atomic_cas_ptr() is for compare-and-swap pointer addresses.
So when building 64 bit:
1. `ptr` is used casted to a pointer value (4-byte into 8-byte).
2. atomic_cas_ptr reads 8-byte at `ptr` and returns
In the case of `uv_async_send`, if handle->pending is 0, cmpxchgi()
actually returns the value of the 4-bytes past handle->pending,
causing uv__async_send to never be called.
The modified test-async-null-cb.c hangs at uv_run in this case.
PR-URL: https://github.com/libuv/libuv/pull/1361
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Filter out anything that is not an IPv4 or IPv6 interface in
uv_interface_addresses().
PR-URL: https://github.com/libuv/libuv/pull/1387
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Some of these compiler flags are not supported. So don't
use them.
PR-URL: https://github.com/libuv/libuv/pull/1373
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
The uv__get_osfhandle() function is a private functio of the
Windows subsystem, and its used to get a Windows HANDLE out
of a file descriptor number.
The motivation behind making this function public is to
allow Node.js programs to pass file descriptors created
using fs.open() to native Node.js C++ add-ons, and be able to
successfully convert them to Windows HANDLEs.
Refs: https://github.com/libuv/libuv/pull/1166
Refs: https://github.com/nodejs/node/issues/6369
Fixes: https://github.com/libuv/libuv/issues/1291
PR-URL: https://github.com/libuv/libuv/pull/1323
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
On Cygwin `recvmsg` always sets `msg_controllen` to zero on a
message received from a named pipe. Therefore we cannot use
`sendmsg` to send handles for ipc. Return failure early from
this code path as `ENOSYS`.
Skip tests requiring this feature since it is not available.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On Cygwin a `connect` to a file that is not a socket fails
with `EBADF` instead of `ENOTSOCK` even though the former is
supposed to be for a bad socket fd. Translate the error.
This fixes the `pipe_connect_to_file` to file test.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On Cygwin a named pipe closed by terminating a child process
may be reported via `ECONNRESET` on the next read. Teach
`uv__read` to interpret this as end of file.
This fixes the `spawn_and_kill_with_std` test.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Cygwin and MSYS2 are POSIX layers implemented on top of Windows.
Use our POSIX `poll(2)` implementation of our poll abstraction.
For most other components we already have dedicated sources
implementing them in terms of APIs available on Cygwin or
providing non-implementations of components not supported.
This leaves only three components that need Cygwin-specific
implementations:
* uv_uptime: implement using sysinfo
* uv_resident_set_memory: add a placeholder returning UV_ENOSYS
* uv_cpu_info: add a placeholder returning UV_ENOSYS
Update our test suite to account for features not available
due to Cygwin platform limitations or our placeholders.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
POSIX defines `poll()` as a portable way to wait on file descriptors.
Use it to provide an implementation of our poll abstraction for use
on platforms that do not provide a more specialized polling method.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On Linux we read `/proc/self/exe` for the path to the executable.
Factor out a dedicated source file to make this implementation
available for use on other platforms that support it.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On Linux we use `sysinfo(2)` to look up the load average.
Factor out a dedicated source file to make this implementation
available for use on other platforms that support it.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On Linux we use `sysinfo(2)` to look up available memory.
Factor out a dedicated source file to make this implementation
available for use on other platforms that support it.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On os390 we implement no support for fsevents. Other platforms may not
support fsevents either, so provide a dedicated source file to use in
this case.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
On SunOS we implement no support for proctitle. Other platforms may not
support proctitle either, so provide a dedicated source file to use in
this case.
PR-URL: https://github.com/libuv/libuv/pull/1312
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
It was previously assumed that OOB data had to be
inlined but that is incorrect.
PR-URL: https://github.com/libuv/libuv/pull/1347
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Remove the kvm.h include since 38323c9fb replaced the use of kvm_open()
with sysctl(),
Conditionally check for kvm_open() on non-OpenBSD platforms so LIBS (and
libuv.pc) does not unnecessarily contain "-lkvm".
Fixes: https://github.com/libuv/libuv/issues/1340
PR-URL: https://github.com/libuv/libuv/pull/1344
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Basic implementation of uv__io_fork on z/OS.
As of now, since filesystem events are not supported,
skip all of those tests on z/OS.
PR-URL: https://github.com/libuv/libuv/pull/1303
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
`int epoll_init()` declares a function that takes any number of
arguments, use `int epoll_init(void)` instead.
PR-URL: https://github.com/libuv/libuv/pull/1314
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: John Barboza <jbarboza@ca.ibm.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
"Fast" path in `uv_udp_send()` attempts to write the data straight away
without queueing the fd with epoll/kqueue/... However, in some cases
this is not possible when `sendmsg()` returns `EAGAIN`. In such event
libuv has to queue the fd anyway.
PR-URL: https://github.com/libuv/libuv/pull/1308
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Fixes deadstore in uv_pipe_bind as 'sockfd' is no longer used.
Fixes deadstore in test related to unused variables or missing
assert.
PR-URL: https://github.com/libuv/libuv/pull/1288
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
This goto target is only one line, and is only reached from
one place.
PR-URL: https://github.com/libuv/libuv/pull/1289
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Fixes a TODO in src/threadpool.c. Updates the Windows code to drop the
unused `loop` parameter in calls to uv_req_init().
PR-URL: https://github.com/libuv/libuv/pull/1091
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Added the uv_loop_fork() API that must be called in a child process to
continue using an existing loop. Internally this calls a uv__io_fork
function for each supported platform, similar to the way
uv__platform_loop_init works.
After this call, existing and new IO, async and signal watchers will
contiue working as before on all platforms, as will the
threadpool (although any threads it was using are of course gone).
On Linux and BSDs that use kqueue, existing and new fsevent watchers
will also continue to work as expected. On OS X, though, directory
fsevents will not be able to use the optimized CoreFoundation path if
they had already been used in the parent process, instead falling back
to the kqueue path used on other BSDs.
Existing fsevent watchers will not function on AIX or SunOS. This
could be relatively easily fixed by someone with AIX knowledge in the
future, but SunOS will require some additional work to keep track if
the watchers.
A new test file, test/test-fork.c, was added to contain fork-related
tests to verify functionality in the child process.
PR-URL: https://github.com/libuv/libuv/pull/846
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Remove the unused `nevents` parameter from `uv__async_event()`
and remove the indirection of having a separate `uv__async`
type. There is only one instance per event loop these days.
This incidentally removes the `assert(n == sizeof(val))` in a
Linux-specific code path that some users seem to hit from time
to time. The cause is not well-understood and I've never been
able to reproduce it myself. Presumably libuv gets an EAGAIN
when trying to read from the eventfd but when and why that
happens is unclear.
Since the byte count is unused, removing the assert seems safe.
Worst case, libuv sometimes iterates over the async watcher list
when it doesn't have to.
Fixes: https://github.com/libuv/libuv/issues/1171
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Supersedes: https://github.com/libuv/libuv/pull/1214
With strict-aliasing rules, a char* type can be cast to any other
pointer type. However, that does not necessarily mean char array
type. So instead of de-referencing a char array, use a union with an
additional char member for aliasing.
PR-URL: https://github.com/libuv/libuv/pull/1211
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Create a dedicated source file to share among platforms on which we find
ifaddrs using the BSD getifaddrs API. De-duplicate our existing copies
of this implementation on such platforms.
PR-URL: https://github.com/libuv/libuv/pull/1240
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On platforms using getifaddrs we iterate over the interfaces two or
three times. First we count them, then we enumerate them, and then we
fill in physical addresses. Each loop needs to do the same filtering,
so factor out the exclusion test into a helper function for each
platform.
PR-URL: https://github.com/libuv/libuv/pull/1240
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
We loop over getifaddrs results multiple times. Fix the filter
conditions to be consistent for all passes.
PR-URL: https://github.com/libuv/libuv/pull/1240
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
"setenv" on z/OS returns successfully on a NULL ptr argument
for name. Check for this on all unixes.
PR-URL: https://github.com/libuv/libuv/pull/1243
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Create a dedicated source file to share among platforms on which we
implement hrtime using POSIX `clock_gettime`. De-duplicate our
existing copies of this implementation on such platforms.
PR-URL: https://github.com/libuv/libuv/pull/1239
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
It behaves as `uv_signal_start` but it resets the signal handler as soon
as the signal is received.
Fixes: https://github.com/libuv/libuv/issues/1104
PR-URL: https://github.com/libuv/libuv/pull/1106
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Bartosz Sosnowski <bartosz@janeasystems.com>
The code path in `uv__io_start` cannot be reached because `pevents` is
always non-zero at that point.
PR-URL: https://github.com/libuv/libuv/pull/1227
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On AIX, watch feature depends on AHAFS based Event infrastructure.
While in principle the watch use case is same across platforms, there
are subtle differences in the way AIX deals with this, with few
behavioral changes (external).
This commit opens up the AIX code for watch feature which was masked
under a macro HAVE_SYS_AHAFS_EVPRODS_H and addresses an assertion
failure on folder watch.
PR-URL: https://github.com/libuv/libuv/pull/1156
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Use built-in proctitle implementation because there is no such thing as
a process title on zOS.
PR-URL: https://github.com/libuv/libuv/pull/1208
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This patch enables querying and setting the process title on AIX.
libuv takes ownership of the memory for argv and returns a deep copy
of the array and its contents.
It also enables the process_title test case.
The process title can be changed on AIX but is handled differently
to Linux/Mac. Commands like ps read the original argv array passed
to the process instead of the memory at argv[0]. To change the process
title we need to update argv[0] to point at a new string and set
argv[1] to NULL (if it exists). This means the process title
can be set to a string of any length.
PR-URL: https://github.com/libuv/libuv/pull/1187
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Fix all functional test cases:
* use PLO compare,swap,store for atomic instruction
* do not use semaphore.h
* use xplink flag when linking
* scandir implementation
* nanosleep implementation
* add proctitle
* uv_loadavg
* uv_fs_event_init/start
* uv_fs_event_stop
* uv_exepath using __getthent syscall
* read free/total memory from mvs data areas
* uv_resident_set_memory implementation
* network interfaces implementation
* cpu_info implementation
* implement uv__hrtime
* make uv__fs_mkdtemp implementation
* epoll implementation for asyncio
* uv__fs_event_close implementation
* set process title
* read ancillary data that remains on queue
* ancillary data
* implement uv__fs_access
* use /dev/urandom for temporary directory name
* disable proctitle on zos completely
PR-URL: https://github.com/libuv/libuv/pull/1037
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Don't compile uv__set_phys_addr() if SUNOS_NO_IFADDRS is defined,
it has been reported to break the build on Solaris 10.
Bug introduced in commit d75e334 ("sunos: set phys_addr of
interface_address using ARP") from last June.
Fixes: https://github.com/nodejs/node/issues/9856
PR-URL: https://github.com/libuv/libuv/pull/1155
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The `out` section is only reachable now if `scandir` returns 0.
PR-URL: https://github.com/libuv/libuv/pull/1189
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
assert expressions should only be scalar
PR-URL: https://github.com/libuv/libuv/pull/1196
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Fix the #ifdef logic to make sure all the available fields filled in the
`stat` struct are correcly copied into the `uv_stat_t` struct. Before
this change, the `n_sec` were being set to 0.
PR-URL: https://github.com/libuv/libuv/pull/1152
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Apple's fsync and fdatasync explicitly do NOT flush the drive write
cache to the drive platters. This is in contrast to Linux's fsync and
fdatasync which do, according to recent man pages. F_FULLFSYNC is
Apple's equivalent for flushing buffered data to permanent storage.
PR-URL: https://github.com/libuv/libuv/pull/1124
Refs: https://github.com/nodejs/node/issues/9439
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
So it aborts in case `pthread_once` fails.
PR-URL: https://github.com/libuv/libuv/pull/1098
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv_translate_sys_error() was a private function for Windows.
This commit adds an equivalent function on other platforms, and
exposes it as public API.
Exposing this is useful in scenarios where the application uses
both libuv functions and platform-specific system calls and wants
to report errors uniformly as libuv errors.
Fixes: https://github.com/libuv/libuv/issues/79
PR-URL: https://github.com/libuv/libuv/pull/1060
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
In unix/internal.h, CoreServices is included for AvailabilityMacros.h.
This commit just includes AvailabilityMacros.h directly instead.
PR-URL: https://github.com/libuv/libuv/pull/1092
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Simplify and optimize uv_thread_create() by casting the function pointer
to the prototype that pthread_create() wants. Avoids the indirection of
an intermediate callback and heap-allocating custom state that is really
only there to placate the compiler.
PR-URL: https://github.com/libuv/libuv/pull/1094
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
alphasort() uses strcoll() in some libcs (notably glibc and musl) which
makes it locale-sensitive.
Replace it with a simple strcmp-based lexicographic comparison for
consistent behavior across platforms.
Remove the special-casing for OpenBSD. It has been exporting function
prototypes conforming to POSIX.1-2008 since OpenBSD 5.3 and that was
released more than three years ago.
OS X 10.7 is now the only special case left.
Fixes: https://github.com/libuv/libuv/issues/1045
PR-URL: https://github.com/libuv/libuv/pull/1046
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
In `musl` libc, defining `_GNU_SOURCE` doesn't automatically define the
other feature definitions, causing that `uv_fs_fstat` would not fill the
`nsec` fields. For the same reason, compile the tests on linux with
`-D_GNU_SOURCE` so the `fs_fstat` tests passes on `musl` platforms.
PR-URL: https://github.com/libuv/libuv/pull/1039
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The type parameter in the "container_of(ptr, type, member)" macro
which uses builtin "offsetof(type, member)" should not require cv
qualifier. Also note that for some platforms, the "offsetof" builtin
does not recognize the cv qualifier in the type.
PR-URL: https://github.com/libuv/libuv/pull/948
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This method uses the poll syscall to determine whether POLLNVAL is
flagged or not.
PR-URL: https://github.com/libuv/libuv/pull/957
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On zOS the listen call does not bind automatically if the socket
is unbound. Hence the manual binding to an arbitrary port is
required to be done manually.
PR-URL: https://github.com/libuv/libuv/pull/949
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
We do this using __fchattr syscall to change file attributes.
PR-URL: https://github.com/libuv/libuv/pull/956
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Use builtins provided that perform compare and swap operations
using the PLO instruction.
PR-URL: https://github.com/libuv/libuv/pull/1008
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
autotools support only, gvp does not support kfreebsd detection.
PR-URL: https://github.com/libuv/libuv/pull/960
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
- zos: disable test cases not applicable
- zos: build options
- zos: semaphore implementation
- zos: use compare and swap builtins
- zos: struct rusage not the same as other platforms
- zos: backlog<=0 produces undefined behaviour
Will redefine backlog in the following way
* if backlog == 0, set it to 1
* if backlog < 0, set it to SOMAXCONN
- zos: define IMAXBEL as empty flag and implement uv__tty_make_raw
- zos: use udp multicast operations from aix
- zos: ESC in ebcdic
- zos: use LIBPATH for dynamic linker path
- zos: uv_udp_set_ttl only works for ipv6
- zos: increase pthread stack size by factor of 4
- zos: return ENODEV instead of ENXIO errors for setsockopt
- zos: use uv_cond_init the same way as aix
- test: enable oob test for zos
- zos: return EINVAL for zos error code EOPNOTSUPP
PR-URL: https://github.com/libuv/libuv/pull/937
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Treat both the base being NULL or the length being 0 as ENOBUFS.
PR-URL: https://github.com/libuv/libuv/pull/997
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On OS X, special files like /dev/null and /dev/tty don't work with
kqueue. Libuv falls back to select() in that case but the initial
probe didn't handle EINTR.
Introduced in August 2012 in commit 731adaca ("unix: use select()
for specific fds on OS X"), this bug was only ten days away from
celebrating its fourth birthday.
PR-URL: https://github.com/libuv/libuv/pull/979
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Libuv leaves loop->data unchanged in uv_loop_init() and uv_loop_done()
on Windows but it clobbered it on UNIX platforms. This commit fixes
that inconsistency.
PR-URL: https://github.com/libuv/libuv/pull/951
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
We not only check the return value, but also check the errno != 0.
Because in rare cases connect() will return -1 but the errno
is 0 (for example, on Android 4.3, OnePlus phone A0001_12_150227)
and actually the tcp three-way handshake is completed.
PR-URL: https://github.com/libuv/libuv/pull/936
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This commit causes uv_get_process_title() to:
- return EINVAL if the buffer is null or size is 0
- return ENOBUFS if the title is too big for the buffer
- null terminate the buffer on success
Fixes: https://github.com/libuv/libuv/issues/315
PR-URL: https://github.com/libuv/libuv/pull/928
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
It was pointed out[0] that libuv could effectively enter an infinite
loop (but not a busy loop) under certain conditions when polling for
events:
1. When the architecture is 32 bits, and
2. When timeout > 0, i.e., finite, and
3. When timeout > max_safe_timeout (~30 minutes), and
4. When epoll_wait(timeout) returns 0, then
5. timeout was not properly updated on the next call to epoll_wait().
Inspection of the code uncovered a secondary bug where under a similar
set of circumstances the timeout could drift when the epoll_wait()
system call returned late.
[0] https://github.com/libuv/libuv/pull/354#discussion_r67837112
PR-URL: https://github.com/libuv/libuv/pull/922
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Don't call uv__to_stat() when the stat/fstat/lstat system call fails;
the stack-allocated buffer contains only garbage in that case.
Not a very serious bug it's technically undefined behavior and it made
valgrind squawk.
Introduced in commit 499c7976 ("unix, windows: nanosecond resolution
for uv_fs_[fl]stat").
PR-URL: https://github.com/libuv/libuv/pull/921
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
If getifaddrs() returns an all zero ifa_addr field, then use ARP to
set the phys_addr field of the uv_interface_address_t on the sunos
platform.
PR-URL: https://github.com/libuv/libuv/pull/907
Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
`unsigned long` is not guaranteed to be 8 bytes long. This prevents
overflow errors from occurring as seen in issue #913
Fixes: https://github.com/libuv/libuv/issues/913
PR-URL: https://github.com/libuv/libuv/pull/914
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This fixes test-stdio-over-pipes and test-pipe-sendmsg test cases.
~~~~
Assertion failed in test/test-stdio-over-pipes.c on line 56: term_signal == 0
Assertion failed in test/test-pipe-sendmsg.c on line 86: pending == UV_NAMED_PIPE
~~~~
This fixes a longstanding issue with nodejs on DragonFly, which was
triggered whenever spawning a process and pipes were involed.
PR-URL: https://github.com/libuv/libuv/pull/884
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Imran Iqbal <imran@imraniqbal.org>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The _SC_PHYS_PAGES and _SC_AVPHYS_PAGES are not POSIX sysconf values, so
the standart C libraries have no obligation to support it, even on
Linux. Use the Linux sysinfo() system call instead.
PR-URL: https://github.com/libuv/libuv/pull/901
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
It's just an indicator that the operation will happen later, so return
success.
PR-URL: https://github.com/libuv/libuv/pull/892
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On unix, scandir() uses the system allocator to allocate memory.
This commit releases the memory with free() instead of uv__free().
uv__free() is still used on Windows, which uses uv__malloc() to
request the memory.
Fixes: https://github.com/libuv/libuv/issues/873
PR-URL: https://github.com/libuv/libuv/pull/874
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
pthread_attr_setstacksize() expects that the stack size is a multiple of
the page size so make sure that it is.
Fixes a regression introduced in commit 3db07cc ("osx: set the default
thread stack size to RLIMIT_STACK") that made the program abort under
certain configurations; GNU Emacs on OS X apparently sets RLIMIT_STACK
to odd values when executing child processes.
Fixes: https://github.com/nodejs/node/issues/6563
PR-URL: https://github.com/libuv/libuv/pull/864
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
There was a memory corruption issue with the pthread barrier
implementation on android, where a barrier could still be in use by one
thread when being freed by another.
This fixes that issue and adds lots of missing error handling.
This implementation is now also used for the OSX fallback.
Fixes: https://github.com/libuv/libuv/issues/615
PR-URL: https://github.com/libuv/libuv/pull/790
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Currently when PATH_MAX is undefined realpath will default to using 4096.
There is a potential stack overflow attack that can be mitigated by having
PATH_MAX defined. This change conservatively errors if a system does not
have PATH_MAX defined.
This change also explicitly includes `limits.h` to ensure that all platforms
have PATH_MAX defined if it is available.
Ref: http://pubs.opengroup.org/onlinepubs/9699919799/functions/realpath.html
Refs: https://github.com/nodejs/node/issues/2680#issuecomment-213521708
PR-URL: https://github.com/libuv/libuv/pull/843
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv__close() won't clobber errno so there is no need to guard calls with
a SAVE_ERRNO(...) wrapper.
PR-URL: https://github.com/libuv/libuv/pull/837
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Libuv expects that free() does not clobber errno. The system allocator
honors that assumption but custom allocators may not be so careful.
PR-URL: https://github.com/libuv/libuv/pull/837
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Originally intended workaround is especially needed for Android <4.4.
However it fails to compare errno collected from pthread_sigmask.
This has been fixed separately in JXcore. See issue:
https://github.com/jxcore/jxcore-cordova/issues/55
PR-URL: https://github.com/libuv/libuv/pull/833
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
It was reported that some node.js tests fail on AIX because the exit
event sometimes comes before the final stdio output of a child process.
Work around that by deferring the signal watcher that is used for
process management until after the dispatch of regular i/o watchers.
Fixes: https://github.com/libuv/libuv/issues/610
PR-URL: https://github.com/libuv/libuv/pull/611
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Remove unused variables and code. Also remove some code duplication. End
result is code is cleaner, readable and easier to step through when
debugging.
PR-URL: https://github.com/libuv/libuv/pull/825
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Return an error instead of aborting when /proc/stat doesn't have the
expected format.
PR-URL: https://github.com/libuv/libuv/pull/822
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
'futimens' is only implemented on AIX 7.1. Other functions like
'utimes' and 'utimes' are merely stub functions that return ENOSYS
on AIX 6.1 and below. Skip test fs_futime for AIX versions below 7.1.
PR-URL: https://github.com/libuv/libuv/pull/811
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
The comment removed by this commit is no longer relevant, as
the function's return type changed from int to void.
PR-URL: https://github.com/libuv/libuv/pull/804
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Find the real name of the tty using ttyname_r(3) instead of
opening "/dev/tty" which causes trouble if the fd doesn't point to the
controlling terminal.
PR-URL: https://github.com/libuv/libuv/pull/779
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Use `uv__close__nocheckstdio` instead of `close`.
PR-URL: https://github.com/libuv/libuv/pull/798
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
We only need the absolute path to create our .mon file under /aha. So
only duplicate the passed string. This keeps behaviour similar to other
platforms.
Fixes test fs_event_getpath.
PR-URL: https://github.com/libuv/libuv/pull/776
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv__makedir_p was not processing all directories in passed string. Now
if a directory already exists (EEXIST) it simply moves onto the next
directory in the provided string.
Fixed bogus assert in uv__ahafs_event.
PR-URL: https://github.com/libuv/libuv/pull/776
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
It can happen that the `parent` end of the `signal_pipe` is given a
STDIO file descriptor, so closing it using `uv__close` fails.
This problem is happening when running the `spawn_closed_process_io`
test in `SmartOS`. The reason being that when creating a socketpair in
`uv__process_init_stdio`, the `Illumos` implementation uses 3 sockets:
one is used as a listener, and the other 2 represent both ends of the
pipe. The listener socket is closed once the pipe is created. In the
test, the listener socket is assigned to the `0` fd, as it is the
first free fd in the system. So the fd `0` remained free after the call
to `socketpair`. Afterwards, when creating the `signal_pipe`, the fd `0`
is being assigned again, so closing it with `uv__close` made the test
fail. This issue is not happening in the other unixes because
`socketpair` doesn't use 3 fd's, but only 2.
To solve the issue, a new `uv__close__nocheckstdio()` function has been
added and used.
PR-URL: https://github.com/libuv/libuv/pull/796
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv__open_cloexec returns either the fd or a libuv error, which is -errno
on Unix, not -1.
PR-URL: https://github.com/libuv/libuv/pull/760
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
When compiling with bionic, st_atime_nsec (and the other nsec members)
canot be found.
In order to compile uv with both NDK and bionic, tv_nsec should be
accessed from st_atimensec (and the others nsec members).
See "Compatibility with glibc" in stat.h from bionic.
Signed-off-by: Robert Chiras <robert.chiras@intel.com>
PR-URL: https://github.com/libuv/libuv/pull/746
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Commit 6798876 ("linux: fix cpu count") switched the /proc/stat parser
to fdopen(). Use fclose() to close the file descriptor to avoid leaking
resources associated with the FILE struct.
PR-URL: https://github.com/libuv/libuv/pull/747
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
'getsockname' does not handle UNIX domain sockets on AIX[1], it does not
fail but simply returns an empty structure with length 0.
If 'getsockname' does not fail (i.e. does not return -1) and the length
is 0 we know, on AIX, that the type is AF_UNIX.
This fixes test pipe_sendmsg.
[1] https://goo.gl/ozqcmS
PR-URL: https://github.com/libuv/libuv/pull/748
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
When libuv is running inside container - eg. lxc container, cpu number
is not obvious. Linux control groups (cgroups) may limit numer of cpus.
As a result of different number cpu cores inside container and
outside container, libuv is crashing.
sysconf(_SC_NPROCESSORS_ONLN) - returns num of host cpus (eg. 32)
`/proc/stat` - sees only cpus limited by cgroups (eg. 4)
When libuv is trying to operate at both numbers and they're different
it's crashing with current test:
run-tests: ../src/unix/linux-core.c:766: read_times: Assertion `num ==
numcpus' failed.
Count the number of cpus based on `/proc/stat` instead.
PR-URL: https://github.com/libuv/libuv/pull/735
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
If an API function returns UV_ENOBUFS make sure we count the terminating
null, which we need space for. In case of success the null byte is not
included in the count, but the buffer *is* null terminated.
PR-URL: https://github.com/libuv/libuv/pull/690
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
libuv has multiple functions which return buffers. Make them consistent
with the following rules: the returned size *does not* include the null
byte, but the buffer *is* null terminated.
There is only one exception to the above: Linux abstract sockets,
because null bytes are not used as string terminators in those.
Refs: https://github.com/libuv/libuv/pull/674
PR-URL: https://github.com/libuv/libuv/pull/690
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
The `__FreeBSD__` macro is not defined by default on OSX, and building
with `-Wundef` will cause a warning to be emitted. This change makes
sure that the macro is only used if defined, removing the warning.
PR-URL: https://github.com/libuv/libuv/pull/710
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Libuv would abort() when trying to watch a file descriptor that is
not compatible with epoll-style polling; file descriptors referring
to on-disk files fall into this category.
File descriptors that libuv creates itself are not an issue but
external ones that come in through the uv_poll_init() API are.
Make uv_poll_init() check whether the file descriptor is accepted by
the underlying system call and return an error when it's not.
Fixes: https://github.com/libuv/libuv/issues/658
PR-URL: https://github.com/libuv/libuv/pull/659
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv__fs_buf_iter currently sets req->bufs to NULL after it is done, but
if the operation fails with EINTR then it will be retried, at which
point it expects the bufs to not be NULL, causing a seg fault as in
https://github.com/nodejs/node/issues/4291.
uv__fs_buf_iter should not set req->bufs to NULL if the operation
fails with EINTR.
Also, when it sets req->bufs to NULL, it should set req->nbufs to 0 as
well, so we don't have the messy situation of a positive nbufs with no
actual bufs.
PR-URL: https://github.com/libuv/libuv/pull/661
Reviewed-By: Fedor Indutny <fedor@indutny.com>
uClibc-ng is currently at v1.0.9. The patch corrects the uClibc
version test so that HAVE_IFADDRS_H is defined for uClibc versions
after v0.9.32.
Signed-off-by: Martin Bark <martin@barkynet.com>
PR-URL: https://github.com/libuv/libuv/pull/653
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The following changeset
442b8a5a84 "unix: use QUEUE_MOVE when iterating over lists"
introduced a new assert failure:
`queue_foreach_delete` failed: exit code 6
Output from process `queue_foreach_delete`:
run-tests: src/unix/linux-inotify.c:244: uv_fs_event_stop: Assertion `w != ((void *)0)' failed.
Simplest test case for this:
1. create and start two uv_fs_event_t for the same path;
2. in the callback for the first one, call uv_close() on it;
3. assert/segfault while accessing the second uv_fs_event_t from uv__inotify_read().
PR-URL: https://github.com/libuv/libuv/pull/621
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Equivalent to realpath(3), returns the full resolved absolute path of a
file or directory.
PR-URL: https://github.com/libuv/libuv/pull/531
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
When `uv_fsevents_t` handle is closed immediately after initializing,
there is a possibility that the `CFRunLoop`'s thread will process both
of these events at the same time. `uv__is_active(handle)` will return
`0`, and the `uv_close()` semaphore will be unblocked, leading
to the use after free in node.js.
See: https://github.com/nodejs/node/issues/4091
PR-URL: https://github.com/libuv/libuv/pull/637
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Remove restriction to Android API level >= 21, now libuv can be used with
Android API level at least 14.
Direct use of getpwuid_r function resulted in linker errors in applications
were built against API level below 21. That function was 'officially'
introduced in Android API level 21, but it exists in libc beginning from API
level at least 14.
So try to get a pointer to getpwuid_r at runtime using dlsym when building
libuv with API level < 21.
PR-URL: https://github.com/libuv/libuv/pull/633
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Like the previous commit, but this time for UNIX platforms other than
Linux.
As far as I have been able to establish, dup2 and dup3 never return
EINTR on OS X and FreeBSD. Even if other platforms do return EINTR,
it's probably still better not to retry because it's unspecified
whether the file descriptor has been closed.
Fixes: https://github.com/libuv/libuv/issues/462
PR-URL: https://github.com/libuv/libuv/pull/608
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Retrying seems like a bad idea in a multi-threaded program because the
man page leaves it unspecified whether the file descriptor is closed
after EINTR.
It's probably an academic issue because as far as I have been able to
establish by checking the kernel sources, dup2 and dup3 never actually
return EINTR. For dup3, this appears to have been the case since its
introduction in v2.6.27; for dup2, it goes back to at least v2.6.18.
Fixes: https://github.com/libuv/libuv/issues/462
PR-URL: https://github.com/libuv/libuv/pull/608
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The pthread_mutex_lock() and pthread_mutex_unlock() calls logically
cannot fail in uv__fs_write() but let's check the return value anyway:
cosmic rays and memory corruption do happen.
PR-URL: https://github.com/libuv/libuv/pull/603
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
fork and the exec functions are marked prohibited on tvOS and watchOS,
so referencing them causes compile errors.
This adds preprocessor conditionals to avoid calling them for those
targets.
PR-URL: https://github.com/libuv/libuv/pull/580
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
The atomics that are in place of a lack of x86 cmpxchg are a GCC
specific intrinsic. When compiling for Solaris on a SPARC platform
with a non-gcc compiler, this prevents a successful build. This commit
will rely on a Solaris Studio specific preprocessor macro to redefine
the GCC intrinsic name to be the Solaris Studio one instead.
PR-URL: https://github.com/libuv/libuv/pull/569
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On some embedded devices (arm-linux-uclibc based ip camera),
sysconf(_SC_IOV_MAX) can not get the correct value. The return
value is -1 and the errno is EINPROGRESS. Degrade the value to 1.
PR-URL: https://github.com/libuv/libuv/pull/573
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Commit 0199955 ("fs: undo uv__req_init when uv__malloc failed")
mistakingly unregisters the requests unconditionally in a few places,
resulting in memory corruption when it hasn't been registered first.
Fixes: https://github.com/libuv/libuv/pull/543
PR-URL: https://github.com/libuv/libuv/pull/567
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Valgrind complains that the msg_control pointer points to uninitialized
memory. While the memory is only used for writing data to, not for
reading data from, and the warning is therefore bogus, it's still
annoying enough that I decided to squelch it by zeroing the memory.
The performance implications should be minimal because this code path
is only used when sending over a handle to another process.
The warning:
==14859== Syscall param sendmsg(msg.msg_control) points to
uninitialised byte(s)
==14859== at 0x5AF1A80: __sendmsg_nocancel (in
/usr/lib64/libpthread-2.21.so)
==14859== by 0x46350E: uv__write (stream.c:810)
==14859== by 0x464B24: uv_write2 (stream.c:1398)
==14859== by 0x421ACE: run_test (test-ipc-send-recv.c:104)
==14859== by 0x421DD1: run_test_ipc_send_recv_tcp
(test-ipc-send-recv.c:156)
==14859== by 0x406D2F: run_test_part (runner.c:404)
==14859== by 0x4058CD: main (run-tests.c:58)
==14859== Address 0xffefff934 is on thread 1's stack
==14859== in frame #1, created by uv__write (stream.c:742)
PR-URL: https://github.com/libuv/libuv/pull/565
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Replace uses of QUEUE_FOREACH when the list can get modified while
iterating over it, in particular when a callback is made into the
user's code. This should fix a number of spurious failures that
people have been reporting.
PR-URL: https://github.com/libuv/libuv/pull/565
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
All uses of QUEUE_SPLIT in libuv split the list at the head so introduce
a QUEUE_MOVE macro that automates that.
PR-URL: https://github.com/libuv/libuv/pull/565
Reviewed-By: Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On AIX, the getsockname API does not operate on the UNIX domain sockets
Please see: https://www-01.ibm.com/support/knowledgecenter/ssw_aix_61/
com.ibm.aix.commtrf2/getsockname.htm
This means that the internal helper routine uv_guess_handle in tty.c
fails in AIX, for common use cases.
There is no direct API support which helps identifying a UNIX domain
socket in AIX. getpeername() retrieves the information pertinent to
connected UNIX domain socket pairs, and using getpeername where
getsockname fails was a good workaround. However, there are edge cases
where one end of a socketpair comes for introspection while the other
end has been closed, which causes getpeername to fail.
A better solution is derived based on these facts:
1.getsockname() on a broken / un-broken socketpair does not actually
fail with -1, instead it returns 0. However, the API does not modify
the socketaddr structure passed to it, with the socket information.
2.This behavior of getsockname is observed only for UNIX domain socket,
among all possible types of sockets which I tested - ~30 of them.
So a practical and stable workaround for AIX is to return
UV_UNKNOWN_HANDLE if getsockname fails. If it passes and the length
of the structure returned is 0 then we know that the type is AF_UNIX
and return UV_NAMED_PIPE , otherwise we inspect in the same way as for
other platforms to derive the type.
PR-URL: https://github.com/libuv/libuv/pull/539
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Fold EAGAIN into EBUSY, and make it the only acceptable error.
PR-URL: https://github.com/libuv/libuv/pull/535
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
On AIX with the native xlC compiler, the compiler supports sync
compare swap and compare in two flavors:
1 __compare_and_swap
2 __compare_and_swaplp
This differs from the gcc version of this function in name. The
second version of this function supports 8-byte boundary for
a doubleword. Therefore, the macro logic checks:
if (AIX OS AND using xlC compiler)
if (64bit)
__compare_and_swaplp /* 64 bit version */
else
__compare_and_swap
endif
endif
Signed-off-by: Neil Mushell <nmushell@bloomberg.net>
PR-URL: https://github.com/libuv/libuv/pull/521
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv__close() does not set errno, it returns the error. It also never
returns EINTR, it always maps it to EINPROGRESS.
PR-URL: https://github.com/libuv/libuv/pull/512
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The code made an implicit assumption that the 'len` variable passed
to the sendfile(2) syscall is not modified by the operating system
in case of an error other than EAGAIN or EINTR.
The man page leaves this unspecified on FreeBSD, DragonFly and
Darwin, so better check the error code which returns a valid
value in `len` explicitly (only EAGAIN and EINTR).
This fixes the test case for sendfile on DragonFly.
PR-URL: https://github.com/libuv/libuv/pull/466
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Makes uv_loop_init return an error code instead of aborting.
Currently uv_loop_init aborts if there are insufficient resources
available. As a user I want to be able to check the return code from
uv_loop_init and decide how I respond rather than having my process die.
PR-URL: https://github.com/libuv/libuv/pull/405
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
At least on OS X 10.10 "Yosemite", an EPROTOTYPE can occur
when trying to write to a socket that is shutting down.
By retrying the write after EPROTOTYPE, we correctly get EPIPE.
PR-URL: https://github.com/libuv/libuv/pull/482
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
For synchronous file operation requests (ones without a callback), there
is no need to make a copy of the arguments because they don't outlive
the scope of the function call.
PR-URL: https://github.com/libuv/libuv/pull/479
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The parentheses are unnecessary because what they wrap are not macro
arguments but function arguments that aren't evaluated by the macro.
PR-URL: https://github.com/libuv/libuv/pull/479
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This allows writing and reading any amount of buffers,
regardless of what IOV_MAX may be defined as.
It also moves the IOV_MAX test from stream to core.
This is based on the excellent work of @bwijen in #269.
Refs: https://github.com/libuv/libuv/pull/269
PR-URL: https://github.com/libuv/libuv/pull/448
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv_exepath() was not behaving correctly on AIX when the process was
invoked with a relative path.
The logic has been reworked to:
1) Use getargs() to retrieve the process arguments
2) Use realpath() to handle absolute and relative paths and
symbolic links
3) Check that a match when searching the PATH is executable (if
not search will continue)
PR-URL: https://github.com/libuv/libuv/pull/464
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Add new UV__POLLRDHUP event to be emitted when EPOLLRDHUP(in Linux) or
EV_EOF(in BSD / OSX) is detected and only if UV_READABLE is set.
When a read returns ECONNRESET after a UV__POLLRDHUP event, emit EOF instead
of the error.
Add tcp-squelch-connreset test. Not to be run on Windows as it returns
ECONNRESET error.
Fixes in test-poll and test-tcp-open so they pass after these changes.
PR-URL: https://github.com/libuv/libuv/pull/403
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Once we are past uv__stream_init, the handle has been added to
loop->handle_queue, it needs to be undone in case of failure.
PR-URL: https://github.com/libuv/libuv/pull/414
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
OSX, other BSDs and SunoS fail with EAFNOSUPPORT when binding a socket created
with AF_INET to an AF_INET6 address or vice versa.
PR-URL: https://github.com/libuv/libuv/pull/407
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Introduce two new APIs:
int uv_tcp_init_ex(uv_loop_t*, uv_tcp_t* handle, int flags)
int uv_udp_init_ex(uv_loop_t*, uv_udp_t* handle, int flags)
The lower 8 bits of the flags field are used for the socket domain.
AF_INET, AF_INET6 and AF_UNSPEC are supported. If AF_UNSPEC is specified
the socket is created lazily, just like uv_{tcp,udp}_init.
Some Windows notes:
getsockname fails with WSAEINVAL if the socket is not bound. This could
potentially be improved by detecting the socket family and filling
the sockaddr_in/6 struct manually.
bind returns WSAEFAULT if we try to bind a socket to the wrong family.
Unix returns EINVAL.
PR-URL: https://github.com/libuv/libuv/pull/400
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This fixes a bug I have noticed with nodejs:
> cat test.js
console.log(process.execPath);
> node test.js (on DragonFly)
node test.js
> node test.js (on Linux)
node
PR-URL: https://github.com/libuv/libuv/pull/399
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Use the same code path as FreeBSD does.
PR-URL: https://github.com/libuv/libuv/pull/399
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Some systems, FreeBSD for example, may return negative values.
sysconf() returns a long, which was being converted to a size_t.
This conversion lead to large allocations, and subsequent out of
memory failures. This commit checks the long value returned by
sysconf() properly, and uses a default value of 4096 if a negative
number is returned.
PR-URL: https://github.com/libuv/libuv/pull/389
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The libuv filesystem events API on AIX requires the non-default IBM
`bos.ahafs` package to be installed. The library should still compile
and work on systems without this package. This patch adds proper
detection to `configure.ac` and makes the API return `-ENOSYS` at
runtime if the system does not support it.
PR-URL: https://github.com/libuv/libuv/pull/372
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Old 32 bits kernels (native and compat) have a bug where timeouts larger
than (LONG_MAX / CONFIG_HZ) milliseconds are treated as infinite.
Work around that by capping the timeout and polling again if necessary.
PR-URL: https://github.com/libuv/libuv/pull/354
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On AIX the length of socket options
for multicast and ttl options is not always sizeof(char).
AIX has the same issue as solaris which was fixed under
PR-URL: https://github.com/libuv/libuv/pull/243
This PR extends the fix to cover AIX as well.
PR-URL: https://github.com/libuv/libuv/pull/345
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Make it consistent with other platforms.
PR-URL: https://github.com/libuv/libuv/pull/303
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Since libuv has it's own syscall wrappers, the standard libc interception
mechanism of the sanitizers do not catch these, leading to false positives
for projects using libuv when running under memory sanitizer.
Fix this by adding the appropriate calls to unpoison manually.
PR-URL: https://github.com/libuv/libuv/pull/311
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Move the logic out of uv__fs_work() and into a function of its own.
PR-URL: https://github.com/libuv/libuv/pull/326
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
We may not have permission to open "/". This fix allows libuv to be
used in a OS X app that has sandboxing enabled, without the need for
a temporary entitlement to allow reading of "/" (Which would never
pass app review).
In some rare cases "/dev" isn't mounted, so we open "/" as a
fallback.
PR-URL: https://github.com/libuv/libuv/pull/328
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
If a descriptor was being copied to an fd that *was not its current
value*, it should get closed after being copied. But the existing code
would close the fd, even when it no longer referred to the original
descriptor. Don't do this, only close fds that are not in the desired
range for the child process.
See https://github.com/iojs/io.js/issues/862
PR-URL: https://github.com/libuv/libuv/pull/309
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
arm64 doesn't have a epoll_wait() system call but a logic error stopped
libuv from falling back to epoll_pwait().
This bug was introduced in commit 67bb2b5 ("linux: fix epoll_pwait()
regression with < 2.6.19") which sadly exchanged one regression for
another.
PR-URL: https://github.com/libuv/libuv/pull/308
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv_tty_reset_mode() is designed to be async signal-safe and is therefore
not allowed to clobber errno.
PR-URL: https://github.com/libuv/libuv/pull/259
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On Solaris and derivatives such as SmartOS, the length of socket options
for multicast and ttl options is not always sizeof(char).
This fixes the udp_options and udp_options6 tests.
PR-URL: https://github.com/libuv/libuv/pull/243
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Apple's `fd_set` stores its bits in an array of 32-bit integers, which
means `FD_ISSET()` may read out of bounds if we allocate storage at
byte granularity. There's also a chance that the `select()` call could
corrupt the heap, although I didn't investigate that.
This issue was discovered by LLVM's AddressSanitizer which caught
`FD_ISSET()` trying to read out of bounds.
PR-URL: https://github.com/libuv/libuv/pull/241
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
With uv_replace_allocator, it's possible to override the default
memory allocator's malloc and free calls with functions of the user's
choosing. This allows libuv to interoperate with projects requiring a
custom memory allocator.
Internally, all calls to malloc and free have been replaced with
uv__malloc and uv__free, respectively. The uv__malloc and uv__free
functions call malloc and free unless they have been overridden by a
previous call to uv_replace_allocator.
As part of this change, the special aligned memory allocations
performed in src/win/fs-event.c have been replaced with standard
allocations. The 4-byte alignment being requested in this file was
unnecessary, since standard allocators already guarantee at least an
8-byte alignment.
PR-URL: https://github.com/libuv/libuv/pull/231
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Work around a bug in kernels 3.10 to 3.19 where passing a struct that
has the EPOLLWAKEUP flag set generates spurious syslog audit warnings.
A bad check makes the kernel read the events field when it should not
and complain when the process lacks the CAP_BLOCK_SUSPEND capability.
PR-URL: https://github.com/libuv/libuv/pull/245
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
In debugging test-tls-ocsp-callback.js for the 0.12.0 Node on
AIX we discovered that there was code missing in
uv__platform_invalidate_fd() in aix.c which removes the file
descriptor from the pollset when needed. We based the
impelementation in aix.c on the unix version in linux-core.cc.
We can see that this issue was addressed for linux in commit
780d8ad.
PR-URL: https://github.com/libuv/libuv/pull/221
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Calling `setsockopt()` on shutdown fds/stdio will result in EINVAL.
There is not much problem here as the OOB data can't be sent to already
shutdown fds. Just ignore it and go on.
PR-URL: https://github.com/libuv/libuv/pull/228
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
In the collaboration with Ben Noordhuis <info@bnoordhuis.nl> and
Saúl Ibarra Corretgé <saghul@gmail.com>.
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
There's a data race in the consuming side of uv_async. The "pending"
flag could be trampled by producing thread causing an async send
event to be missed.
PR-URL: https://github.com/libuv/libuv/pull/189
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
On Solaris derivatives, cfmakeraw is not available. Instead, set the
termios flags manually. The set of flags to use so that the behavior of
the terminal is similar to what it is after a call to cfmakeraw was
taken from
http://www.perkin.org.uk/posts/solaris-portability-cfmakeraw.html.
PR-URL: https://github.com/libuv/libuv/pull/210
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Before this commit, UNIX returned -1 on failure. Windows sometimes
returned a UV_E* error code and sometimes a bogus status code, courtesy
of errno values not mapping to UV_E* error codes on that platform.
PR-URL: https://github.com/libuv/libuv/pull/204
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Commit 393c1c5 ("unix: set non-block mode in uv_{pipe,tcp,udp}_open")
causes a regression in the io.js cluster module.
The io.js documentation states that `worker.send()` and `process.send()`
are synchronous but they no longer were after upgrading to libuv v1.2.1.
The reason they are synchronous is because of backpressure - or rather,
lack of backpressure: a slow consumer eventually causes a fast producer
to run out of memory because the backlog of pending messages in the
producer can grow unchecked.
Ergo, implement uv_stream_set_blocking() on UNIX platforms to let io.js
enable the old blocking behavior for pipes again.
Refs: https://github.com/iojs/io.js/issues/760
PR-URL: https://github.com/libuv/libuv/pull/187
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Since API level 21 (Android 5.0) the non-standard pthread function
pthread_cond_timedwait_monotonic_np has been removed in favour of
the standard pthread_cond_timedwait which libuv normally uses on
Linux.
This commit changes the detection of the Android OS to account for
the removal of the function via the #define
HAVE_PTHREAD_COND_TIMEDWAIT_MONOTONIC which is only present in older
API levels.
Fixes: https://github.com/libuv/libuv/issues/172
PR-URL: https://github.com/libuv/libuv/pull/176
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Linux before kernel 2.6.19 does not support epoll_pwait(). Due to a
logic error in commit 2daf944 ("unix: add flag for blocking SIGPROF
during poll"), the fallback path for ENOSYS was not taken.
This commit also adds epoll_pwait() emulation using pthread_sigmask().
The block/unblock operations are not atomic but that is fine for our
particular use case, to wit, sleep through SIGPROF signals.
PR-URL: https://github.com/libuv/libuv/pull/162
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
libuv requires that the socket/fd is in non-blocking mode, so do it
internally so the user doesn't need to write platform specific code to
do so.
This also makes the API consistent with uv_{tcp,udp,pipe}_open, since
it's not required to pass the fd in non-blocking mode there either.
PR-URL: https://github.com/libuv/libuv/pull/136
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
The contract specifies that the file descriptor should already be in
non-blocking mode before passing it to libuv.
However, node users don't really have an opportunity to do so, never
mind the fact that the call to uv_pipe_open() or uv_tcp_open() is an
implementation detail that most users won't be aware of.
Let's be nice and set the non-blocking flag explicitly. It's a cheap
operation anyway.
Fixes: https://github.com/libuv/libuv/issues/124
PR: https://github.com/libuv/libuv/pull/134
Reviewed-by: Saúl Ibarra Corretgé <saghul@gmail.com>
The mode argument is an enum now and the signedness of an enum is
implementation-defined when it doesn't have negative members.
Cast it to int in the comparison to tty->mode because the latter is
still an int.
PR: https://github.com/libuv/libuv/pull/134
Reviewed-by: Saúl Ibarra Corretgé <saghul@gmail.com>
I'm not 100% sure whether the behavior of readlink() with size=0 is
well-defined on Solaris; it can return either 0 or fail with EINVAL.
POSIX is not explicit about the mandated behavior.
This commit adds a little caution and makes it skip the readlink()
call when size=1 and simply writes the terminating zero byte.
PR-URL: https://github.com/libuv/libuv/pull/104
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
uv_exepath() should write as much as possible to the output buffer.
In the case of size=1, skip the call to readlink() because it will
fail with EINVAL; just write the terminating zero byte.
PR-URL: https://github.com/libuv/libuv/pull/104
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Write as much of the path as possible to the output buffer. Before this
commit, passing in a buffer that was too small to hold the result failed
with a UV_EINVAL error.
PR-URL: https://github.com/libuv/libuv/pull/104
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Passing a buffer that was too small to hold the result made it fail
with UV_EPERM because the -1 status code from _NSGetExecutablePath()
was returned unmodified to the caller.
This commit rewrites uv_exepath() to:
1. Not allocate heap memory, and
2. Not clobber the result buffer on error, and
3. Handle _NSGetExecutablePath()'s and realpath()'s idiosyncracies, and
4. Store as much of the path in the output buffer as possible, don't
fail with an error. Makes it behave the same as other platforms.
The result is always zero-terminated.
Fixes: https://github.com/libuv/libuv/issues/103
PR-URL: https://github.com/libuv/libuv/pull/104
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Make the behavior of a call to uv_exepath() with a size argument of zero
consistent with the Windows implementation where it returns UV_EINVAL.
PR-URL: https://github.com/libuv/libuv/pull/104
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
The fix from commit 269ff0be is wrong in that it sniffs system headers
to determine if preadv() and pwrite() are supported.
It breaks the fs_write_multiple_bufs test on one of my ARM systems;
its glibc has preadv() and pwritev() wrappers but its kernel has no
support for the corresponding system calls.
Simply try the system call and fall back to a normal read or write
loop if not supported. Move the no_preadv and no_pwritev variables
around to squelch two -Wpedantic warnings about mixing declarations
with code.
PR-URL: https://github.com/libuv/libuv/pull/90
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This reverts commit a9e95a33bc.
This seems to cause regressions in the test suite when IPv6 is disabled.
As optimizations go, it's a very minor one so let's not spend too much
effort trying to track down the cause of the regressions but revert the
change before it makes its way into a release.
PR-URL: https://github.com/libuv/libuv/pull/92
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Only call setsockopt() when the initial bind() fails with EADDRINUSE,
then retry. Saves a system call in what is normally the common case.
PR-URL: https://github.com/libuv/libuv/pull/88
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Introduce a uv_tty_mode_t enum for uv_tty_set_mode(), with backward
compatible values. Add a new mode UV_TTY_MODE_IO, which uses
cfmakeraw() internally.
PR-URL: https://github.com/libuv/libuv/pull/86
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Revisit the fix from commit b705b53. The problem with using sigset_t
and _NSIG is that the size of sigset_t and the value of _NSIG depend
on what headers libuv picks up first, <signal.h> or <asm/signal.h>.
With the former, sizeof(sigset_t) = 128; with the latter, it's 8.
Simply sidestep the issue by calculating the signal mask as a 64 bits
integer, without using sigset_t or _NSIG.
PR-URL: https://github.com/libuv/libuv/pull/83
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
If pending I/O callbacks were ran before polling, do a zero timeout
poll.
PR-URL: https://github.com/libuv/libuv/pull/58
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
The modes are not meant to be combined, and doing so may hide problems
in the future.
PR-URL: https://github.com/libuv/libuv/pull/58
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This aligns the behavior with the Windows implementation.
PR-URL: https://github.com/libuv/libuv/pull/63
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
As pointed out by clang-analyzer.
PR-URL: https://github.com/libuv/libuv/pull/13
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This was pointed out as possible undefined allocation of 0 bytes by
clang-analyzer. `malloc` of 0 bytes can return either `NULL` or a
valid pointer which can't be dereferenced. On systems which return
`NULL`, we would consider it a `malloc` error and return an `ENOMEM` to
the caller.
PR-URL: https://github.com/libuv/libuv/pull/13
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
As pointed out by clang-analyzer.
PR-URL: https://github.com/libuv/libuv/pull/13
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This aligns the behavior with Windows, where users don't need to
call `uv_read_stop` or `uv_close` if they get an error in the read
callback.
PR-URL: https://github.com/libuv/libuv/pull/47
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Add a per-event loop flag for blocking SIGPROF signals when polling for
events.
The motivation for this addition is to reduce the number of wakeups and
subsequent clock_gettime() system calls when using a sampling profiler.
On Linux, this switches from epoll_wait() to epoll_pwait() when enabled.
Other platforms bracket the poll syscall with pthread_sigmask() calls.
Refs strongloop/strong-agent#3 and strongloop-internal/scrum-cs#37.
PR-URL: https://github.com/libuv/libuv/pull/15
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
This reverts commit 0f25560c8a.
This change was introduced to make it possible to use the TTY for binary
I/O but unfortunately it breaks libuv users that depend on the OPOST and
ONLCR flags, like node.js.
There is no point in adding those flags after the call to cfmakeraw()
because that would once again make the TTY unsuitable for binary I/O.
Let's revert it for now and revisit it again later.
Fixeslibuv/libuv#32, reverts joyent/libuv#1567.
PR-URL: https://github.com/libuv/libuv/pull/33
Reviewed-By: Saúl Ibarra Corretgé <saghul@gmail.com>
Fix various typos and spelling mistakes in comments.
Does not affect any code, just changes comments.
PR-URL: https://github.com/libuv/libuv/pull/17
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Make sure we initialize it after the handle was removed from the pending
queue so that QUEUE_REMOVE doesn't do an invalid write when the process
is closed.
Valgrind output:
==4362== Invalid write of size 8
==4362== at 0x407DB8: uv__process_close (process.c:515)
==4362== by 0x404F94: uv_close (core.c:138)
==4362== by 0x4037C5: main (invalid_write.c:33)
==4362== Address 0xffeffc820 is not stack'd, malloc'd or (recently) free'd
==4362==
==4362== Invalid write of size 8
==4362== at 0x407DC3: uv__process_close (process.c:515)
==4362== by 0x404F94: uv_close (core.c:138)
==4362== by 0x4037C5: main (invalid_write.c:33)
==4362== Address 0xffeffc828 is not stack'd, malloc'd or (recently) free'd
Refs: joyent/libuv/issues/1584
linux-syscalls.h assumes that on all Linux platforms the value
of O_NONBLOCK is the same as SOCK_NONBLOCK.
This commit fixes it, as it is at least not true for hppa, and
resolves#1442.
Signed-off-by: Luca Bruno <lucab@debian.org>
sizeof(sigset_t) = 128 whereas the kernel expects 8, the size of a long.
It made the system call fail with EINVAL when a non-NULL sigset was
passed in. Fortunately, it's academic because there is just one call
site and it passes in NULL.
Fixeslibuv/libuv#4.
The existing probes, all two of them, cause a great deal of pain for
people trying to build libuv on Linux because of SystemTap's dtrace(1)
utilitity not understanding the -xnolibs flag.
We could hack around that but it's easier to just remove the probes:
they are largely useless and unused while still needing a lot of
supporting infrastructure. This commit removes 200 lines of code
and configuration.
Refs joyent/libuv#1478.
59658a8de7 changed uv_thread_self()
to return uv_thread_t, but uv_thread_t is a thread's HANDLE while
uv_thread_self() returns the current thread's id.
This means that uv_thread_equal() is also broken, as we are
potentially comparing HANDLES to ids.
Changed uv_thread_self() to return the current thread's creation handle.
Fixed small doc issue.
It seems that epoll_wait is implemented in glibc in terms of epoll_pwait and
new architectures (like arm64) do not implement the epoll_wait syscall at all.
So if epoll_wait errors with ENOSYS, just call epoll_pwait.
This is a backport of 861de3d71d for v0.10
branch.
It seems that epoll_wait is implemented in glibc in terms of epoll_pwait and
new architectures (like arm64) do not implement the epoll_wait syscall at all.
So if epoll_wait errors with ENOSYS, just call epoll_pwait.
`fd_set`s are way too small for `select()` hack when stream's fd is
bigger than 1023. Make `fd_set`s a part of `uv__stream_select_t`
structure.
fix#1461
Conflicts:
src/unix/stream.c
`fd_set`s are way too small for `select()` hack when stream's fd is
bigger than 1023. Make `fd_set`s a part of `uv__stream_select_t`
structure.
fix#1461
Returns the platform specific file descriptor for handles that are
backed by one. The datatype is abstracted as uv_os_fd_t, which maps to
int on Unices and HANDLE on Windows.
Users can use this function to set specific socket options, for example,
in a non portable way.
This function is essentially a shotgun, you better be careful with
whatever you do with it, don't blame me if you used it to get the fd of
a stream, close it yourself and expect things to Just Work.
If the connect wouldn't go off (no such tcp remote or any other failure),
the subsequent writes made would not be called. Now we call the writes
in the queue with ECANCELED if the connect fails.
Fix#1432
This fixes a SmartOS specific issue that happens when reading from
a stream that is the reading end of a pipe that has been closed by
the parent process.
In this case, a UV__POLLHUP event would be set on the stream and would
prevent the event loop from closing it. As a result, the event loop
would think there are stil handles open, and leave the process
hanging.
Fixes#1419.
Introduce:
int uv_fs_readdir_next(uv_fs_t* req, uv_dirent_t* ent);
`uv_fs_readdir()` is not returning a file names list in `req->ptr`
anymore, the proper way to gather them is to call `uv_fs_readdir_next()`
in a callback.
After 41891222bc landed it's possible that uv__udp_sendmsg is called
even if there are no pending write nor write completed requests:
1. User calls uv_udp_send and the request is sent immediately. The
request is the added to the completed queue and we 'feed' the uv__io
handle so that we process the completed request in the next
iteration.
2. User calls uv_udp_send again but the request is not completed
immediately, so it's queued in the write_queue.
3. The uv__io handle gets a UV__POLLOUT event and uv__udp_sendmsg is
run, which completes the send request and puts it in the
write_completed_queue. Afterwards, uv__udp_run_completed is executed
and the write_completed queue is drained.
4. At this point, the uv__io handle was made pending in step 3, in
uv__udp_sendmsg, but we no longer have requests to write or to complete,
so we skip processing.
This functionality is present in stream and uv_udp_t has a queue
as well so it makes sense for udp to have a send_write_size.
Since udp sends entire messages atomically, the send_queue_count field
lets the user determine how many messages are there left to send.
According to @aktau, the call to `mach_timebase_info` costs 90% of the
total execution time of `uv_hrtime()`. The result of the call is static
on all existing platforms, so there is no need in invoking it multiple
times.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Make uv_barrier_wait() return a boolean value indicating whether this
thread was the "serializer".
From `man pthread_barrier_wait`:
Upon successful completion, the pthread_barrier_wait() function
shall return PTHREAD_BARRIER_SERIAL_THREAD for a single (arbitrary)
thread synchronized at the barrier and zero for each of the other
threads.
Exposing that information from libuv is useful because it can make
cleanup significantly easier:
if (uv_barrier_wait(&barrier) > 0)
uv_barrier_destroy(&barrier);
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Only these functions will trigger an implicit binding of a UDP handle:
- uv_udp_send
- uv_udp_recv_start
- uv_udp_set_membership
All other functions will return UV_EBADF in case the socket was not
bound.
Note: currently the socket is created and bound at the same time. This
may change in the future.
If the handle was opened using `uv_udp_open` ift's possible that the
kernel doesn't fill in the msg_name field, so return NULL as the address
in that case.
fixes#1252
Reopen one of the pipe descriptors as read/write through the procfs.
Allows us to close the original pipe file descriptors, saving a file
descriptor on kernels that don't support eventfd(2).
The VirtualBox shared file system doesn't seem to be capable of
`readv()` syscall. Since we don't need them in the most of the cases
anyway, it won't hurt to just use `read()` and `write()` when we are
reading just one buffer.
see https://github.com/joyent/node/issues/7343 for details.
Add UV_UDP_REUSEADDR flag instead, which can be passed to uv_udp_bind.
If the udp handle is unbound when uv_udp_set_memberhsip or
uv_udp_set_multicast_interface is called, the handle will be bound with
UV_UDP_REUSEADDR set.
In case reopening the file descriptor fails, fallback to the old
behavior where we do blocking writes, to avoid disrupting other
processes potentially using the file descriptor.
Invalidate file descriptor when closing `uv_fs_event_t` handle. Note
that `uv__io_stop` is just removing `fd` from `loop->watchers` and not
actually invalidating all consequent events in a `kevent()` results.
fixjoyent/node#1101
Reopen the file descriptor when it refers to a tty. This lets us put the
tty in non-blocking mode without affecting other processes that share it
with us.
This brings back commit 31f9fbc, which was reverted in 20bb1bf. The OSX
select trick is working now.
Original patch by @bnoordhuis
If the same file description is open in two different processes, then
closing the file descriptor is not sufficient to deregister it from the
epoll instance (as described in epoll(7)), resulting in spurious events
that cause the event loop to spin repeatedly. So always explicitly
deregister it.
Fixes#1099.
Conflicts:
test/test-spawn.c
Introduce `int uv_pipe_pending_count(uv_pipe_t*)` and
`uv_handle_type uv_pipe_pending_type(uv_pipe_t*)`. They should be
used in IPC pipe's read cb to accept incoming handles:
int count = uv_pipe_pending_count(pipe);
int i;
for (i = 0; i < count; i++) {
uv_handle_type type = uv_pipe_pending_type(pipe);
/* ... */
uv_accept(...);
}
When the supplied buffer is not big enough and UV_ENOBUFS is
returned, hint the user about the required size by setting
the len paramemeter to the required value.
Applies to:
- uv_pipe_getsockname
- uv_fs_event_getpath
- uv_fs_poll_getpath
This improves API consistency with uv_read and uv_write and may
improve efficiency for some uses. Vectored IO is emulated when the
requisite system calls are unavailable.
These functions supersede uv_loop_new and uv_loop_delete.
uv_loop_init initialized a user allocated loop and uv_loop_close
removes all associated resources a loop uses after it has finished
execution.
uv_loop_new and uv_loop_delete are now deprecated.
If the same file description is open in two different processes, then
closing the file descriptor is not sufficient to deregister it from the
epoll instance (as described in epoll(7)), resulting in spurious events
that cause the event loop to spin repeatedly. So always explicitly
deregister it.
Fixes#1099.
Replace the red-black tree with a heap. The most common operation that
libuv performs on timers is looking up the first timer to expire. With
a red-black tree, that operation is O(log n). With a heap, it's O(1).
This improves API consistency with uv_read and uv_write and may
improve efficiency for some uses. Vectored IO is emulated when the
requisite system calls are unavailable.
Every file descriptor opened using libuv should be automatically marked
as CLOEXEC to prevent it from leaking to a child process. Note that
since we are opening fds in a thread pool, there is a possible race
condition between `uv_spawn()` and the `open()` + `uv__cloexec()`. The
rwlock was added to avoid it.
see https://github.com/joyent/node/issues/6905
Changed the order of the member assignments since the thread
may start before the parameters are assigned. This especially
happens when the osx scheduler is very busy.
When building using gyp and BUILDTYPE=Release using clang 3.4 received
this warning:
../src/unix/linux-core.c:640:34: warning: variable 'n' is uninitialized
for (len = sizeof("cpu0"); n /= 10; len++);
^
Initializing n = 0 silences this build warning.
Thus allow passing the same file descriptor as the source of multiple
stdios.
Committed with the help and code parts from:
* Sam Roberts <sam@strongloop.com>
fix#1074
CPU entries in /proc/stat are not guaranteed to be monotonically
increasing, asserting on this assumption can break in cases such
as the UltraSparc II machine shown in #1080.
Signed-off-by: Luca Bruno <lucab@debian.org>
uv__stream_osx_interrupt_select was only defined on OS X, but is used
elsewhere on all platforms. The intention was to implement it as a
no-op on other platforms, but the entire definition was inside
"#ifdef __APPLE__", so this didn't actually work.
Fix: move it above the #ifdef.
Relying on `readable`/`writable` when polling will lead to the select
thread spinning in the loop and calling `uv_async_send()`, because
stream may never become not `readable`/`writable`.
We should rely on `uv__io_active()` instead and interruprt select thread
every time it changes.
If multiple handles arrive to the IPC pipe at the same time (happens on
some platforms), libuv will queue them internally, and call `read2_cb`
multiple times with a null-buffer and proper `handle_type`.
In our build infrastructure, I've seen a lot of segfaults recently that
were all only happening on OSX. Upon inspecting the coredumps, it
appearded that all segfaults happened at the same instruction, and upon
translating the assembly back to the source, I found that an array could
be indexed with a -1 index before the index was checked to be not -1.
As concrete evidence, here is the situation that I found caused the
segfault. The instruction in question along with the relevant register
values was:
mov (%r8,%r15,8),%r12
r8 = 0x7fb0ba800000
r15 = 0xffffffffffffffff
r8 + r15 * 8 == 0x7fb0ba7ffff8
It appears that the base of loop->watchers was page aligned, and by
going back one word I guess that the page wasn't mapped, causing our
segfaults.
In our build infrastructure, I've seen a lot of segfaults recently that
were all only happening on OSX. Upon inspecting the coredumps, it
appearded that all segfaults happened at the same instruction, and upon
translating the assembly back to the source, I found that an array could
be indexed with a -1 index before the index was checked to be not -1.
As concrete evidence, here is the situation that I found caused the
segfault. The instruction in question along with the relevant register
values was:
mov (%r8,%r15,8),%r12
r8 = 0x7fb0ba800000
r15 = 0xffffffffffffffff
r8 + r15 * 8 == 0x7fb0ba7ffff8
It appears that the base of loop->watchers was page aligned, and by
going back one word I guess that the page wasn't mapped, causing our
segfaults.
Useful to know when the the event loop is empty, this can't be done with
uv_run() without possibly blocking, or running some events (which might
empty the event loop as a side-effect).
Some scandir implementations allocate the dirent struct even if the
directory is empty, so if `scandir` returns 0 there may still be memory
that needs to get deallocated. I have altered uv__fs_readdir to go to
the "deallocation exit area" when 0 files are found in the directory
and continue to return early on a return value of -1.
I went to add a test for this functionality, but it appears that one
already exists (reading an empty directory), so I imagine that the
valgrind builds must only be happening on linux instead of OSX as well?
I have confirmed manually that a program without this fix will
infinitely leak memory, and with this fix the memory usage stays
constant.
`uv_try_write(stream, buf, size)` acts like `uv_write()`,
but without queueing actual write until UV_POLLOUT (or IOCP completion).
This is useful for doing writes using on-stack `uv_write_t` requests.
fix#1025
First of all, a bit of explanation of what happens there:
1. FSEvents emits absolute paths to changed files or directories
2. We cut off the first part of such paths, which is equal to handle's
real path ('/dir/subdir/subsubdir`, without trailing slash)
3. Then, if we are running in non-recursive mode, we discard paths that
have slashes ('/') as a non-first character in them
Commit 3d2c820 back-ports a patch from the master branch that disables
the use of SO_REUSEPORT on Linux for reasons mentioned in the commit
log.
Unfortunately, the back-port was incomplete; another setsockopt() call
site in src/unix/udp.c was overlooked. This commit rectifies that.
Hat tip to Luca Bruno for helping troubleshoot the issue.
Fix a bug that was introduced in commit 3ee4d3f ("unix, windows:
return error codes directly") and add a regression test for good
measure.
Hat tip to Fedor for pointing out the issue.
Fixes#1007.
Print the error message rather than just the errno. The fact that the
errno is 24 is only informative to people that have their operating
system's error codes memorized.
Set the close-on-exec flag on file descriptors that we've received with
recvmsg() so we don't leak them when calling fork() afterwards.
On Linux, we use the MSG_CMSG_CLOEXEC flag when supported (2.6.23 and
up.)
On older Linux versions and other platforms, we walk the received file
descriptors and set the close-on-exec flag for each fd manually. That
won't entirely avoid race conditions when other threads call fork() or
clone() but at least we're less likely to leak file descriptors now.
When fd is closed and new one (with the same number) is opened inside
kqueue/epoll/port loop's callback - stale events might invoke callbacks
on wrong watchers.
Check if watcher was changed after invocation and invalidate all events
with the same fd.
fix#826
Otherwise `FSEventStreamCreate()` will coalesce events, even if they're
happening in the interval, bigger than supplied `latency`. In other
words, if this flag is not set events will happen in separate callback
only if there was a delay bigger than `latency` between two consecutive
events.
Work around an epoll quirk where it sometimes reports just the EPOLLERR
or EPOLLHUP event. In order to force the event loop to move forward,
we merge in the read/write events that the watcher is interested in;
uv__read() and uv__write() will then deal with the error or hangup in
the usual fashion.
Fixes#982.
This is a back-port of commit 24bfef2 from the master branch.
Work around an epoll quirk where it sometimes reports just the EPOLLERR
or EPOLLHUP event. In order to force the event loop to move forward,
we merge in the read/write events that the watcher is interested in;
uv__read() and uv__write() will then deal with the error or hangup in
the usual fashion.
Fixes#982.
Create file paths with CFStringCreateWithFileSystemRepresentation(),
not CFStringCreateWithCString().
Reapplies 3780e12 ("fsevents: support japaneese characters in path")
from the v0.10 branch. Was dropped in the last v0.10 -> master merge
for failing to apply.
Drops commit 3780e12 ("fsevents: support japaneese characters in path")
for being quite inapplicable to the master branch. Will be reworked
and applied in a follow-up commit.
Conflicts:
README.md
build.mk
src/unix/fsevents.c
src/unix/udp.c
If spawning a process fails due to an exec() failure (but it succeeded
in forking), then this should be considered a spawn failure instead of
an asynchronous termination of the process. This allows to check for
common exec() failure conditions such as a bad path quickly instead of
having to rely on keeping track of the async callback.
Additionally, the meaning of the two fields returned in the callback are
now exactly what they advertise to be. The process exit argument is not
one of two values depending on what happened to the child.
Fixes#978.
On the BSDs, SO_REUSEPORT is pretty much SO_REUSEADDR with some special
casing for IP multicast. When two processes (that don't do multicast)
bind to the same address, only the last one receives traffic. It allows
one to "steal" the bound address from another process. (Both processes
have to enable SO_REUSEPORT though, so it only works in a cooperative
setting.)
On Linux however, it enables port sharing, not stealing - both processes
receive a share of the traffic. This is a desirable trait but pre-3.9
kernels don't support the socket option and a libuv program therefore
behaves differently with older kernels or on another platform.
This is a back-port of commit 9d60f1e from the master branch.
Fixesjoyent/node#6454.
Watchers could be stopped between two `kevent()`/`epoll_wait()` calls
(which may happen in the same loop in `uv__io_poll()`), in such cases
`watcher->events` could be stale and won't be updated to
`watcher->pevents`.
Try to use and rely on `watcher->pevents` instead of blindly expecting
`watcher->events` to be always correct.
On some systems, clock_gettime(CLOCK_MONOTONIC) is only serviced from
the vDSO when the __vdso_clock_gettime() wrapper is confident enough
that the vDSO timestamp is highly accurate. When in doubt, it falls
back to making a traditional SYS_clock_gettime system call with all
the overhead that entails.
While a commendable approach, it's overkill for our purposes because we
don't usually need high precision time. That's why this commit switches
to CLOCK_MONOTONIC_COARSE for low-precision timekeeping, provided said
clock has at least a one millisecond resolution.
This change should eliminate the system call on almost all systems,
including virtualized ones, provided the kernel is >= 2.6.32 and glibc
is new enough to find and parse the vDSO.
Make it possible to call uv_tty_reset_mode() from inside a signal
handler. The primary motivation is to make it possible to restore
the TTY from inside a SIGINT or SIGTERM signal handler.
Fixes#954.
Before this commit, multiple event loops raced with each other when a
SIGCHLD signal was received. More concretely, it was possible for
event loop A to consume waitpid() events that should have been
delivered to event loop B.
This commit addresses that by doing a linear scan over the list of
child processes. An O(n) scan is not terribly efficient but the
actual performance impact is not measurable in a benchmark that spawns
rounds of several thousands instances of /bin/false. For the time
being, this patch will suffice; we can always revisit it later.
Fixes#887.
Before this commit, uv_uptime() returned the nanoseconds as the
fractional part of the uptime. It's a bit inconsistent with the
output on other platforms because those only return whole seconds
(with the possible exception of Windows.)
This commit changes linux-core.c to only return whole seconds.
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().
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.
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.
Fixesjoyent/node#6297.
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.
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 8fe4ca686b.
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.
Work around an 'initializer element is not constant' build error in
src/unix/fsevents.c by turning the const int flags into #defines.
Only an issue on OS X 10.6 due to the old compiler it uses.
Fixes#908.
This is a back-port of commit 82f2472 from the master branch.
