Date.now() indirectly calls gettimeofday() on Linux and that's a system
call that is extremely expensive on virtualized systems when the host
operating system has to emulate access to the hardware clock.
Case in point: output from `perf record -c 10000 -e cycles:u -g -i`
for a benchmark/http_simple bytes/8 benchmark with a light load of
50 concurrent clients:
53.69% node node [.] v8::internal::OS::TimeCurrentMillis()
|
--- v8::internal::OS::TimeCurrentMillis()
|
|--99.77%-- v8::internal::Runtime_DateCurrentTime(v8::internal::Arguments, v8::internal::Isolate*)
| 0x23587880618e
That's right - over half of user time spent inside the V8 function that
calls gettimeofday().
Notably, nearly all system time gets attributed to acpi_pm_read(), the
kernel function that reads the ACPI power management timer:
32.49% node [kernel.kallsyms] [k] acpi_pm_read
|
--- acpi_pm_read
|
|--98.40%-- __getnstimeofday
| getnstimeofday
| |
| |--71.61%-- do_gettimeofday
| | sys_gettimeofday
| | system_call_fastpath
| | 0x7fffbbaf6dbc
| | |
| | |--98.72%-- v8::internal::OS::TimeCurrentMillis()
The cost of the gettimeofday() system call is normally measured in
nanoseconds but we were seeing 100 us averages and spikes >= 1000 us.
The numbers were so bad, my initial hunch was that the node process was
continuously getting rescheduled inside the system call...
v8::internal::OS::TimeCurrentMillis()'s most frequent caller is
v8::internal::Runtime_DateCurrentTime(), the V8 run-time function
that's behind Date.now(). The timeout handling logic in lib/http.js
and lib/net.js calls into lib/timers.js and that module will happily
call Date.now() hundreds or even thousands of times per second.
If you saw exports._unrefActive() show up in --prof output a lot,
now you know why.
That's why this commit makes V8 switch over to clock_gettime() on Linux.
In particular, it checks if CLOCK_REALTIME_COARSE is available and has
a resolution <= 1 ms because in that case the clock_gettime() call can
be fully serviced from the vDSO.
It speeds up the aforementioned benchmark by about 100% on the affected
systems and should go a long way toward addressing the latency issues
that StrongLoop customers have been reporting.
This patch will be upstreamed as a CR against V8 3.26. I'm sending it
as a pull request for v0.10 first because that's what our users are
running and because the delta between 3.26 and 3.14 is too big to
reasonably back-port the patch. I'll open a pull request for the
master branch once the CR lands upstream.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Forcibly flushes the request headers. You need this with long-lived
HTTP connections where the first data isn't written until the connection
has been established (think: tunneling requests over HTTP CONNECT.)
Fixes#7296.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
In tls.connect a unix socket connection to a path may be made in
recent versions of node by specifying the value for the path
property.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
OpenSSL behaves oddly: on client `cert_chain` contains
the `peer_certificate`, but on server it doesn't.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
When creating a TLSSocket instance based on the existing connecting
socket, `_connecting` property is copied after the initialization of
`net.Socket`. However, since `net.Socket` constructor will call
`.read(0)` if the `readable` is true - error may happen at this code
chunk in net.js:
Socket.prototype._read = function(n) {
debug('_read');
if (this._connecting || !this._handle) {
debug('_read wait for connection');
this.once('connect', this._read.bind(this, n));
...
Leading to a test failures on windows:
- test/simple/test-tls-connect-given-socket.js
Signed-off-by: Fedor Indutny <fedor@indutny.com>
* isaacs, Robert Kowalski, Benjamin Coe: Test Improvements
* isaacs doc: Add canonical url
* isaacs view: handle unpublished packages properly
* Raynos (Jake Verbaten) do not log if silent
* Julian Gruber fix no such property error
* isaacs npmconf@0.1.14
* Thorsten Lorenz adding save-prefix configuration option
* isaacs npm-registry-client@0.4.7
* isaacs cache: treat missing versions as a 404
* isaacs cache: Save shasum, write resolved/etc data to cache
* isaacs cache: Always fetch root doc
* isaacs cache: don't repack unnecessarily from tmp
* Andrey Kislyuk Don't crash if shrinkwrap-dependencies were not passed in pkginfo
* Robert Kowalski fix link in faq
* Jean Lauliac Add a peerDependencies section in package.json doc
* isaacs read-installed@2.0.2
Oversight to not pass blksize to fs.Stats on initialization.
Also added a test to make sure the object property has been set. Since
now on Windows both blksize and blocks will simply be set to undefined.
Fix possible deadlock, when handles are sent in both direction
simultaneously. In such rare cases, both sides may queue their
`NODE_HANDLE_ACK` replies and wait for them.
fix#7465
Fix issue where a signed integer is returned.
Example:
var b = new Buffer(4);
b.writeUInt32BE(0xffffffff);
b.readUInt32BE(0) == -1
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
When our estimates for a storage size are higher than the actual length
of decoded data, the destination buffer should be truncated. Otherwise
`Buffer::Length` will give misleading information to C++ layer.
fix#7365
Signed-off-by: Fedor Indutny <fedor@indutny.com>
`process.uptime()` interface will return the amount of time the
current process has been running. To achieve this it was caching the
`uv_uptime` value at program start, and then on the call to
`process.uptime()` returning the delta between the two values.
`uv_uptime` is defined as the number of seconds the operating system
has been up since last boot. On sunos this interface uses `kstat`s
which can be a significantly expensive operation as it requires
exclusive access, but because of the design of `process.uptime()` node
*had* to always call this on start. As a result if you had many node
processes all starting at the same time you would suffer lock
contention as they all tried to read kstats.
Instead of using `uv_uptime` to achieve this, the libuv loop already
has a concept of current loop time in the form of `uv_now()` which is
in fact monotonically increasing, and already stored directly on the
loop. By using this value at start every platform performs at least
one fewer syscall during initialization.
Since the interface to `uv_uptime` is defined as seconds, in the call
to `process.uptime()` we now `uv_update_time` get our delta, divide by
1000 to get seconds, and then convert to an `Integer`. In 0.12 we can
move back to `Number::New` instead and not lose precision.
Caveat: For some platforms `uv_uptime` reports time monotonically
increasing regardless of system hibernation, `uv_now` interface is
also monotonically increasing but may not reflect time spent in
hibernation.
Introduce new signature for both `dgram.createSocket` method and
`dgram.Socket` constructor:
dgram.createSocket(options, [listener])
Options should contain `type` property and may contain `reuseAddr`
property. When `reuseAddr` is `true` - SO_REUSEADDR will be issued on
socket on bind.
fix#7415
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Socket may become not `readable`, but http should not rely on this
property and should not think that it means that no data will ever
arrive from it. In fact, it may arrive in a next tick and, since
`this.push(null)` was already called, it will result in a error like
this:
Error: stream.push() after EOF
at readableAddChunk (_stream_readable.js:143:15)
at IncomingMessage.Readable.push (_stream_readable.js:123:10)
at HTTPParser.parserOnBody (_http_common.js:132:22)
at Socket.socketOnData (_http_client.js:277:20)
at Socket.EventEmitter.emit (events.js:101:17)
at Socket.Readable.read (_stream_readable.js:367:10)
at Socket.socketCloseListener (_http_client.js:196:10)
at Socket.EventEmitter.emit (events.js:123:20)
at TCP.close (net.js:479:12)
fix#6784
This prevents segfaults when a native method is reassigned to a
different object (which corrupts args.This()). When unwrapping,
clients should use args.Holder() instead of args.This().
Closes#6690.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Increase the performance and simplify the logic of Buffer#write{U}Int*
and Buffer#read{U}Int* methods by placing the byte manipulation code
directly inline.
Also improve the speed of buffer-write benchmarks by creating a new
call directly to each method by using Function() instead of calling by
buff[fn].
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The two biggest changes are that v8::Script::New() has been removed and
that a v8::Script object now has to be explicitly bound to a context if
you want to run it from another context.
We can accommodate both changes without breaking the vm module's public
API or even the internal JS API.
The test/simple/test-smalloc.js has an implicit assumption
of the byte order of the data stored for Double and Uint32
values. On a big endian platform this test fails without
these patches.
Use os.endianness() to detect the endian of the platform
and use it to gate the static value used for comparison.