darwin: create fsevents thread on demand

* Move CF run loop code to fsevents.c.

* Create the fsevents thread on demand rather than at startup.

* Remove use of ACCESS_ONCE.

darwin: create fsevents thread on demand

* Move CF run loop code to fsevents.c.

* Create the fsevents thread on demand rather than at startup.

* Remove use of ACCESS_ONCE. All accesses to loop->cf_loop are
protected by full memory barriers so no reordering can take place.

Fixes #872.

Conflicts:
src/unix/darwin.c

show more ...

fsevents: use shared FSEventStream

It seems that number of simultaneously opened FSEventStreams is
limited on OSX (i.e. you can have only fixed number of them on
one running system),

fsevents: use shared FSEventStream

It seems that number of simultaneously opened FSEventStreams is
limited on OSX (i.e. you can have only fixed number of them on
one running system), getting past through this limit will cause
`FSEventStreamCreate` to return false and write following message
to stderr:

(CarbonCore.framework) FSEventStreamStart: register_with_server:
ERROR: f2d_register_rpc() => (null) (-21)

To prevent this, we must use only one shared FSEventStream with a
paths for all uv_fsevent_t handles, and then filter out events for
each handle using this paths again.

See https://github.com/joyent/node/issues/5463

show more ...

Now working on v0.10.15

2013.08.22, Version 0.10.14 (Stable)

Changes since version 0.10.13:

* unix: retry waitpid() on EINTR (Ben Noordhuis)

Now working on v0.11.9

2013.08.22, Version 0.11.8 (Unstable)

Changes since version 0.11.7:

* unix: fix missing return value warning in stream.c (Ben Noordhuis)

* build: serial-tests was added in

2013.08.22, Version 0.11.8 (Unstable)

Changes since version 0.11.7:

* unix: fix missing return value warning in stream.c (Ben Noordhuis)

* build: serial-tests was added in automake v1.12 (Ben Noordhuis)

* windows: fix uninitialized local variable warning (Ben Noordhuis)

* windows: fix missing return value warning (Ben Noordhuis)

* build: fix string comparisons in autogen.sh (Ben Noordhuis)

* windows: move INLINE macro, remove UNUSED (Ben Noordhuis)

* unix: clean up __attribute__((quux)) usage (Ben Noordhuis)

* sunos: remove futimes() macro (Ben Noordhuis)

* unix: fix uv__signal_unlock() prototype (Ben Noordhuis)

* unix, windows: allow NULL async callback (Ben Noordhuis)

* build: apply dtrace -G to all object files (Timothy J. Fontaine)

* darwin: fix indentation in uv__hrtime() (Ben Noordhuis)

* darwin: create fsevents thread on demand (Ben Noordhuis)

* darwin: reduce fsevents thread stack size (Ben Noordhuis)

* darwin: call pthread_setname_np() if available (Ben Noordhuis)

* build: fix automake serial-tests check again (Ben Noordhuis)

* unix: retry waitpid() on EINTR (Ben Noordhuis)

* darwin: fix ios build error (Ben Noordhuis)

* darwin: fix ios compiler warning (Ben Noordhuis)

* test: simplify test-ip6-addr.c (Ben Noordhuis)

* unix, windows: fix ipv6 link-local address parsing (Ben Noordhuis)

* fsevents: FSEvents is most likely not thread-safe (Fedor Indutny)

* windows: omit stdint.h, fix msvc 2008 build error (Ben Noordhuis)

show more ...

windows: omit stdint.h, fix msvc 2008 build error

Don't include <stdint.h>, it's not available when compiling with MSVC
2008 and we don't need it in the first place - just cast the argum

windows: omit stdint.h, fix msvc 2008 build error

Don't include <stdint.h>, it's not available when compiling with MSVC
2008 and we don't need it in the first place - just cast the argument
to `char *` rather than `uintptr_t`.

Fixes #893.

show more ...

fsevents: FSEvents is most likely not thread-safe

Perform all operation with FSEventStream in the same thread, where it'll
be used.

unix, windows: fix ipv6 link-local address parsing

uv_ip6_addr() copies the address part to a temporary buffer when it
contains a link-local suffix ('<address>%<interface>'). Before this

unix, windows: fix ipv6 link-local address parsing

uv_ip6_addr() copies the address part to a temporary buffer when it
contains a link-local suffix ('<address>%<interface>'). Before this
commit, it didn't zero-terminate the address properly: it put the nul
byte at the end of the temporary buffer rather than at the end of the
address string, meaning the buffer looked like this:

<address> <garbage> '\0'

Fixes the following valgrind warning:

==16170== Conditional jump or move depends on uninitialised value(s)
==16170== at 0x43602C: inet_pton6 (inet.c:228)
==16170== by 0x435CE1: uv_inet_pton (inet.c:163)
==16170== by 0x436FD0: uv_ip6_addr (uv-common.c:175)
==16170== by 0x434717: test_ip6_addr_scope (test-ip6-addr.c:89)
==16170== by 0x43455B: call_iface_info_cb (test-ip6-addr.c:45)
==16170== by 0x43462B: foreach_ip6_interface (test-ip6-addr.c:59)
==16170== by 0x434791: run_test_ip6_addr_link_local (test-ip6-add
==16170== by 0x4061E8: run_test_part (runner.c:396)
==16170== by 0x404F4B: main (run-tests.c:58)
==16170==
==16170== Conditional jump or move depends on uninitialised value(s)
==16170== at 0x4C2AD8A: __GI_strchr (mc_replace_strmem.c:224)
==16170== by 0x435ECB: inet_pton6 (inet.c:231)
==16170== by 0x435CE1: uv_inet_pton (inet.c:163)
==16170== by 0x436FD0: uv_ip6_addr (uv-common.c:175)
==16170== by 0x434717: test_ip6_addr_scope (test-ip6-addr.c:89)
==16170== by 0x43455B: call_iface_info_cb (test-ip6-addr.c:45)
==16170== by 0x43462B: foreach_ip6_interface (test-ip6-addr.c:59)
==16170== by 0x434791: run_test_ip6_addr_link_local (test-ip6-add
==16170== by 0x4061E8: run_test_part (runner.c:396)
==16170== by 0x404F4B: main (run-tests.c:58)

Fixes #890.

show more ...

test: simplify test-ip6-addr.c

Drop the callback-driven approach in favor of a simple loop. Reduces
the line count and makes the flow of execution a little easier to
follow.

darwin: fix ios compiler warning

uv__fsevents_loop_delete() returns void. Remove the 'return 0'
statement.

darwin: fix ios build error

Include <errno.h> and <stdlib.h>. They're being pulled in implicitly
on OS X but apparently that's not the case with iOS builds.

Build breakage intr

darwin: fix ios build error

Include <errno.h> and <stdlib.h>. They're being pulled in implicitly
on OS X but apparently that's not the case with iOS builds.

Build breakage introduced in 5ff6f85 by yours truly. Mea culpa.

Fixes #885 and #891.

show more ...

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

unix: retry waitpid() on EINTR

Before this commit, libuv would abort() if waitpid() failed with EINTR.

It's unlikely that anyone actually hit this error condition: the major
UNI

unix: retry waitpid() on EINTR

Before this commit, libuv would abort() if waitpid() failed with EINTR.

It's unlikely that anyone actually hit this error condition: the major
UNIX platforms - with the possible exception of Solaris - don't return
EINTR when the WNOHANG flag is specified, as libuv does.

However, POSIX allows for an implementation to do whatever here: unless
explicitly forbidden, it's allowed and POSIX doesn't restrict
implementers in this particular area.

Let's opt for robustness and handle EINTR.

show more ...

build: fix automake serial-tests check again

This effectively undoes the following commits:

a97685e build: add automake serial-tests version check
e4c1483 build: serial-

build: fix automake serial-tests check again

This effectively undoes the following commits:

a97685e build: add automake serial-tests version check
e4c1483 build: serial-tests was added in automake v1.12

After much trial and error I've come to the conclusion that you cannot
reliably test for the automake version inside configure.ac itself.

Feature checks (testing for the presence of macros with m4_ifdef) is
not reliable when the macro is "lazy-loaded" by aclocal: m4 won't see
its definition unless it's actually used in configure.ac, hence checking
for obsolete macros like AM_ENABLE_MULTILIB and AM_WITH_REGEX is not a
reliable proxy for the automake version - both are are lazy-loaded.

That's why this commit moves the version check to autogen.sh, creates
a m4 file with automake options on the fly and includes that in
configure.ac.

Thank you, automake maintainers, for making hard what should be easy.
That's an hour of my life I won't be getting back!

show more ...

darwin: call pthread_setname_np() if available

When setting the process title, also call pthread_setname_np() when
supported (OS X >= 10.6 and iOS >= 3.2.)

darwin: reduce fsevents thread stack size

The fsevents watcher thread only needs a minimal amount of stack space
in order to run. Restrict its stack size so there's more address space

darwin: reduce fsevents thread stack size

The fsevents watcher thread only needs a minimal amount of stack space
in order to run. Restrict its stack size so there's more address space
left for the libuv user.

show more ...

darwin: create fsevents thread on demand

* Move CF run loop code to fsevents.c.

* Create the fsevents thread on demand rather than at startup.

* Remove use of ACCESS_ONCE.

darwin: create fsevents thread on demand

* Move CF run loop code to fsevents.c.

* Create the fsevents thread on demand rather than at startup.

* Remove use of ACCESS_ONCE. All accesses to loop->cf_loop are
protected by full memory barriers so no reordering can take place.

Fixes #872.

show more ...

darwin: fix indentation in uv__hrtime()

build: apply dtrace -G to all object files

fixes #871

unix, windows: allow NULL async callback

Allow a NULL callback so the user doesn't have to provide a dummy when
the actual event is processed by e.g. a check handle callback.

unix: fix uv__signal_unlock() prototype

sunos: remove futimes() macro

Remove a macro from src/unix/internal.h that aliases futimes() to
futimesat(). Deal with the platform inconsistency in src/unix/fs.c.

unix: clean up __attribute__((quux)) usage

Macro-ify __attribute__((destructor)) and __attribute__((unused)).

The macros are no-ops when the compiler does not support function
a

unix: clean up __attribute__((quux)) usage

Macro-ify __attribute__((destructor)) and __attribute__((unused)).

The macros are no-ops when the compiler does not support function
attributes.

show more ...

windows: move INLINE macro, remove UNUSED

Move the INLINE macro from src/uv-common.h to src/win/internal.h, it's
not used in src/unix/ or src/

Remove the UNUSED macro. It's, wel

windows: move INLINE macro, remove UNUSED

Move the INLINE macro from src/uv-common.h to src/win/internal.h, it's
not used in src/unix/ or src/

Remove the UNUSED macro. It's, well, unused.

show more ...