test: fix the flaky test-tcp-reuseport

Start connecting to the peers after all threads
to poll for accepting connections.

Ref: #4407

doc: add entries for extended getpw

This patch adds documentation for the introduced `uv_os_get_passwd2`,
`uv_os_get_group`, `uv_os_free_group` methods in
https://github.com/libuv/li

doc: add entries for extended getpw

This patch adds documentation for the introduced `uv_os_get_passwd2`,
`uv_os_get_group`, `uv_os_free_group` methods in
https://github.com/libuv/libuv/pull/3523

Fixes: https://github.com/libuv/libuv/issues/4007
Signed-off-by: Juan José Arboleda <soyjuanarbol@gmail.com>

show more ...

unix: support SO_REUSEPORT with load balancing for TCP

bsd: support pipe2() on *BSD (#4412)

---------

Signed-off-by: Andy Pan <i@andypan.me>

unix: use the presence of SOCK_* instead of OS macros for socketpair

---------

Signed-off-by: Andy Pan <i@andypan.me>

test: remove the obsolete HAVE_KQUEUE macro

---------

Signed-off-by: Andy Pan <i@andypan.me>

dragonfly: disable SO_REUSEPORT for UDP socket bindings (#4410)

---------

Signed-off-by: Andy Pan <i@andypan.me>

linux,darwin: make `uv_fs_copyfile` behaves like `cp -r` (#4396)

This commit changes the timestamps in the file, the ownership and the
group.

Fixes: https://github.com/libuv/lib

linux,darwin: make `uv_fs_copyfile` behaves like `cp -r` (#4396)

This commit changes the timestamps in the file, the ownership and the
group.

Fixes: https://github.com/libuv/libuv/issues/3125

Signed-off-by: Juan José Arboleda <soyjuanarbol@gmail.com>

show more ...

doc: fix some typos

win: remove no longer needed define

Windows 7 is no longer supported.

mailmap: update saghul's main email address

darwin: simplify uv_hrtime

mach_continuous_time is available since macOS 10.12, but our minimum
version is 11, so no need for a workaround.

Also, prefer that to `clock_gettime_n

darwin: simplify uv_hrtime

mach_continuous_time is available since macOS 10.12, but our minimum
version is 11, so no need for a workaround.

Also, prefer that to `clock_gettime_nsec_np(CLOCK_MONOTONIC_RAW)` which
the documentation suggests
(https://developer.apple.com/documentation/driverkit/3438077-mach_continuous_time)
since the latter calls mach_timebase_info every time, unnecessarify: https://github.com/apple-open-source/macos/blob/49dcc07a40d19fa97384033a8398dae5d00d11a1/Libc/gen/clock_gettime.c#L107

show more ...

unix: de-duplicate conditions for using kqueue (#4378)

Our platform-specific headers provide a dedicated indicator.

test: use newer ASSERT_MEM_EQ macro (#4346)

Should hopefully make it easier to debug CI flakiness because
currently the test sometimes fails without a clear indication why.

Refs

test: use newer ASSERT_MEM_EQ macro (#4346)

Should hopefully make it easier to debug CI flakiness because
currently the test sometimes fails without a clear indication why.

Refs: https://github.com/libuv/libuv/issues/4106

show more ...

win: robustify uv_os_getenv() error checking (#4339)

Make it less likely for the thread-local error value to get
clobbered between performing the operation and checking the result.

win: robustify uv_os_getenv() error checking (#4339)

Make it less likely for the thread-local error value to get
clobbered between performing the operation and checking the result.

Refs: https://github.com/libuv/libuv/issues/4338

show more ...

unix: fix fd leaks in SCM_RIGHTS error path (#4358)

The file descriptor leak in the inner path was pointed out by @theanarkh
and I subsequently spotted another one in the outer loop. Rew

unix: fix fd leaks in SCM_RIGHTS error path (#4358)

The file descriptor leak in the inner path was pointed out by @theanarkh
and I subsequently spotted another one in the outer loop. Rewrite the
function to process all control messages.

Refs: https://github.com/libuv/libuv/pull/4357

show more ...

linux: don't use sendmmsg() for single datagrams (#4366)

Benchmarking shows that sendmsg() is persistently around 1% faster for
single datagrams, and that kind of stands to reason becaus

linux: don't use sendmmsg() for single datagrams (#4366)

Benchmarking shows that sendmsg() is persistently around 1% faster for
single datagrams, and that kind of stands to reason because there is
less setup overhead, and the kernel has to copy in less data.

Fixes: https://github.com/libuv/libuv/issues/4320

show more ...

doc: correct udp socket options documentation (#4371)

uv_udp_init() creates the UDP socket lazily but to set socket options
there must be, well, a socket to set the options on. Document

doc: correct udp socket options documentation (#4371)

uv_udp_init() creates the UDP socket lazily but to set socket options
there must be, well, a socket to set the options on. Document how and
when that requirement is met.

Fixes: https://github.com/libuv/libuv/issues/4370

show more ...

win: simplify uv_once implementation

* win: simplify uv_once implementation

InitOnceExecuteOnce is available in Windows >= Vista.

Ref: https://learn.microsoft.com/en-us/win

win: simplify uv_once implementation

* win: simplify uv_once implementation

InitOnceExecuteOnce is available in Windows >= Vista.

Ref: https://learn.microsoft.com/en-us/windows/win32/api/synchapi/nf-synchapi-initonceexecuteonce

show more ...

unix,win: error on zero delay tcp keepalive

Closes: https://github.com/libuv/libuv/pull/4350
Closes: https://github.com/libuv/libuv/issues/3487

doc: fix typos in ChangeLog (#4355)

Signed-off-by: tgolang <seekseat@aliyun.com>

linux: don't delay EPOLL_CTL_DEL operations (#4328)

Perform EPOLL_CTL_DEL immediately instead of going through
io_uring's submit queue, otherwise the file descriptor may
be closed by

linux: don't delay EPOLL_CTL_DEL operations (#4328)

Perform EPOLL_CTL_DEL immediately instead of going through
io_uring's submit queue, otherwise the file descriptor may
be closed by the time the kernel starts the operation.

Fixes: https://github.com/libuv/libuv/issues/4323

show more ...

unix: reinstate preadv/pwritev fallback code (#4345)

I removed the fallback code back in October but it prevents Node.js
from upgrading libuv in their v20.x release line because they sup

unix: reinstate preadv/pwritev fallback code (#4345)

I removed the fallback code back in October but it prevents Node.js
from upgrading libuv in their v20.x release line because they support
systems older than we do. Bring back a dlsym-based fallback path.

Fixes: https://github.com/libuv/libuv/issues/4332

show more ...

win: remove _alloca usage (#4361)

Remove it since it can cause stack overflows. Use heap allocation
instead.

Fixes: https://github.com/libuv/libuv/issues/4348

test,ci: fix [AM]San, disable ASLR (#4365)

The kernel that ships with the new Ubuntu 22.04 CI image seems to have a
PIE slide that is bigger than the sanitizer runtimes can handle.

test,ci: fix [AM]San, disable ASLR (#4365)

The kernel that ships with the new Ubuntu 22.04 CI image seems to have a
PIE slide that is bigger than the sanitizer runtimes can handle.

It makes ASan fail with thousands of "AddressSanitizer:DEADLYSIGNAL"
warnings, and MSan error with complaints about memory accesses outside
known ranges. Disabling address space layout randomization fixes both.

This commit also fixes a small bug in the platform_output test where
the cgroups v1 logic did not handle the "unlimited quota" special case
properly. Ubuntu 20.04 still uses cgroups v1.

show more ...