fsevents: fix clever rescheduling

There're could be a situation, where one fsevents handle gets created
and another one is destroyed simultaneously. In such cases
`fsevent_need_resch

fsevents: fix clever rescheduling

There're could be a situation, where one fsevents handle gets created
and another one is destroyed simultaneously. In such cases
`fsevent_need_reschedule` will be set to 1 twice and reset only once,
leaving handle destructor hanging in uv_sem_wait().

show more ...

unix: revert recent FSEvent changes

This commit reverts the following commits:

983fa68 darwin: fix 10.6 build error in fsevents.c
684e212 fsevents: use shared FSEventStr

unix: revert recent FSEvent changes

This commit reverts the following commits:

983fa68 darwin: fix 10.6 build error in fsevents.c
684e212 fsevents: use shared FSEventStream
ea4cb77 fsevents: FSEvents is most likely not thread-safe
9bae606 darwin: create fsevents thread on demand

Several people have reported stability issues on OS X 10.8 and bus
errors on the 10.9 developer preview.

See also joyent/node#6296 and joyent/node#6251.

show more ...

unix: don't close inherited fds on uv_spawn() fail

The cleanup-after-error code path in uv_spawn() was closing file
descriptors indiscriminately. Only close file descriptors that we

unix: don't close inherited fds on uv_spawn() fail

The cleanup-after-error code path in uv_spawn() was closing file
descriptors indiscriminately. Only close file descriptors that we
created ourselves, not the ones that are passed in by the user.

Fixes joyent/node#6297.

show more ...

unix: fix uv_spawn() NULL pointer deref on ENOMEM

In the cleanup-after-error section of uv_spawn(), check that the pointer
is non-NULL - we might end up in said section due to a malloc()

unix: fix uv_spawn() NULL pointer deref on ENOMEM

In the cleanup-after-error section of uv_spawn(), check that the pointer
is non-NULL - we might end up in said section due to a malloc() failure.

show more ...

unix: remove uv__pipe_accept()

It's basically a less advanced version of uv__server_io(). Drop the
former in favor of the latter.

unix: sanity-check fds before closing

Ensure that close() system calls don't close stdio file descriptors
because that is almost never the intention.

This is also a partial work

unix: sanity-check fds before closing

Ensure that close() system calls don't close stdio file descriptors
because that is almost never the intention.

This is also a partial workaround for a kernel bug that seems to affect
all Linux kernels when stdin is a pipe that gets closed: fd 0 keeps
signalling EPOLLHUP but a subsequent call to epoll_ctl(EPOLL_CTL_DEL)
fails with EBADF. See joyent/node#6271 for details and a test case.

show more ...

Merge remote-tracking branch 'origin/v0.10'

Conflicts:
ChangeLog
include/uv-darwin.h
src/unix/darwin.c
src/unix/fsevents.c
src/ver

Merge remote-tracking branch 'origin/v0.10'

Conflicts:
ChangeLog
include/uv-darwin.h
src/unix/darwin.c
src/unix/fsevents.c
src/version.c
src/win/fs.c

show more ...

Revert "unix: set O_NONBLOCK in uv_pipe_open()"

It turns out that node.js relies on the blocking behavior of pipes in
some cases, notably when forking worker processes. Reopens #941.

Revert "unix: set O_NONBLOCK in uv_pipe_open()"

It turns out that node.js relies on the blocking behavior of pipes in
some cases, notably when forking worker processes. Reopens #941.

This reverts commit 8fe4ca686bcb069f670b0381e89c008ca814f8ba.

show more ...

unix: set O_NONBLOCK in uv_pipe_open()

Don't rely on the caller to set the O_NONBLOCK flag on the file
descriptor.

Prevents sporadic stalls when the file descriptor is in blocki

unix: set O_NONBLOCK in uv_pipe_open()

Don't rely on the caller to set the O_NONBLOCK flag on the file
descriptor.

Prevents sporadic stalls when the file descriptor is in blocking mode
and exactly as many bytes are read as there are available; in that case,
libuv will try to read again and block. Node.js was guilty of this.

Fixes #941.

show more ...

windows: remove duplicate check in stream.c

Fixes #937.

Now working on v0.10.18

2013.09.25, Version 0.10.17 (Stable)

Changes since version 0.10.16:

* build: remove GCC_WARN_ABOUT_MISSING_NEWLINE (Ben Noordhuis)

* darwin: fix 10.6 build error in fsevent

2013.09.25, Version 0.10.17 (Stable)

Changes since version 0.10.16:

* build: remove GCC_WARN_ABOUT_MISSING_NEWLINE (Ben Noordhuis)

* darwin: fix 10.6 build error in fsevents.c (Ben Noordhuis)

show more ...

doc: update readme

Update the intro blurb and spend a few more words on libuv's features.

build: use -pthread on sunos

When building on sunos with autoconf make sure to specify -pthread
otherwise there will be race conditions with errno and etc

bench: measure total/init/dispatch/cleanup times

* Add finer-grained measurements to the million_timers benchmark.
Before this commit it only measured the total running time.

bench: measure total/init/dispatch/cleanup times

* Add finer-grained measurements to the million_timers benchmark.
Before this commit it only measured the total running time.

* Bump the number of inserted timers from 1M to 10M. With one million
timers, it finishes too quickly to get useful profiling data out of
it with perf or oprofile. The name of the benchmark is now a lie but
such is life.

show more ...

unix, windows: make uv_is_*() always return 0 or 1

Ensure that the following API functions always return either 0 or 1:

* uv_is_active()
* uv_is_closing()
* uv_is_read

unix, windows: make uv_is_*() always return 0 or 1

Ensure that the following API functions always return either 0 or 1:

* uv_is_active()
* uv_is_closing()
* uv_is_readable()
* uv_is_writable()

show more ...

unix: wrap long lines at 80 columns

test: wrap long lines at 80 columns

include: make uv_process_options_t.cwd const

There is no reason for it to be non-const, libuv treats it as a
read-only input. Make it a pointer to a const string.

include: update uv_is_active() documentation

Seems to have been overlooked in the reference counting refactor back
in May 2012 in commit 9efa8b3. Clarify what "active" means for
dif

include: update uv_is_active() documentation

Seems to have been overlooked in the reference counting refactor back
in May 2012 in commit 9efa8b3. Clarify what "active" means for
different kinds of handles.

show more ...

include: fix typo in comment in uv.h

include: remove UV_IO_PRIVATE_FIELDS macro

It's an alias for UV_IO_PRIVATE_PLATFORM_FIELDS. Remove the macro and
use UV_IO_PRIVATE_PLATFORM_FIELDS directly.

include: clean up includes in uv.h

* size_t is defined in <stddef.h>, not <sys/types.h>

* Simplify the platform check.

include: clarify uv_tcp_bind() behavior

On BSD-like platforms, EADDRINUSE is returned by the bind() system call.
On other platforms, it's returned by the listen() system call.

I

include: clarify uv_tcp_bind() behavior

On BSD-like platforms, EADDRINUSE is returned by the bind() system call.
On other platforms, it's returned by the listen() system call.

In other words, some platforms are 'first to bind wins', others are
'first to listen wins' - but only with TCP sockets: UNIX domain sockets
always return EADDRINUSE from the bind() system call, UDP sockets don't
call listen() in the first place.

Fixes #769.

show more ...

windows: run close callbacks after polling for i/o

Makes the uv-win behavior compatible with uv-unix.

Fixes #796.