When a vm script aborted after a timeout/signal interruption, test
whether the local timeout/signal watchdog was responsible for
terminating the execution.
Without this, when a shorter timer from an outer `vm.run*` invocation
fires before an inner timeout, the inner timeout would throw an error
instead of the outer one, but because it did not witness the timeout
itself, it would assume the termination was the result of a signal
interruption.
PR-URL: https://github.com/nodejs/node/pull/7373
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
- Adds the `breakEvalOnSigint` option to `vm.runIn(This)Context`.
This uses a watchdog thread to wait for SIGINT and generally works
just like the existing `timeout` option.
- Adds a method to the existing timer-based watchdog to check if it
stopped regularly or by running into the timeout. This is used to
tell a SIGINT abort from a timer-based one.
- Adds (internal) `process._{start,stop}SigintWatchdog` methods to
start/stop the watchdog thread used by the above option manually.
This will be used in the REPL to set up SIGINT handling before
entering terminal raw mode, so that there is no time window in
which Ctrl+C fully aborts the process.
PR-URL: https://github.com/nodejs/node/pull/6635
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
For consistency with the newly added src/base64.h header, check that
NODE_WANT_INTERNALS is defined and set in internal headers.
PR-URL: https://github.com/nodejs/node/pull/6948
Refs: https://github.com/nodejs/node/pull/6910
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Remove the Watchdog class' dependency on Environment.
No functional changes, only code cleanup.
PR-URL: https://github.com/nodejs/node/pull/3274
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.
Eliminate a race condition between uv_async_send and the closing of the
corresponding handle.
Also made errors in Watchdog constructor call abort()
Fixes#6088
Previous code was calling uv_loop_delete() directly on a running loop,
which led to race condition aborts/segfaults within libuv. This change
changes the watchdog thread to call uv_run() with UV_RUN_ONCE so that
the call exits after either the timer times out or uv_async_send() is
called from the main thread in Watchdog::Destroy(). The timer/async
handles are then closed and uv_run() with UV_RUN_DEFAULT is called so
that libuv has a chance to cleanup before the thread exits. The main
thread meanwhile calls uv_thread_join() and then uv_loop_delete() to
complete the cleanup.
Add a watchdog class which executes a timer in a separate event loop in
a separate thread that will terminate v8 execution if it expires.
Add timeout argument to functions in vm module which use the watchdog
if a non-zero timeout is specified.
At the same time implement synchronous wrappers of the POSIX functions.
These will be undocumented until we settle on an API. Works like this
// returns promise as before
posix.mkdir("test").addCallback(function () {
sys.puts("done");
});
// returns undefined, executed synchronously.
posix.mkdirSync("test");
sys.puts("done");
This refactoring is a step towards allowing promises to be implemented
purely in javascript.
This is because it would call the javascript initializer which executed
Promise::New, and then it would rewrap the handle. Instead I make an
explicit inheritance from EIOPromise to Promise.
This seems to fix a memory leak which was reported by Ray Morgan:
http://groups.google.com/group/nodejs/browse_thread/thread/e38949b1989da1d7
All the c++ code is now reduced to simple wrappers. The node.fs.File object
is defined entirely in javascript now. As is the actionQueue methods.
This makes the boundaries much cleaner. There is still some thought that
needs to go into how exactly the API should behave but this simplification
is a first step.
This is sloppy: after each ObjectWrap allocation the user needs to
call ObjectWrap::InformV8ofAllocation(). In addition each class deriving
from ObjectWrap needs to implement the virtual method size() which should
return the size of the derived class. If I was better at C++ I could
possibly make this less ugly. For now this is how it is.
Memory usage looks much better after this commit.
Anna Henningsen
Ben Noordhuis
Ido Ben-Yair
isaacs
Fedor Indutny
Saúl Ibarra Corretgé
Alexis Campailla
Fedor Indutny
Andrew Paprocki
Ryan Dahl
Paul Querna
Connor Dunn