White [Tue, 2 Sep 2014 13:26:14 +0000 (09:26 -0400)]
crypto: avoid memory leak
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Jackson Tian [Sat, 23 Aug 2014 14:48:13 +0000 (22:48 +0800)]
src: reading/owner/onread/onconnection for tcp
Initialize fields to avoid Hidden Class creation in runtime.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Isaac Burns [Fri, 29 Aug 2014 19:47:42 +0000 (14:47 -0500)]
crypto: wrap ECDH constants in HAVE_OPENSSL
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Shigeki Ohtsu [Thu, 28 Aug 2014 07:51:42 +0000 (16:51 +0900)]
tls: add DHE-RSA-AES128-SHA256 to the def ciphers
`!EDH` is also removed from the list in the discussion of #8272
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Shigeki Ohtsu [Wed, 27 Aug 2014 09:00:13 +0000 (18:00 +0900)]
tls, crypto: add DHE support
In case of an invalid DH parameter file, it is sliently discarded. To
use auto DH parameter in a server and DHE key length check in a
client, we need to wait for the next release of OpenSSL-1.0.2.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Wed, 27 Aug 2014 14:01:01 +0000 (18:01 +0400)]
crypto: introduce ECDH
Jackson Tian [Mon, 25 Aug 2014 12:55:11 +0000 (20:55 +0800)]
src: Add function name for .byteLength/.compare
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Sat, 23 Aug 2014 13:38:32 +0000 (17:38 +0400)]
crypto: allow padding in RSA methods
Reviewed-By: Trevor Norris <trevnorris@gmail.com>
Fedor Indutny [Sat, 23 Aug 2014 19:22:23 +0000 (23:22 +0400)]
crypto: fix memory leak in Connection::New
Do not create `SSL` instance twice, `SSL_new` is called from `SSLBase`
constructor anyway.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Jackson Tian [Thu, 21 Aug 2014 16:05:42 +0000 (00:05 +0800)]
http: avoid create difference hidden class
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Chris Dickinson [Tue, 19 Aug 2014 21:38:55 +0000 (14:38 -0700)]
net,stream: add isPaused, don't read() when paused
net Sockets were calling read(0) to start reading, without
checking to see if they were paused first. This would result
in paused Socket objects keeping the event loop alive.
Fixes #8200
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
cjihrig [Fri, 15 Aug 2014 14:40:09 +0000 (10:40 -0400)]
dns: throw if hostname is not string or falsey
Fix assertion failure from poor argument parsing logic introduced in
6ea5d16. Add tests to make sure arguments are properly parsed.
Fixes: 6ea5d16 "dns: always set variable family in lookup()"
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ben Noordhuis [Tue, 19 Aug 2014 17:28:06 +0000 (10:28 -0700)]
node: add missing Isolate::Scope at startup
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Yazhong Liu [Wed, 13 Aug 2014 18:02:37 +0000 (02:02 +0800)]
src: add missing Isolate arguments
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Alex Kocharin [Thu, 14 Aug 2014 14:54:44 +0000 (18:54 +0400)]
http: fix bailout for writeHead
Reported-by: Jackson Tian <shyvo1987@gmail.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fedor Indutny [Sun, 17 Aug 2014 10:17:30 +0000 (14:17 +0400)]
gyp: preserve v8dbg syms on freebsd too
Alexis Campailla [Tue, 12 Aug 2014 14:48:02 +0000 (16:48 +0200)]
test: fix dns test
Fix a few issues in test/internet/test-dns.js:
- 'hint' should be 'hints'
- reverse name lookup is not guaranteed to return 'localhost'
- V4MAPPED hint requires IPV6 address family
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Julien Gilli [Mon, 11 Aug 2014 20:29:00 +0000 (13:29 -0700)]
tests: don't assume IPv4 only in remote addr tests
Tests in test-net-remote-address-port.js assume that client and server
sockets always use IPv4. However, depending on the OS and the network
interfaces setup, this is not true. This change makes the test consider
that both IPv4 or IPv6 sockets are valid
Fixes #8096.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Fedor Indutny [Sat, 2 Aug 2014 08:42:42 +0000 (12:42 +0400)]
test: check ipv6 support before testing it
fix #7983
fix #8049
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Julien Gilli [Wed, 6 Aug 2014 21:32:46 +0000 (14:32 -0700)]
tests: do not hardcode service name in test-dns.
Instead of hard-coding http service name in test-dns, retrieve it from
/etc/services. This is not ideal, but it's still better than hard-coding
it.
Fixes #8047.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Julien Gilli [Mon, 11 Aug 2014 22:15:21 +0000 (15:15 -0700)]
tests: fix invalid hints flags dns test.
1 is actually a valid flag on SmartOS. More generally, hints flags'
values are defined by the underlying native flags, and these can have
different values on different systems.
Using (ADDRCONFIG | V4MAPPED) + 1 ensure that the flag will be invalid,
since it will always be different from ADDRCONFIG, V4MAPPED, ADDRCONFIG
| V4MAPPED, 0 and any other combination of even flags.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Cheng Zhao [Sun, 10 Aug 2014 02:46:11 +0000 (10:46 +0800)]
src: change kIsolateSlot to 3
The slot 0 and 1 had already been taken by "gin" and "blink" in Chrome,
and the size of isolate's slots is 4 by default, so using 3 should hopefully
make node work independently when embedded into other application.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Trevor Norris [Wed, 13 Aug 2014 04:50:21 +0000 (21:50 -0700)]
src: require EventEmitter via NativeModule
Fixes a recent change causing test-process-kill-pid.js to fail.
Fixes: 931cbc1 "lib: don't use emitter.listeners(type).length"
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Jackson Tian [Wed, 13 Aug 2014 04:03:25 +0000 (12:03 +0800)]
dgram: remove new keyword from errnoException
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Kyle Robinson Young [Fri, 8 Aug 2014 17:49:24 +0000 (10:49 -0700)]
doc: typo fixes on stream, tls and http
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Ezequiel Rabinovich [Wed, 13 Aug 2014 02:32:22 +0000 (23:32 -0300)]
querystring: remove prepended ? from query field
Fixes an issue that caused the first querystring to be parsed prepending
a "?" in the first variable name on relative urls with no #fragment
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Jackson Tian [Mon, 11 Aug 2014 08:41:18 +0000 (16:41 +0800)]
lib: don't use emitter.listeners(type).length
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Trevor Norris [Tue, 12 Aug 2014 00:10:43 +0000 (17:10 -0700)]
node: set names for prototype methods
Fix issue where output of a native prototype method would simply print
[Function]. It will now print [Function: name].
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
seishun [Mon, 5 May 2014 14:35:28 +0000 (17:35 +0300)]
crypto: add RSA encryption
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Alexis Campailla [Tue, 5 Aug 2014 13:33:16 +0000 (15:33 +0200)]
windows: fix memory leak in WinapiErrnoException
Fix https://github.com/joyent/node/issues/2341
Reviewed-By: Fedor Indutny <fedor@indutny.com>
cjihrig [Fri, 8 Aug 2014 02:36:56 +0000 (22:36 -0400)]
dns: always set variable family in lookup()
Regression occurred that prevented the variable "family" from being set
properly when the lookup() function's "options" parameter was passed a
number instead of an object.
Also included a sanity check by setting the default value of "family" to
a value that will not pass verification.
Fixes: e643fe4 "dns: fix GetAddrInfo assert"
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fedor Indutny [Wed, 11 Dec 2013 17:19:04 +0000 (21:19 +0400)]
openssl: fix keypress requirement in apps on win32
Original source:
http://openssl.6102.n7.nabble.com/PATCH-s-client-Fix-keypress-requirement-with-redirected-input-on-Windows-td46787.html
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Thorsten Lorenz [Wed, 6 Aug 2014 18:00:31 +0000 (14:00 -0400)]
build: linking CoreFoundation framework for OSX
Linking CoreFoundation for OSX is needed for OSX debugging features to
function properly.
For instance Instruments cannot record Heap Allocations if the
CoreFoundation is not linked.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Alexis Campailla [Thu, 7 Aug 2014 14:45:53 +0000 (16:45 +0200)]
test: fix test-process-kill-pid on Windows
Disabling the part of the test that relies on dispatching SIGHUP,
because sending SIGHUP is not supported on Windows.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Jackson Tian [Wed, 6 Aug 2014 06:35:35 +0000 (14:35 +0800)]
doc: document max `new Buffer(size)`
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Timothy J Fontaine [Thu, 7 Aug 2014 23:33:35 +0000 (16:33 -0700)]
Merge remote-tracking branch 'upstream/v0.10' into v0.12
Conflicts:
ChangeLog
Makefile
deps/uv/ChangeLog
deps/uv/build.mk
deps/uv/src/unix/darwin.c
deps/uv/src/unix/getaddrinfo.c
deps/uv/src/version.c
deps/v8/src/checks.h
deps/v8/src/isolate.h
lib/cluster.js
lib/module.js
lib/timers.js
lib/tls.js
src/node_version.h
Julien Gilli [Thu, 31 Jul 2014 23:16:42 +0000 (16:16 -0700)]
tests: fix child-process-fork-dgram on SmartOS.
Send messages until both the parent and the child process have received
at least one message. If at least one of them doesn't receive any
message, the test runner will make the test timeout.
Fixes #8046.
Maciej Małecki [Thu, 7 Aug 2014 18:56:34 +0000 (20:56 +0200)]
doc: document arguments for 'error' event on a stream
Fixes #6361.
Jackson Tian [Wed, 6 Aug 2014 03:21:59 +0000 (11:21 +0800)]
fs: fix fs.readFileSync fd leak when get RangeError
Saúl Ibarra Corretgé [Sat, 5 Jul 2014 09:02:33 +0000 (11:02 +0200)]
src: handle UV_EAGAIN in TryWrite
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Alexis Campailla [Thu, 24 Jul 2014 09:59:00 +0000 (11:59 +0200)]
Cluster: fix shared handles on Windows
This is the Node side of the fix for Node's cluster module on Windows.
https://github.com/joyent/node/issues/7691
The other required part is
https://github.com/joyent/libuv/pull/1384
Windows and Unix return certain socket errors (i.e. EADDRINUSE) at
different times: bind on Windows, and listen on Unix.
In an effort to hide this difference, libuv on Windows stores such
errors in the bind_error field of uv_tcp_t, to defer raising it at
listen time.
This worked fine except for the case in which a socket is shared in
a Node cluster and a bind error occurs.
A previous attempt to fix this (
https://github.com/joyent/libuv/commit/
d1e6be1460f555a1f8a4063d7642696aa7238769
https://github.com/joyent/node/commit/
3da36fe00e5d85414031ae812e473f16629d8645
) was flawed becaused in an attempt to relay the error at the JS level
it caused the master to start accepting connections.
With this new approach, libuv itself is relaying the bind errors,
providing for a uniform behavior of uv_tcp_listen.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Thu, 7 Aug 2014 11:03:17 +0000 (15:03 +0400)]
deps: update libuv to v0.11.28
Fedor Indutny [Thu, 7 Aug 2014 09:36:56 +0000 (13:36 +0400)]
deps: update openssl to v1.0.1i
Kevin Simper [Sun, 3 Aug 2014 00:27:02 +0000 (02:27 +0200)]
doc: clarify factory methods for net.Socket
Jackson Tian [Tue, 5 Aug 2014 09:34:56 +0000 (17:34 +0800)]
fs: fs.readFile should not throw uncaughtException
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Trevor Norris [Tue, 5 Aug 2014 01:12:54 +0000 (18:12 -0700)]
dns: fix GetAddrInfo assert
The method GetAddrInfo() is used by more than just dns.lookup(), and in
those cases a third argument isn't passed. This caused the following
check to abort:
assert(args[3]->IsInt32());
Fixes: 4306786 "net: don't prefer IPv4 addresses during resolution"
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
cjihrig [Thu, 22 May 2014 02:13:09 +0000 (22:13 -0400)]
net: don't prefer IPv4 addresses during resolution
Currently the address resolution family defaults to IPv4. Instead remove
the preference and instead resolve to a family suitable for the host.
Expose the getaddrinfo flags and allow them to be passed.
Add documentation about new flags.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Timothy J Fontaine [Mon, 4 Aug 2014 20:43:50 +0000 (13:43 -0700)]
v8: Fix compliation with GCC 4.8
Supresses a very loud warning from GCC 4.8 about unused typedefs
Original url https://codereview.chromium.org/
69413002
Julien Gilli [Fri, 1 Aug 2014 18:06:54 +0000 (11:06 -0700)]
test: fix test-net-remote-address-port
Do not use first socket in second socket's connect handler. Probably a
copy/paste mistake.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Herman Lee [Mon, 4 Aug 2014 14:18:54 +0000 (22:18 +0800)]
path: isAbsolute() should always return boolean
On Windows, path.isAbsolute() returns an empty string on failed cases.
This forces the return value to always be boolean.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Julien Gilli [Mon, 4 Aug 2014 16:52:35 +0000 (09:52 -0700)]
timers: fix timers with non-integer delay hanging.
When backporting f8193ab into v0.10, a regression was introduced. Timers
with non-integer timeout could trigger a infinite recursion with 100%
cpu usage. This commit backports 93b0624 which fixes the regression.
After backporting f8193ab, instead of using Date.now(), timers would use
Timer.now() to determine if they had expired. However, Timer.now() is
based on loop->time, which is not updated when a timer's remaining time
is > 0 and < 1. Timers would thus never timeout if their remaining time
was at some point > 0 and < 1.
With this commit, Timer.now() updates loop->time itself, and timers
always timeout eventually.
Fixes #8065 and #8068.
Jackson Tian [Fri, 1 Aug 2014 11:26:09 +0000 (19:26 +0800)]
process: improve process binding
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Sat, 2 Aug 2014 11:06:25 +0000 (15:06 +0400)]
test: disable 0-dgram tests on darwin
fix #8023
Sam Roberts [Thu, 31 Jul 2014 23:44:15 +0000 (16:44 -0700)]
cluster: disconnect should not be synchronous
Callbacks in node are usually asynchronous, and should never be
sometimes synchronous, and sometimes asynchronous.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Lucio M. Tato [Sat, 2 Aug 2014 05:33:35 +0000 (02:33 -0300)]
path: fix slice OOB in trim
Internal function trim(arr). 2nd parameter of slice() should be slice's
end index (not included). Because of function normalize() (called before
trim()), "start" is always zero so the bug -for now- has no effect, but
its a bug waiting to happen.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Sam Roberts [Thu, 31 Jul 2014 23:49:28 +0000 (16:49 -0700)]
test: assert cluster.disconnect is async
See joyent/node#8043, test passed on v0.11 already, but this makes the
test stronger.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
James Halliday [Mon, 9 Jun 2014 01:58:53 +0000 (18:58 -0700)]
stream: fix Readable.wrap objectMode falsy values
A streams1 stream will have its falsy values such as 0, false, or ""
eaten by the upgrade to streams2, even when objectMode is enabled.
Include test for said cases.
Reviewed-by: isaacs <i@izs.me>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Jackson Tian [Fri, 1 Aug 2014 06:08:52 +0000 (14:08 +0800)]
tls: throw an error, not string
Reviewed-By: Fedor Indutny <fedor@indutny.com>
cjihrig [Thu, 31 Jul 2014 02:20:52 +0000 (22:20 -0400)]
cluster: support options in Worker constructor
This commit moves some common Worker code into the constructor
via support for an options argument.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Jakob Gillich [Wed, 25 Jun 2014 17:08:46 +0000 (19:08 +0200)]
docs: fix non-string ignore note in path.resolve
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Gabriel Wicke [Tue, 1 Jul 2014 19:28:49 +0000 (12:28 -0700)]
Add fast path for simple URL parsing
This patch adds a fast path for parsing of simple path-only URLs, as commonly
found in HTTP requests received by a server.
Benchmark results [ms], before / after patch:
/foo/bar 0.008956 0.000418 (fast path used)
http://example.com/ 0.011426 0.011437 (normal slow path, no change)
In a simple 'ab' benchmark of a single-threaded web server, this patch
increases the request rate from around 6400 to 7400 req/s.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Timothy J Fontaine [Thu, 31 Jul 2014 18:12:08 +0000 (11:12 -0700)]
Now working on 0.10.31
Timothy J Fontaine [Thu, 31 Jul 2014 18:11:53 +0000 (11:11 -0700)]
Merge branch 'v0.10.30-release' into v0.10
Timothy J Fontaine [Thu, 31 Jul 2014 17:24:06 +0000 (10:24 -0700)]
2014.07.31, Version 0.10.30 (Stable)
* uv: Upgrade to v0.10.28
* npm: Upgrade to v1.4.21
* v8: Interrupts must not mask stack overflow.
* Revert "stream: start old-mode read in a next tick" (Fedor Indutny)
* buffer: fix sign overflow in `readUIn32BE` (Fedor Indutny)
* buffer: improve {read,write}{U}Int* methods (Nick Apperson)
* child_process: handle writeUtf8String error (Fedor Indutny)
* deps: backport
4ed5fde4f from v8 upstream (Fedor Indutny)
* deps: cherry-pick
eca441b2 from OpenSSL (Fedor Indutny)
* lib: remove and restructure calls to isNaN() (cjihrig)
* module: eliminate double `getenv()` (Maciej Małecki)
* stream2: flush extant data on read of ended stream (Chris Dickinson)
* streams: remove unused require('assert') (Rod Vagg)
* timers: backport f8193ab (Julien Gilli)
* util.h: interface compatibility (Oguz Bastemur)
* zlib: do not crash on write after close (Fedor Indutny)
Timothy J Fontaine [Thu, 31 Jul 2014 16:31:01 +0000 (09:31 -0700)]
uv: Upgrade to v0.10.28
Fedor Indutny [Wed, 30 Jul 2014 22:33:52 +0000 (15:33 -0700)]
v8: Interrupts must not mask stack overflow.
Backport of https://codereview.chromium.org/
339883002
Timothy J Fontaine [Thu, 31 Jul 2014 16:05:30 +0000 (09:05 -0700)]
npm: Upgrade to v1.4.21
Maciej Małecki [Tue, 25 Mar 2014 21:47:05 +0000 (04:47 +0700)]
module: eliminate double `getenv()`
`process.env` access results in a synchronous `getenv` call. Cache the
first result instead and save one syscall.
Rod Vagg [Sun, 30 Mar 2014 08:51:01 +0000 (19:51 +1100)]
streams: remove unused require('assert')
Julien Gilli [Wed, 23 Jul 2014 01:03:10 +0000 (18:03 -0700)]
timers: backport f8193ab
Original commit message:
timers: use uv_now instead of Date.now
This saves a few calls to gettimeofday which can be expensive, and
potentially subject to clock drift. Instead use the loop time which
uses hrtime internally.
In addition to the backport, this commit:
- keeps _idleStart timers' property which is still set to
Date.now() to avoid breaking existing code that uses it, even if
its use is discouraged.
- adds automated tests. These tests use a specific branch of
libfaketime that hasn't been submitted upstream yet. libfaketime
is git cloned if needed when running automated tests.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Fedor Indutny [Thu, 31 Jul 2014 08:38:46 +0000 (12:38 +0400)]
Revert "Cluster: fix shared handles on Windows"
This reverts commit
4e68a28e20b348f3519b359a17fcb941b235202b.
Alexis Campailla [Thu, 24 Jul 2014 09:59:00 +0000 (11:59 +0200)]
Cluster: fix shared handles on Windows
This is the Node side of the fix for Node's cluster module on Windows.
https://github.com/joyent/node/issues/7691
The other required part is
https://github.com/joyent/libuv/pull/1384
Windows and Unix return certain socket errors (i.e. EADDRINUSE) at
different times: bind on Windows, and listen on Unix.
In an effort to hide this difference, libuv on Windows stores such
errors in the bind_error field of uv_tcp_t, to defer raising it at
listen time.
This worked fine except for the case in which a socket is shared in
a Node cluster and a bind error occurs.
A previous attempt to fix this (
https://github.com/joyent/libuv/commit/
d1e6be1460f555a1f8a4063d7642696aa7238769
https://github.com/joyent/node/commit/
3da36fe00e5d85414031ae812e473f16629d8645
) was flawed becaused in an attempt to relay the error at the JS level
it caused the master to start accepting connections.
With this new approach, libuv itself is relaying the bind errors,
providing for a uniform behavior of uv_tcp_listen.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Thu, 31 Jul 2014 08:30:46 +0000 (12:30 +0400)]
repl: proper `setPrompt()` and `multiline` support
fix #8031
Fedor Indutny [Thu, 31 Jul 2014 08:12:18 +0000 (12:12 +0400)]
Revert "repl: fix overwrite for this._prompt"
This reverts commit
7166b55015261de8ab69758320f3d9159b3eaadd.
Sam Roberts [Mon, 28 Jul 2014 23:53:10 +0000 (16:53 -0700)]
cluster: test events emit on cluster.worker
v0.10 and node docs specific that in a worker, the 'message' and 'error'
event emits on process, and on cluster.worker.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fedor Indutny [Wed, 30 Jul 2014 10:54:36 +0000 (14:54 +0400)]
gyp: do not rm unused openssl syms on osx/linux
fix #8026
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Julien Gilli [Wed, 30 Jul 2014 01:06:47 +0000 (18:06 -0700)]
tests: fix internet/test-dns.js
internet/test-dns.js assumes that ::1 always resolves to "localhost" on
all platforms. This is not what happens in reality. Some platforms
resolve it to "ip6-localhost" too. There doesn't seem to be any consensus
on what's the right thing to do. However, most sane platforms will use
either one of these two values.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>