2019.02.11, Version 1.26.0 (Stable)

Changes since version 1.25.0:

* doc: fix uv_get_free_memory doc (Stephen Belanger)

* unix: fix epoll cpu 100% issue (yeyuanfeng)

2019.02.11, Version 1.26.0 (Stable)

Changes since version 1.25.0:

* doc: fix uv_get_free_memory doc (Stephen Belanger)

* unix: fix epoll cpu 100% issue (yeyuanfeng)

* openbsd,tcp: special handling of EINVAL on connect (ptlomholt)

* win: simplify registry closing in uv_cpu_info() (cjihrig)

* src,include: define UV_MAXHOSTNAMESIZE (cjihrig)

* win: return product name in uv_os_uname() version (cjihrig)

* thread: allow specifying stack size for new thread (Anna Henningsen)

* win: fix duplicate tty vt100 fn key (erw7)

* unix: don't attempt to invalidate invalid fd (Ben Noordhuis)

show more ...

unix: don't attempt to invalidate invalid fd

Add a missing check in uv__io_close() where it called
uv__platform_invalidate_fd() without checking that the
watcher actually has a valid

unix: don't attempt to invalidate invalid fd

Add a missing check in uv__io_close() where it called
uv__platform_invalidate_fd() without checking that the
watcher actually has a valid file descriptor assigned.

Fixes: https://github.com/libuv/libuv/issues/2181
PR-URL: https://github.com/libuv/libuv/pull/2182
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

win: fix duplicate tty vt100 fn key

Refs: https://github.com/libuv/libuv/pull/2114
Refs: https://github.com/nodejs/node/issues/25875
Refs: https://github.com/nodejs/node/issues/26013

win: fix duplicate tty vt100 fn key

Refs: https://github.com/libuv/libuv/pull/2114
Refs: https://github.com/nodejs/node/issues/25875
Refs: https://github.com/nodejs/node/issues/26013
Fixes: https://github.com/libuv/libuv/issues/2168
PR-URL: https://github.com/libuv/libuv/pull/2160
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

thread: allow specifying stack size for new thread

PR-URL: https://github.com/libuv/libuv/pull/2179
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé

thread: allow specifying stack size for new thread

PR-URL: https://github.com/libuv/libuv/pull/2179
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>

show more ...

win: return product name in uv_os_uname() version

Currently, on Windows the uv_utsname_t's version field can
be an empty string if no service packs are installed. This isn't
very hel

win: return product name in uv_os_uname() version

Currently, on Windows the uv_utsname_t's version field can
be an empty string if no service packs are installed. This isn't
very helpful, and a lot more information is available in the
Windows registry. This commit prepends the full product name
to the existing service pack information.

Refs: https://github.com/nodejs/node/issues/25843
Refs: https://github.com/libuv/libuv/pull/2128#issuecomment-452109129
PR-URL: https://github.com/libuv/libuv/pull/2170
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>

show more ...

src,include: define UV_MAXHOSTNAMESIZE

This commit adds UV_MAXHOSTNAMESIZE for working with
uv_os_gethostname(). Prior to this commit, this logic was
duplicated in several places acr

src,include: define UV_MAXHOSTNAMESIZE

This commit adds UV_MAXHOSTNAMESIZE for working with
uv_os_gethostname(). Prior to this commit, this logic was
duplicated in several places across libuv and Node.js alone.

PR-URL: https://github.com/libuv/libuv/pull/2175
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>

show more ...

win: simplify registry closing in uv_cpu_info()

RegCloseKey() is the first thing executed in both the success
and error case, so combine the calls.

PR-URL: https://github.com/li

win: simplify registry closing in uv_cpu_info()

RegCloseKey() is the first thing executed in both the success
and error case, so combine the calls.

PR-URL: https://github.com/libuv/libuv/pull/2171
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>

show more ...

openbsd,tcp: special handling of EINVAL on connect

Map `EINVAL` to `ECONNREFUSED` and return error on the next tick.

Fixes: https://github.com/libuv/libuv/issues/2155
PR-URL: ht

openbsd,tcp: special handling of EINVAL on connect

Map `EINVAL` to `ECONNREFUSED` and return error on the next tick.

Fixes: https://github.com/libuv/libuv/issues/2155
PR-URL: https://github.com/libuv/libuv/pull/2154
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

show more ...

unix: fix epoll cpu 100% issue

Fixes: https://github.com/libuv/libuv/issues/2162
PR-URL: https://github.com/libuv/libuv/pull/2166
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>

unix: fix epoll cpu 100% issue

Fixes: https://github.com/libuv/libuv/issues/2162
PR-URL: https://github.com/libuv/libuv/pull/2166
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Li YuBei <leeight@gmail.com>
Reviewed-By: Jameson Nash <vtjnash@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

show more ...

doc: fix uv_get_free_memory doc

There was a formatting error that prevented the uv_get_free_memory
function from appearing on the page.

PR-URL: https://github.com/libuv/libuv/pu

doc: fix uv_get_free_memory doc

There was a formatting error that prevented the uv_get_free_memory
function from appearing on the page.

PR-URL: https://github.com/libuv/libuv/pull/2163
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>

show more ...

Now working on version 1.25.1

Fixes: https://github.com/libuv/libuv/issues/2134

Add SHA to ChangeLog

2019.01.19, Version 1.25.0 (Stable)

Changes since version 1.24.1:

* Revert "win,fs: retry if uv_fs_rename fails" (Ben Noordhuis)

* aix: manually trigger fs event monitoring

2019.01.19, Version 1.25.0 (Stable)

Changes since version 1.24.1:

* Revert "win,fs: retry if uv_fs_rename fails" (Ben Noordhuis)

* aix: manually trigger fs event monitoring (Gireesh Punathil)

* unix: rename WRITE_RETRY_ON_ERROR macro (Ben Noordhuis)

* darwin: DRY platform-specific error check (Ben Noordhuis)

* unix: refactor uv__write() (Ben Noordhuis)

* unix: don't send handle twice on partial write (Ben Noordhuis)

* tty,win: fix Alt+key under WSL (Bartosz Sosnowski)

* build: support running tests in out-of-tree builds (Jameson Nash)

* fsevents: really watch files with fsevents on macos 10.7+ (Jameson
Nash)

* thread,mingw64: need intrin.h header for SSE2 MemoryBarrier (Jameson
Nash)

* win: fix sizeof-pointer-div warning (cjihrig)

* unix,win: add uv_os_uname() (cjihrig)

* win, tty: fix CreateFileW() return value check (Bartosz Sosnowski)

* unix: enable IPv6 tests on OpenBSD (ptlomholt)

* test: fix test-ipc spawn_helper exit_cb (Santiago Gimeno)

* test: fix test-ipc tests (Santiago Gimeno)

* unix: better handling of unsupported F_FULLFSYNC (Victor Costan)

* win,test: de-flake fs_event_watch_dir_short_path (Refael Ackermann)

* win: fix msvc warning (sid)

* openbsd: switch to libuv's barrier implementation (ptlomholt)

* unix,stream: fix zero byte writes (Santiago Gimeno)

* ibmi: return EISDIR on read from directory fd (Kevin Adler)

* build: wrap long lines in Makefile.am (cjihrig)

show more ...

Revert "ibmi: add support for IBM i to cmake build"

This reverts commit 8d62d933ef31fd47ae269395364a280a12eb9636.
The corresponding autotools update broke the CI. This commit
is reve

Revert "ibmi: add support for IBM i to cmake build"

This reverts commit 8d62d933ef31fd47ae269395364a280a12eb9636.
The corresponding autotools update broke the CI. This commit
is reverted for consistency.

Refs: https://github.com/libuv/libuv/pull/2148
PR-URL: https://github.com/libuv/libuv/pull/2156
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>

show more ...

Revert "ibmi: add support for IBM i to autotools build"

This reverts commit ca02c36c6ddcf1f74215a63b3fbff04c1a79a1c0.
This broke SmartOS compilation in the CI.

Refs: https://git

Revert "ibmi: add support for IBM i to autotools build"

This reverts commit ca02c36c6ddcf1f74215a63b3fbff04c1a79a1c0.
This broke SmartOS compilation in the CI.

Refs: https://github.com/libuv/libuv/pull/2148
PR-URL: https://github.com/libuv/libuv/pull/2156
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>

show more ...

build: wrap long lines in Makefile.am

PR-URL: https://github.com/libuv/libuv/pull/2151
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoor

build: wrap long lines in Makefile.am

PR-URL: https://github.com/libuv/libuv/pull/2151
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

show more ...

ibmi: add support for IBM i to autotools build

PR-URL: https://github.com/libuv/libuv/pull/2148
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmai

ibmi: add support for IBM i to autotools build

PR-URL: https://github.com/libuv/libuv/pull/2148
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

ibmi: add support for IBM i to cmake build

PR-URL: https://github.com/libuv/libuv/pull/2148
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.co

ibmi: add support for IBM i to cmake build

PR-URL: https://github.com/libuv/libuv/pull/2148
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

ibmi: return EISDIR on read from directory fd

On IBM i PASE, EOPNOTSUPP is returned when reading a directory instead
of EISDIR, like (seemingly) every other platform which doesn't suppor

ibmi: return EISDIR on read from directory fd

On IBM i PASE, EOPNOTSUPP is returned when reading a directory instead
of EISDIR, like (seemingly) every other platform which doesn't support
reading directories. To ensure compatibility with software expecting
EISDIR, we map the EOPNOTSUPP to EISDIR when the fd passed in was a
directory.

This is a partial revert of 25a3894, but scoped to PASE and the fstat
call is moved to the end so it's out of the fast path.

Refs: https://github.com/nodejs/node/issues/25433
Fixes: https://github.com/libuv/libuv/issues/2147
PR-URL: https://github.com/libuv/libuv/pull/2148
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

unix,stream: fix zero byte writes

Fixes a regression where a write request to write a zero byte buffer
would never complete.

Refs: https://github.com/libuv/libuv/pull/2097
R

unix,stream: fix zero byte writes

Fixes a regression where a write request to write a zero byte buffer
would never complete.

Refs: https://github.com/libuv/libuv/pull/2097
Refs: https://github.com/libuv/libuv/issues/2134
PR-URL: https://github.com/libuv/libuv/pull/2149
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

show more ...

openbsd: switch to libuv's barrier implementation

On OpenBSD 6.4 x86_64 the barrier_serial_thread test fails due to
OpenBSD's pthread_barrier_wait() implementation returning
PTHREAD_

openbsd: switch to libuv's barrier implementation

On OpenBSD 6.4 x86_64 the barrier_serial_thread test fails due to
OpenBSD's pthread_barrier_wait() implementation returning
PTHREAD_BARRIER_SERIAL_THREAD from the first thread that releases the
wait. Switch to libuv's internal implementation to address the issue.

Fixes: https://github.com/libuv/libuv/issues/2145
PR-URL: https://github.com/libuv/libuv/pull/2144
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>

show more ...

win: fix msvc warning

PR-URL: https://github.com/libuv/libuv/pull/2091
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By:

win: fix msvc warning

PR-URL: https://github.com/libuv/libuv/pull/2091
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Jameson Nash <vtjnash@gmail.com>

show more ...

win,test: de-flake fs_event_watch_dir_short_path

New versions of Windows ship with 8.3 short-names disabled.
This commit adds 8.3 detection logic in the
fs_event_watch_dir_short_path

win,test: de-flake fs_event_watch_dir_short_path

New versions of Windows ship with 8.3 short-names disabled.
This commit adds 8.3 detection logic in the
fs_event_watch_dir_short_path test.

PR-URL: https://github.com/libuv/libuv/pull/2103
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...

unix: better handling of unsupported F_FULLFSYNC

Apple's released source code at https://opensource.apple.com/ shows at
least three different error codes being returned when a filesyste

unix: better handling of unsupported F_FULLFSYNC

Apple's released source code at https://opensource.apple.com/ shows at
least three different error codes being returned when a filesystem does
not support F_FULLFSYNC.

fcntl() is implemented in xnu-4903.221.2 in bsd/kern/kern_descrip.c,
where it delegates to fcntl_nocancel(). The documentation for
fcntl_nocancel() mentions error codes for some operations, but does not
include F_FULLFSYNC. The F_FULLSYNC branch in fcntl_nocancel() calls
VNOP_IOCTL(_, F_FULLSYNC, NULL, 0, _), whose return value sets the error
code.

VNOP_IOCTL() is implemented in bsd/vfs/kpi_vfs.c and calls the ioctl
function in the vnode's operation vector. The filesystem-level function
names follow the pattern _vnop_ioctl() for all the instances in
opensource code -- {hfs,msdosfs,nfs,ntfs,smbfs,webdav,zfs}_vnop_ioctl().

hfs-407.30.1, msdosfs-229.200.3, and nfs in xnu-4903.221.2 handle
F_FULLFSYNC. ntfs-94.200.1 and smb-759.40.1 do not handle F_FULLFSYNC,
and the default branch returns ENOSUP. webdav-380.200.1 also does not
handle F_FULLFSYNC, but the default branch returns EINVAL. zfs-59 also
does not handle F_FULLSYNC, and its default branch returns ENOTTY.

From a different angle, Apple's ntfs-94.200.1 includes utility code that
uses fcntl(F_FULLFSYNC) and falls back to fsync() just like this patch
does, supporting the hypothesis that there is no good way to detect lack
of F_FULLFSYNC support.

PR-URL: https://github.com/libuv/libuv/pull/2135
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Joran Dirk Greef <joran@ronomon.com>

show more ...

test: fix test-ipc tests

A couple of ipc tests failures where undercover by the previous commit.

- In `ipc-heavy-traffic-deadlock-bug`, `uv_shutdown` can fail with
`UV_ENOTCON

test: fix test-ipc tests

A couple of ipc tests failures where undercover by the previous commit.

- In `ipc-heavy-traffic-deadlock-bug`, `uv_shutdown` can fail with
`UV_ENOTCONN`. Allow it.
- `ipc_closed_handle` was failing consistently on `Centos6` because the
first large write was completed in just one syscall. Also, there were
issues with the lifetime of the `uv_write_t` requests. Refactor the
test so it passes consistently on `Centos6` while fixing the
`uv_write_t` issues.

PR-URL: https://github.com/libuv/libuv/pull/2108
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>

show more ...