uv_close() is an asynchronous operation. Calling it on a data member
inside the destructor is unsound because its memory is about to be
reclaimed but libuv is not done with it yet.
PR-URL: https://github.com/nodejs/node/pull/7907
Reviewed-By: Ali Ijaz Sheikh <ofrobots@google.com>
Reviewed-By: Eugene Ostroukhov <eostroukhov@chromium.org>
Reviewed-By: James M Snell <jasnell@gmail.com>
Make the inspector code easier to reason about by restructuring it
to avoid manual memory allocation and copying as much as possible.
An amusing side effect is that it reduces the total amount of memory
used in the test suite.
Before:
$ valgrind ./out/Release/cctest 2>&1 | grep 'total heap' | cut -c31-
1,017 allocs, 1,017 frees, 21,695,456 allocated
After:
$ valgrind ./out/Release/cctest 2>&1 | grep 'total heap' | cut -c31-
869 allocs, 869 frees, 14,484,641 bytes allocated
PR-URL: https://github.com/nodejs/node/pull/7906
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Node process will no longer terminate with an assertion if the
inspector port is not available.
PR-URL: https://github.com/nodejs/node/pull/7874
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: jasnell - James M Snell <jasnell@gmail.com>
This commit attempts to address one of the items in #4641 which is
related to src/pipe_wrap.cc and src/tcp_wrap.cc. Currently both
pipe_wrap.cc and tcp_wrap.cc contain a class that are almost
identical. This commit extracts these parts into a separate class
that both can share.
PR-URL: https://github.com/nodejs/node/pull/7501
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Mac OSX 10.9 has switched to using libc++ by default. libc++
provides a C++11 <type_traits> implementation, so we only need
to use the TR1 version when targetting OSX 10.8 or 10.7.
PR-URL: https://github.com/nodejs/node/pull/7778
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
85af1a6 was added (squashed into another commit) without running
through CI. Unfortunately, it breaks some of the CI builds, notably on
three of the CentOS setups.
Undoing that one small change to get builds green again.
Refs: https://github.com/nodejs/node/pull/7547#issuecomment-235045450
PR-URL: https://github.com/nodejs/node/pull/7873
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Johan Bergström <bugs@bergstroem.nu>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
One of the issues in #4641 concerns OnConnection in pipe_wrap and
tcp_wrap which are very similar with some minor difference in how
they are coded. This commit extracts OnConnection so both these
classes can share the same implementation.
PR-URL: https://github.com/nodejs/node/pull/7547
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This patch
1. moves the basic validation of arguments to `truncate` family
of functions to the JavaScript layer from the C++ layer.
2. makes sure that the File Descriptors are validated strictly.
PR-URL: https://github.com/nodejs/node/pull/2498
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
API function callbacks run inside an implicit HandleScope. We don't
need to explicitly create one and in fact introduce some unnecessary
overhead when we do.
PR-URL: https://github.com/nodejs/node/pull/7711
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Create a handle scope before performing a check that creates a handle,
otherwise the handle is leaked into the handle scope of the caller.
PR-URL: https://github.com/nodejs/node/pull/7711
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Create a handle scope before performing a check that creates a handle,
otherwise the handle is leaked into the handle scope of the caller.
PR-URL: https://github.com/nodejs/node/pull/7711
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Fix handle leaks in Buffer::New() and Buffer::Copy() by creating the
handle scope before looking up the env with Environment::GetCurrent().
Environment::GetCurrent() calls v8::Isolate::GetCurrentContext(), which
creates a handle in the current scope, i.e., the scope created by the
caller of Buffer::New() or Buffer::Copy().
PR-URL: https://github.com/nodejs/node/pull/7711
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Current implementation tracks connected/disconnected status separately
which potentially introduces race condition.
This change introduces notion of session IDs and also posts
connect/disconnect events into the same queue as the messages. This way
Node knows what session given response belongs to and can discard
messages if the frontend for that session had disconnected.
This also fixes an issue when frontend was unable to attach to V8
instance that was running infinite loop.
PR-URL: https://github.com/nodejs/node/pull/7271
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
Disable stdio buffering, it interacts poorly with printf() calls from
elsewhere in the program (e.g., any logging from V8.) Unbreaks among
other things the `--trace_debug_json` switch.
Undoes commit 0966ab99 ("src: force line buffering for stderr"), which
in retrospect is not a proper fix. Turning on line buffering fixed a
flaky test on SmartOS but the test wasn't failing on other platforms,
where stderr wasn't line-buffered either. Mark the test flaky again,
it failed once in a run of 333 tries on the smartos-64 buildbot.
Disabling buffering should be safe even when mixed with non-blocking
stdio I/O because libuv goes to great lengths to reopen the tty file
descriptors and falls back to blocking I/O when that fails.
PR-URL: https://github.com/nodejs/node/pull/7610
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Many extensions are unknown to the `ClientHelloParser::ParseExtension`,
do not cast user-supplied `uint16_t` to `enum`.
PR-URL: https://github.com/nodejs/node/pull/7494
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
`offset` is user supplied variable and may be bigger than
`ts_obj_length`. There is no need to subtract them and pass along, so
just throw when the subtraction result would overflow.
PR-URL: https://github.com/nodejs/node/pull/7494
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Coverity marked a change in 630096b as a constant expression.
However, on platforms where sizeof(int64_t) > sizeof(size_t),
this should not be the case. This commit flags the comparison
as OK to coverity.
PR-URL: https://github.com/nodejs/node/pull/7587
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
ParseArrayIndex() would wrap around large (>=2^32) index values on
platforms where sizeof(int64_t) > sizeof(size_t). Ensure that the
return value fits in a size_t.
PR-URL: https://github.com/nodejs/node/pull/7497
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
It's not used anywhere else so move it out of src/node_internals.h.
PR-URL: https://github.com/nodejs/node/pull/7497
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
`--debug=1.2.3.4:5678` and `--debug=example.com:5678` are now accepted,
likewise the `--debug-brk` and `--debug-port` switch. The latter is
now something of a misnomer but it's undocumented and for internal use
only so it shouldn't matter too much.
`--inspect=1.2.3.4:5678` and `--inspect=example.com:5678` are also
accepted but don't use the host name yet; they still bind to the
default address.
Fixes: https://github.com/nodejs/node/issues/3306
PR-URL: https://github.com/nodejs/node/pull/3316
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
In `AppendExceptionLine()`, which is used both by the `vm`
module and the uncaught exception handler, don’t print anything
to stderr when called from the `vm` module, even if the
thrown object is not a native error instance.
Fixes: https://github.com/nodejs/node/issues/7397
PR-URL: https://github.com/nodejs/node/pull/7398
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
ICU has a punycode implementation built in. Use it instead of the
javascript implementation because it's much faster.
PR-URL: https://github.com/nodejs/node/pull/7355
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Inspector socket implementation was notifying handshake callback before
performing the cleanups, which meant that callback could not reclaim
resources allocated by the client. New implementation will free all
resource not allocated by the client before calling the callback,
allowing the client to complete the cleanup.
Fixes: https://github.com/nodejs/node/pull/7418
PR-URL: https://github.com/nodejs/node/pull/7450
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
Print a C backtrace on fatal errors to make it easier to debug issues.
PR-URL: https://github.com/nodejs/node/pull/6734
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Don't inline calls to node::DumpBacktrace() and fflush(), it makes the
generated code bigger. A secondary benefit of moving it to a function
is that it gives you something to put a breakpoint on.
PR-URL: https://github.com/nodejs/node/pull/6734
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Print a C backtrace on fatal errors to make it easier to debug issues.
PR-URL: https://github.com/nodejs/node/pull/6734
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Pointed out by Coverity. Introduced in commits 3546383c ("process_wrap:
avoid leaking memory when throwing due to invalid arguments") and
fa4eb47c ("bindings: add spawn_sync bindings").
The return statements inside the if blocks were dead code because their
guard conditions always evaluated to false. Remove them.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Pointed out by Coverity. Introduced in commit 05d30d53 from July 2015
("fs: implemented WriteStream#writev").
WriteBuffers() leaked memory in the synchronous uv_fs_write() error path
when trying to write > 1024 buffers.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Pointed out by Coverity. Introduced in commit 5b8e1dab from September
2011 ("Initial pass at zlib bindings".)
The asynchronous version of Write() used a pointer to a stack-allocated
buffer on flush. A mitigating factor is that zlib does not dereference
the pointer for zero-sized writes but it's still technically UB.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
The code assigned the result of EVP_get_digestbyname() to data members
called md_ that were not used outside the initialization functions.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Remove TLSWrap::write_queue_size_, it's not used anywhere.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
This commit adds a CHECK that verifies that the file event watcher is
not started twice, which would be indicative of a bug in lib/fs.js.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Pointed out by Coverity. Not really a bug because it's assigned before
use but explicit assignment in the constructor is more obviously
correct.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Pointed out by Coverity.
PR-URL: https://github.com/nodejs/node/pull/7374
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>