* fixes #2110
* includes V8 postmortem metadata in Solaris builds
* adds GYP support for DTrace probes and ustack helper
* ustack helper derives constants dynamically from libv8_base.a
* build with DTrace support by default on SunOS

_snprintf() doesn't zero-terminate the buffer on overflow.

It returns void on some platforms, notably FreeBSD.

In this case, backwards compatibility is not worth the API
inconsistency.  We can just document the change.

This reverts commit b521ff3b.

The locker makes node crash in debug mode sometimes.
For example, test/simple/test-repl.js triggers it.

This reverts commit 9a6012ed.

Conflicts:

	src/node.cc

* Upgrade V8 to 3.9.24.7

* Upgrade npm to 1.1.15

* Handle Emoji characters properly (Erik Corry, Bert Belder)

* readline: migrate ansi/vt100 logic from tty to readline (Nathan Rajlich)

* readline: Fix multiline handling (Alex Kocharin)

* add a -i/--interactive flag to force the REPL (Nathan Rajlich)

* debugger: add breakOnException command (Fedor Indutny)

* cluster: kill workers when master dies (Andreas Madsen)

* cluster: add graceful disconnect support (Andreas Madsen)

* child_process: Separate 'close' event from 'exit' (Charlie McConnell)

* typed arrays: add Uint8ClampedArray (Mikael Bourges-Sevenier)

* buffer: Fix byte alignment issues (Ben Noordhuis, Erik Lundin)

* tls: fix CryptoStream.setKeepAlive() (Shigeki Ohtsu)

* Expose http parse error codes (Felix Geisendörfer)

* events: don't delete the listeners array (Ben Noordhuis, Nathan Rajlich)

* process: add process.config to view node's ./configure settings (Nathan Rajlich)

* process: process.exe...

Technically saying `tty.ReadStream#setRawMode()` is correct,
but since a typical use cannot instantiate `tty.ReadStream` themselves,
and 99% of the time the only instance is `process.stdin`,
then a little clarification seemed necessary.

Travis-CI is failing on this test repeatedly, but other Linux systems
seem to be fine with it.  Alter the logging so it's more clear which
part is timing out.

Said test takes over 90 seconds on my Core 2 Duo which is too long, even for
the pummel tests.

Instead of allocating a new 64KB buffer each time when checking if there is
something to transform, continue to use the same buffer. Once the buffer is
exhausted, allocate a new buffer. This solves the problem of huge allocations
when small fragments of data are processed, but will also continue to work
well with big pieces of data.

The test is too slow / CPU intensive for the main test harness.