Changes since version 0.10.28:
* darwin: allocate enough space for select() hack (Fedor Indutny)
* linux: try epoll_pwait if epoll_wait is missing (Michael Hudson-Doyle)
* windows: map ERROR_INVALID_DRIVE to UV_ENOENT (Saúl Ibarra Corretgé)
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.
When multiple bufs are specified, overlapped needs to be advanced
manually between each write. Without this, each buf will be written
to the same position (if offset is specified.)
ReadFile() automatically advances, but WriteFile() does not.
uv_interface_addresses was using the linked list pointed to by
the FirstPrefix member of IP_ADAPTER_ADDRESSES in order to compute the
network prefix / network mask.
This was flawed in several ways:
- FirstPrefix can be NULL, and we would crash.
- On Windows Vista and later, the prefix list includes three IP adapter prefixes
for each IP address assigned to the adapter. We were assuming a 1:1 mapping
with the unicast address list.
- Even on Windows versions (i.e. XP) where the prefix list is supposed to have
one and only one element for each unicast address, the order of the two lists
is not guaranteed to be the same.
This fix was inspired and adapted from a commit in the Chromium project:
https://codereview.chromium.org/25167002/diff/6001/net/base/net_util_win.cc
See MSDN article for reference:
http://msdn.microsoft.com/en-us/library/windows/desktop/aa366058(v=vs.85).aspx
Excerpt from MSDN below:
In addition, the linked IP_ADAPTER_UNICAST_ADDRESS structures pointed to
by the FirstUnicastAddress member and the linked IP_ADAPTER_PREFIX
structures pointed to by the FirstPrefix member are maintained as separate
internal linked lists by the operating system. As a result, the order of
linked IP_ADAPTER_UNICAST_ADDRESS structures pointed to by the
FirstUnicastAddress member does not have any relationship with the order
of linked IP_ADAPTER_PREFIX structures pointed to by the FirstPrefix member.
On Windows Vista and later, the linked IP_ADAPTER_PREFIX structures pointed to
by the FirstPrefix member include three IP adapter prefixes for each IP address
assigned to the adapter. These include the host IP address prefix, the subnet
IP address prefix, and the subnet broadcast IP address prefix. In addition, for
each adapter there is a multicast address prefix and a broadcast address prefix.
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.
- Remove the UV_HANDLE_ACTIVE flag. It's a duplicate from
UV__HANDLE_ACTIVE, which was used solely on timers and loop watchers.
- Avoid duplicated code when running timers by stopping the handle and
rearming it with the repeat time, thus having a single place where the
timers are added and removed to and from the RB tree, respectively.
`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.
Support all possible types on Unix, and files, directories and links on
Windows. Some systems (hello SunOS!) don't have the d_type field on struct
dirent, so mark them as UV_DIRENT_UNKNOWN.
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.
This implements locking around the blocking call to ReadFile to get
around a Windows kernel bug where a blocking ReadFile operation on a
stream can deadlock the thread. This allows uv_read_stop to immediately
cancel a pending IO operation, and allows uv_pipe_getsockname to
"pause" any pending read (from libuv) while it retrieves the
sockname information.
If unsupported by the OS (pre-Vista), this reverts to the old
(e.g. deadlock-prone) behavior
Closes#1313
The unix and windows process implementations diverge in their behavior
when dealing with subprocesses that are spawned with a relative path.
With unix the *child's* PATH environment variable is read, whereas
with windows the *parent's* environment variable is read.
This commit brings the two implementation in line with respect to
their behavior of reading PATH by having both read the *child's* PATH
environment variable. This involves looking into the user-provided
environment on windows and extracting the PATH variable specifically
so it can be inspected later on.
Changes since version 0.10.27:
* windows: fix handling closed socket while poll handle is closing (Saúl
Ibarra Corretgé)
* unix: return system error on EAI_SYSTEM (Saúl Ibarra Corretgé)
* unix: fix bogus structure field name (Saúl Ibarra Corretgé)
* darwin: invoke `mach_timebase_info` only once (Fedor Indutny)
This is the libuv side of the fix for Node's cluster module on Windows.
https://github.com/joyent/node/issues/7691
Windows and Unix return certain socket errors (i.e. EADDRINUSE) at
different times: bind on Windows, and listen on Unix.
In an effort to hide this difference, libuv on Windows stores such
errors in the bind_error field of uv_tcp_t, to defer raising it at
listen time.
This worked fine except for the case in which a socket is shared in
a Node cluster and a bind error occurs.
A previous attempt to fix this (
d1e6be14603da36fe00e
) was flawed becaused in an attempt to relay the error at the JS level
it caused the master to start accepting connections.
With this new approach, libuv itself is relaying the bind errors,
providing for a uniform behavior of uv_tcp_listen.
If libuv is loaded as a DLL and is later unloaded deadlocks can happen
when running atexit handlers, so we can't use synchronization
priomitives or join threads there.
For reference see https://github.com/saghul/pyuv/issues/171
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 makes libuv more tolerant to the properties of the pipes
that it can use without any issue. This is necessary because Cygwin
(and hence Mintty) opens STDIN without FILE_WRITE_ATTRIBUTES.
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.
Debug build failed on mingw32 because CRT assertion disable code was stubbed out.
Replace __declspec(thread) with UV_THREAD_LOCAL which is defined as __thread on GCC.
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>
We add a function called uv_loop_size which returns the size of
the loop. It is an additional function which complements
`uv_handle_size` and` uv_req_size` in order to provide full FFI
support.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Switches from using ReadConsoleA to ReadConsoleW.
Uses an auxiliary buffer to store the temporary utf-16 string.
Signed-off-by: Peter Atashian <retep998@gmail.com>
Signed-off-by: Saúl Ibarra Corretgé <saghul@gmail.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>
Remove should shuffle items in both directions, not just down. It is
required, because `max` node could be not the actual maximum value in
the tree.
fix#1267
Signed-off-by: Fedor Indutny <fedor@indutny.com>
